Test Bank for Statistics for the Behavioral Sciences, 5th Edition by Nolan, Heinzen by ACADEMIAMILL

Name:

Class:

Date:

Chap 01_5e Indicate whether the statement is true or false. 1. Preregistration is a technique that is now discouraged by proponents of open science. a. True b. False 2. Discrete observations can typically take on more values than continuous observations. a. True b. False 3. In a between-groups research design, a participant experiences one and only one of the levels of the independent variable. a. True b. False 4. In a within-groups research design, a participant experiences more than one of the levels of the dependent variable. a. True b. False 5. By definition a population is smaller than its respective sample. a. True b. False 6. A test that has low reliability cannot have good validity. a. True b. False 7. In a between-groups research design, a participant experiences one and only one of the levels of the dependent variable. a. True b. False 8. Square footage, such as that used to describe home size, is a ratio variable. a. True b. False 9. Mattress size, including twin, full, queen, king, and California king, is an ordinal variable. a. True b. False

Page 1

Name:

Class:

Date:

Chap 01_5e 10. Square footage, such as that used to describe home size, is an interval variable. a. True b. False 11. In a within-groups research design, a participant experiences more than one of the levels of the independent variable. a. True b. False 12. In a between-groups research design, a participant experiences more than one of the levels of the dependent variable. a. True b. False 13. In a between-groups research design, a participant experiences more than one of the levels of the independent variable. a. True b. False 14. Open science is an approach to research that encourages researchers to work together and share methodology, data, and statistical analyses in order to allow for replication of their findings. a. True b. False 15. Continuous observations can typically take on more values than discrete observations. a. True b. False 16. A major electronics store calculates holiday spending for both its store-based sales (2.3 million) and online sales (1.4 million). These are both descriptive statistics. a. True b. False 17. A test that has poor validity cannot have good reliability. a. True b. False 18. Mattress size, including twin, full, queen, king, and California king, is a nominal variable. a. True b. False

Page 2

Name:

Class:

Date:

Chap 01_5e 19. Confounding variables are connected to the dependent variable such that we cannot know which variable causes the effect we observe. a. True b. False 20. By definition a sample is smaller than its respective population. a. True b. False 21. Confounding variables are connected to the independent variable such that we cannot know which variable causes the effect we observe. a. True b. False 22. In a within-groups research design, a participant experiences only one of the levels of the dependent variable. a. True b. False 23. In a within-groups research design, a participant experiences only one of the levels of the independent variable. a. True b. False 24. Inferential statistics summarize a group, while descriptive statistics help us make estimates about a larger population. a. True b. False 25. If a test is valid, then we know that it is reliable. a. True b. False 26. Descriptive statistics summarize a group, while inferential statistics help us make estimates about a larger population. a. True b. False

Page 3

Name:

Class:

Date:

Chap 01_5e Indicate the answer choice that best completes the statement or answers the question. 27. Melina is a student at a large university. When visiting professors during their office hours, she has noticed that many have refrigerators in their offices. She decides to survey 60 faculty and count the total number of refrigerators they have. What type of observation is she making? a. discrete b. continuous c. discrete and ordinal d. continuous and interval 28. A nutritional researcher was interested in whether the size of breakfast could decrease overall food consumption throughout the rest of the day. He creates two breakfast groups, a 350-calorie breakfast and a 700-calorie breakfast, assigns eight participants to each group, and tracks their total calories eaten in one day. Because of the detailed attention needed to accurately interview participants about their eating, he works with the high-calorie group and has his assistant interview the low-calorie group. What is the independent variable in this study? a. total calories consumed b. the breakfasts c. weight loss experienced in the day d. the researcher conducting the interviews 29. In 2010, there was an interesting lawsuit about bagels. A company claimed to have created a new way to recreate "Brooklyn style" bagels and then reported that another bagel producer stole its recipe. A researcher wonders if bagel sales might have been affected simply by the story making the national news, so she tracks total bagel sales in dollars for one year before and after the news story hits. What is the independent variable in this study? a. the types of bagels sold b. total sales c. the news story d. the lawsuit 30. In 2010, there was an interesting lawsuit about bagels. A company claimed to have created a new way to recreate "Brooklyn style" bagels and then reported that another bagel producer stole its recipe. A researcher wonders if bagel sales might have been affected simply by the story making the national news, so she tracks total bagel sales in dollars for one year before and after the news story hits. What type of variable is total bagel sales? a. nominal b. ordinal c. scale d. independent

Page 4

Name:

Class:

Date:

Chap 01_5e 31. A behavioral economist wanted to explore the connection between the number of bathrooms in a house and the sale price of the house. She studied 1384 home sales in an economically diverse, medium-sized city and found that the average sale price went up by $63,000 for each full bath. What is the descriptive statistic in this study? a. 1384 home sales b. houses in economically diverse, medium-sized cities c. number of bathrooms d. average increase in sale price of $63,000 per bathroom 32. Prof. Acosta was interested in which of two popular statistics textbooks (Statistics: It Will Change Your Life and Statistics: Bigger, Better, Stronger) was better for students. Prof. Acosta compared the two texts by assigning one text to a section of statistics taught by Prof. Agnew from 10 to 11 a.m. on Monday, Wednesday, and Friday, and the other text to a section of statistics taught by Prof. Alvarez from 7 to 10 p.m. on Wednesday evenings. At the end of the term, all students took the same comprehensive test. Students to whom Statistics: Bigger, Better, Stronger was assigned performed better on the test than did students to whom Statistics: It Will Change Your Life was assigned. Therefore, Prof. Acosta concluded that the former textbook was the better one. What was the independent variable in this study? a. statistics textbooks b. professors c. comprehensive test d. students 33. Hypothesis testing refers to: a. drawing conclusions about whether a particular relation between variables is supported by the evidence. b. the direct manipulation of an independent variable in an attempt to assess its effects on a dependent variable. c. summarizing data using descriptive statistics. d. measuring a variable of interest using an operational definition. 34. An education researcher studies length of time in college, first through fourth year, and its relation to academic motivation. To get the most detail out of her measures, she assesses each student in both the fall and spring semesters of each of their four years in school. She finds that students have increasingly higher motivation from their first semester to their seventh semester (the start of their fourth year), with a trailing off in the last semester. What is the independent variable in this study? a. year in school b. semester in school c. academic motivation d. time of year in which the assessment was completed

Page 5

Name:

Class:

Date:

Chap 01_5e 35. An article from the New York Times, published on April 24, 2007, summarized research conducted by Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above, he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. What type of research design did Dr. Vallortigara employ? a. between-groups b. within-groups c. non-experimental d. quasi-experimental 36. Random assignment refers to a situation in which: a. participants self-select into a particular condition in the study. b. the experimenter randomly determines whether to use a single-blind or double-blind research design. c. every person in the population has an equal chance of being selected for participation in the study. d. every participant in the study has an equal chance of being assigned to any condition or level of the independent variable. 37. The purpose of random assignment to groups is to: a. control confounding variables. b. ensure that you have a representative sample. c. control extraneous variables. d. reduce the noise in your study. 38. The term level refers to: a. a variable that is manipulated to determine its effects on another variable. b. the discrete values that a variable can take on. c. a situation in which two variables have the same value. d. a situation in which there are no confounding variables. 39. A population is generally defined as: a. the entire group of interest about which we want to make conclusions. b. a single number or group of numbers that organize, summarize, and communicate a group of numerical observations. c. a subset, or smaller collection, of observations from the overall group of interest. d. using data to make general estimates about the overall group of interest.

Page 6

Name:

Class:

Date:

Chap 01_5e 40. A behavioral economist wanted to explore the connection between the number of bathrooms in a house and the sale price of the house. She studied 1384 home sales in an economically diverse, medium-sized city and found that the average sale price went up by $63,000 for each full bath. What is the sample in this study? a. 1384 home sales b. houses in economically diverse, medium-sized cities c. number of bathrooms d. average increase in sale price of $63,000 per bathroom 41. In 2000, Bartels and Zeki conducted a study in which they hypothesized that there may be special pathways in the brain that support the feeling of romantic love. To test their hypothesis they recruited volunteers who reported themselves to be "truly, deeply, and madly in love." They then used brain imaging methods to determine which areas of the volunteers' brains were active when looking at pictures of their loved one. How did these researchers operationally define romantic love? a. They asked volunteers if they were in a romantic relationship. b. They gave volunteers a Depth of Love scale. c. They used self-reports of volunteers who claimed to be "truly, deeply, and madly in love." d. These researchers did not operationally define romantic love. 42. When researchers push for a "severe test" of a hypothesis, they are referring to approaches that: a. utilize rigorous analyses aimed at identifying flaws in the hypothesis. b. analyze data repeatedly and in different ways so as to support the research hypothesis. c. reject any collected data that do not match the stated hypothesis. d. only report statistically significant results. 43. The purpose of preregistering a study is to enable the researcher to: a. obtain research funding before they have collected any data. b. receive scientific recognition before other researchers can claim credit for the ideas. c. demonstrate that they conducted their research study as they originally intended. d. obtain permission to conduct the proposed research study. 44. Lewis has taken the GRE three times. Every time he takes the test he gets a 500 on the math section. This implies that: a. the GRE is a valid test. b. the GRE is a reliable test. c. the GRE is neither a valid nor a reliable test. d. Lewis is not motivated to improve his score on the math section.

Page 7

Name:

Class:

Date:

Chap 01_5e 45. Teachers and administrators debate whether standardized tests, such as the ACT and SAT, are predictors of college performance. This is essentially a debate about: a. validity. b. reliability. c. confounding variables. d. hypothesis testing. 46. A team of researchers interested in asthma symptoms wanted to know how symptoms were affected in dry versus humid conditions. The researchers recruited 18 asthma patients to spend four weeks under two conditions: sleeping with a dehumidifier for two weeks to create a "dry" environment and sleeping with a humidifier for the remaining two weeks to create a "humid" environment. Patients were asked to rate their symptoms at regular intervals using a scale from "0 – no symptoms" to "20 – maximum asthma symptoms." The change in asthma symptoms from dry to humid conditions was 5.82, showing a reduction of symptoms in humid conditions. What was the descriptive statistic in this study? a. change in responses on the scale of 5.82 on average b. four weeks, with two weeks under each condition c. 18 asthma patients d. two sleeping conditions, dry and humid 47. Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a pre-recorded psychology lecture. At the end of the 20 minute lecture, students answered a 17 question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who hadn't texted during the lecture. Which of these requires an inferential statistic? a. the random assignment of students into texting and non-texting groups b. recruiting the sample of 99 college students c. the average performance on the post-lecture quiz d. the conclusion that texting interferes with student learning 48. Which types of variables are considered scale variables by statistical computing packages such as SPSS? a. continuous and ratio b. continuous and interval c. discrete and interval d. ratio and interval

Page 8

Name:

Class:

Date:

Chap 01_5e 49. The variable that is manipulated or observed in order to determine its effects on another variable is the _____ variable. a. scale b. independent c. dependent d. confounding 50. A student's grade point average on a scale from 0 to 4.0 is a(n) _____ variable. a. nominal b. ordinal c. interval d. ratio 51. Francisco has a history of depression. As part of his self-care, he takes a depression assessment every six months. His results tend to be very consistent, except when he is in a serious depression and his results show elevated levels of depression. The tendency for his results to be consistent supports the _____ of the assessment. a. validity b. reliability c. continuous nature d. confounding nature 52. Why does random assignment help control for confounding variables? a. Random assignment ensures that participants in a study are properly motivated to perform the experimental task that will be required of them. b. Random assignment eliminates individual differences by removing individuals with the same characteristics from the study and only using individuals who have different characteristics. c. By randomly assigning people to groups, individual differences that may influence the dependent variable are randomly distributed throughout the conditions, rather than being systematically related to the independent variable. d. By randomly assigning people to groups, all individuals with similar characteristics are grouped together in the same condition.

Page 9

Name:

Class:

Date:

Chap 01_5e 53. An article from the New York Times, published on April 24, 2007, summarized research conducted by Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First, Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above, he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. What type of measure was the dependent variable in this study? a. nominal b. ordinal c. interval d. ratio 54. A research approach that encourages scientists to collaborate with each other and share their data, methodology and analyses is known as: a. data transparency. b. open science. c. data ethics. d. HARKing. 55. A descriptive statistic is generally defined as: a. the entire group of interest about which we want to make conclusions. b. a single number or group of numbers that organize, summarize, and communicate a group of numerical observations. c. a subset, or smaller collection, of observations from the overall group of interest. d. using data to make general estimates about the overall group of interest. 56. When a test or inventory measures what it is intended to measure, the test is said to be: a. a scale variable. b. continuous. c. reliable. d. valid. 57. A social psychologist was interested in the effects of gender on attitudes toward women in leadership positions. The researcher surveyed a group of individuals, 12 of whom were men and 12 of whom were women. In this example, gender is the _____ variable. a. level of the independent b. independent c. dependent d. confounding

Page 10

Name:

Class:

Date:

Chap 01_5e 58. Variables are: a. specific values (in whole numbers) that represent an individual's category membership. b. the value of physical, attitudinal, or behavioral characteristics for a given individual. c. hypothetical ideas that have been developed to describe and explain human behavior. d. observations of physical, attitudinal, or behavioral characteristics that can take on different values. 59. An inferential statistic is generally defined as: a. the entire group of interest about which we want to make conclusions. b. a single number or group of numbers that organize, summarize, and communicate a group of numerical observations. c. a subset, or smaller collection, of observations from the overall group of interest. d. using data to make general estimates about the overall group of interest. 60. A team of researchers interested in asthma symptoms wanted to know how symptoms were affected in dry versus humid conditions. The researchers recruited 18 asthma patients to spend four weeks under two conditions: sleeping with a dehumidifier for two weeks to create a "dry" environment and sleeping with a humidifier for the remaining two weeks to create a "humid" environment. Patients were asked to rate their symptoms at regular intervals using a scale from "0 – no symptoms" to "20 – maximum asthma symptoms." The change in asthma symptoms from dry to humid conditions was 5.82, showing a reduction of symptoms in humid conditions. What is MOST likely the population of interest for these researchers? a. the patients who participated in the study b. all asthma sufferers c. everyone who sleeps d. patients during the four weeks of the study 61. An article from the New York Times, published on April 24, 2007, summarized research conducted by Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. What is the dependent variable in this study? a. finding that dogs' tails went rightward for the owner and leftward for an unfamiliar dog b. the 30 dogs that were recruited for the study c. whether the dog wagged its tail to the right or left d. the type of visual stimulus provided to the dog

Page 11

Name:

Class:

Date:

Chap 01_5e 62. A nutritional researcher was interested in whether the size of breakfast could decrease overall food consumption throughout the rest of the day. He creates two breakfast groups, a 350-calorie breakfast and a 700-calorie breakfast, assigns eight participants to each group, and tracks their total calories eaten in one day. Because of the detailed attention needed to accurately interview participants about their eating, he works with the high-calorie group and has his assistant interview the low-calorie group. How many levels does the independent variable have in this study? a. 1 b. 2 c. 8 d. 16 63. The amount of food a person eats each week (as measured in calories) is: a. continuous and ratio. b. continuous and interval. c. discrete and interval. d. continuous and ordinal. 64. A nutritional researcher was interested in whether the size of breakfast could decrease overall food consumption throughout the rest of the day. He creates two breakfast groups, a 350-calorie breakfast and a 700-calorie breakfast, assigns eight participants to each group, and tracks their total calories eaten in one day. Because of the detailed attention needed to accurately interview participants about their eating, he works with the high-calorie group and has his assistant interview the low-calorie group. What is the confounding variable in this study? a. total calories consumed b. the low- and high-calorie breakfasts c. weight loss experienced in the day d. the individual conducting the interviews 65. A variable for which there is an infinite number of values between any two points on the scale is: a. discrete. b. ratio. c. continuous. d. confounding.

Page 12

Name:

Class:

Date:

Chap 01_5e 66. In a variant of the Coke/Pepsi Challenge, tasters try to identify regular and diet versions of these popular beverages under "blind" conditions, where they can't see the two products. How many levels are there to the independent variable? a. 1 b. 2 c. 4 d. 8 67. A behavioral psychologist wanted to determine whether eating sugary cereal for breakfast increased the aggression of first graders during their morning play period. After feeding a group of 20 students sugary cereal for breakfast she observed that, on average, the students committed 5.1 aggressive behaviors during their morning play period. In this example, the descriptive statistic is: a. the 5.1 aggressive behaviors. b. the 20 students the researcher observed. c. all first graders. d. all first graders who ate sugary cereal for breakfast. 68. A preschool school teacher is interested in the association between sugar consumption and activity level in preschool children. The teacher gives 30 preschool children from Preppy Preschool Playland 0 milligrams, 25 milligrams, or 50 milligrams of sucrose (sugar) in a breakfast drink. He then observes their behavior for 30 minutes during their morning outdoor play period and codes their activity level. In this example, the sample is: a. 30 preschool children. b. the amount of sucrose. c. all preschool children. d. activity level. 69. Five people run in an election for student body president. The votes are tallied to create a list of candidates from most to least popular. When the number of votes are actually presented, this is a(n) _____ variable. a. nominal b. ordinal c. interval d. ratio

Page 13

Name:

Class:

Date:

Chap 01_5e 70. Five people run in an election for student body president. The votes are tallied to create a list of candidates from most to least popular. The number of votes is then removed so that a list of candidates from most to least popular is presented. This is a(n) _____ variable. a. nominal b. ordinal c. interval d. ratio 71. Imagine that a variable "sensitivity to others" is measured from 0 (low) to 80 (high). Although it is possible to have low sensitivity to others, it is not conceptually possible to have no sensitivity at all. What type of variable is this MOST likely to be? a. nominal b. ordinal c. interval d. ratio 72. A behavioral psychologist wanted to determine whether eating sugary cereal for breakfast increased the aggression of first graders during their morning play period. After feeding a group of 20 students sugary cereal for breakfast she observed that, on average, the students committed 5.1 aggressive behaviors during their morning play period. In this example, the sample is: a. the 5.1 aggressive behaviors. b. the 20 students the researcher observed. c. all first graders. d. all first graders who ate sugary cereal for breakfast. 73. Research published by Hsee and Tang (2007) described the results of a study in which 195 college students completed a happiness scale (from 1 to 7) just before taking a midterm exam. On this scale, 1 corresponded to very unhappy and 7 to very happy. On average, the students rated their happiness as 6.18. In this study, which of these would require an inferential statistic? a. the average rating of happiness at 6.18 b. the conclusion that college students, on average, are rather happy prior to taking midterm exams c. the conclusion that these 195 college students are rather happy prior to taking this midterm exam d. the 195 college students who completed the happiness scale 74. What research technique is crucial to drawing the conclusion that the independent variable caused the change in the dependent variable? a. random selection b. random assignment to groups c. double-blind experiment d. quasi-experiment Copyright Macmillan Learning. Powered by Cognero.

Page 14

Name:

Class:

Date:

Chap 01_5e 75. A correlation measures the relationship among _____ or more variables. a. two b. three c. four d. five 76. The measurement of the performance of bicyclists in a race based on their finishing places is a(n) _____ variable. The measurement of the performance of bicyclists in a race based on their times to complete the race is a(n) _____ variable. a. ratio; interval b. ordinal; ratio c. ordinal; nominal d. nominal; ordinal 77. Variables on which measurement scale are always discrete? a. ratio and ordinal b. ratio and interval c. nominal and ordinal d. nominal and interval 78. In 2010, there was an interesting lawsuit about bagels. A company claimed to have created a new way to recreate "Brooklyn style" bagels and then reported that another bagel producer stole its recipe. A researcher wonders if bagel sales might have been affected simply by the story making the national news, so she tracks total bagel sales in dollars for one year before and after the news story hits. What is the dependent variable in this study? a. the types of bagels sold b. total sales c. the news story d. the lawsuit 79. Melina is a student at a large university. When visiting professors during their office hours, she has noticed that many have refrigerators in their offices. She decides to survey 60 faculty and count the total number of refrigerators they have. What is the variable in this study? a. the university where the data are collected b. number of faculty, 60 c. total number of refrigerators d. location of refrigerators

Page 15

Name:

Class:

Date:

Chap 01_5e 80. A psychological researcher wants to assess well-being among dog and cat owners. She administers a wellbeing assessment to 125 dog owners and 163 cat owners. What type of research design is being used? a. experimental research b. random assignment design c. between-groups d. within-groups 81. Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a pre-recorded psychology lecture. At the end of the 20 minute lecture, students answered a 17 question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who hadn't texted during the lecture. What is the sample in this study? a. the 99 college students who participated in the study b. all college students, those who text in class and those who don't text in class c. the number of questions answered correctly on the post-lecture quiz d. the 20-minute duration of the pre-recorded lecture 82. The number of times a person eats fast food each week is: a. discrete and ratio. b. continuous and interval. c. discrete and interval. d. continuous and ordinal. 83. A sample is generally defined as: a. the entire group of interest about which we want to make conclusions. b. a single number or group of numbers that organize, summarize, and communicate a group of numerical observations. c. a subset, or smaller collection, of observations from the overall group of interest. d. using data to make general estimates about the overall group of interest. 84. The outcome variable that we expect to change with changes in the independent variable is the _____ variable. a. confounding b. noise c. dependent d. scale

Page 16

Name:

Class:

Date:

Chap 01_5e 85. A behavioral economist wanted to explore the connection between the number of bathrooms in a house and the sale price of the house. She studied 1384 home sales in an economically diverse, medium-sized city and found that the average sale price went up by $63,000 for each full bath. What is a logical population to which the researcher would want to extend this finding? a. the 1384 homes involved in the research b. home sales across the country in which the research was conducted c. all home sales in diverse, medium-sized cities d. home sales throughout the last decade 86. Stacey is a weight-loss group instructor. To get a better idea of how to help those she will be working with to achieve their weight-loss goals, she wishes to know the average weight-loss goal of the individuals in her group. What kind of statistic should Stacey use? a. reliability b. population c. inferential d. descriptive 87. The statement "12405 college-aged students participated in a study examining the relationship between gender and depression" is an example of a(n) _____ in research and statistics. a. population b. sample c. descriptive statistic d. inferential statistic 88. A preschool school teacher is interested in the association between sugar consumption and activity level in preschool children. The teacher gives 30 preschool children from Preppy Preschool Playland 0 milligrams, 25 milligrams, or 50 milligrams of sucrose (sugar) in a breakfast drink. He then observes their behavior for 30 minutes during their morning outdoor play period and codes their activity level. In this example, the dependent variable is: a. 30 preschool children. b. the amount of sucrose. c. all preschool children. d. activity level. 89. The difference between an interval and a ratio variable is that: a. ratio scales indicate only difference, but interval scales indicate difference and order. b. interval scales indicate only difference, but ratio scales indicate difference and order. c. on a ratio scale, the number 0 corresponds to an absence of the quality, but this is not true for an interval scale. d. there are equal intervals between points on an interval scale, but this is not true for a ratio scale. Copyright Macmillan Learning. Powered by Cognero.

Page 17

Name:

Class:

Date:

Chap 01_5e 90. When reading your college textbooks, you may sometimes find errors in them. If you track the number of errors based on the edition of the textbook, you might find that 1st editions have more errors than 3rd, 5th, and 10th editions. What type of variable is the edition of the text you are assessing? a. nominal b. ordinal c. scale d. dependent 91. An operational definition is one that: a. can be flexibly implemented by any researcher. b. defines a variable in terms of observable and measurable behaviors. c. defines a variable in terms of a hypothetical construct. d. is used to determine the independent variable of an experiment. 92. The Hamilton Depression Rating Scale (HDRS) is a scale intended to measure depression levels, with higher scores indicating higher levels of depression. If the HDRS is a valid measure of depression, we would expect that: a. the results of the HDRS could not be consistently replicated. b. a person's score on the HDRS would not be related to his or her level of depression. c. people who get higher scores on the HDRS would be more depressed than people who get lower scores. d. people who get lower scores on the HDRS would be more depressed than people who get higher scores. 93. A medical researcher is interested in the effectiveness of natural remedies for allergies. The researcher randomly assigns to 36 allergy sufferers a treatment of herbal tea, homeopathic doses of the allergens, or a traditional antihistamine. What type of research design has the researcher employed? a. within-groups b. experimental c. correlational d. operational research 94. A person's religious affiliation is a(n) _____ variable. a. nominal b. ordinal c. interval d. ratio

Page 18

Name:

Class:

Date:

Chap 01_5e 95. Reading times are collected for bilingual participants, comparing their reading speed across their two languages. What type of research design would MOST likely be used in this study? a. experimental research b. random assignment design c. between-groups d. within-groups 96. A five-star rating system for restaurants is a(n) _____ variable. a. nominal b. ordinal c. interval d. ratio 97. A variable that consists of separate specific categories for which there are no values between categories is: a. discrete. b. ratio. c. continuous. d. confounding. 98. An article from the New York Times, published on April 24, 2007, summarized research conducted by Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First, Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above, he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. What is the independent variable in this study? a. finding that dogs' tails went rightward for the owner and leftward for an unfamiliar dog b. the 30 dogs recruited for the study c. how far each dog wagged its tail to the right or left d. the type of visual stimulus provided to the dog

Page 19

Name:

Class:

Date:

Chap 01_5e 99. An article from the New York Times, published on April 24, 2007, summarized research conducted by of Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First, Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above, he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. What type of measure was the independent variable in this study? a. nominal b. ordinal c. interval d. ratio 100. A behavioral psychologist wanted to determine whether eating sugary cereal for breakfast increased the aggression of first graders during their morning play period. After feeding a group of 20 students sugary cereal for breakfast she observed that, on average, the students committed 5.1 aggressive behaviors during their morning play period. In this example, the population is: a. the 5.1 aggressive behaviors. b. the 20 students the researcher observed. c. all first graders. d. all first graders who eat sugary cereal for breakfast. 101. A nutritional researcher was interested in whether the size of breakfast could decrease overall food consumption throughout the rest of the day. He creates two breakfast groups, a 350-calorie breakfast and a 700-calorie breakfast, assigns eight participants to each group, and tracks their total calories eaten in one day. Because of the detailed attention needed to accurately interview participants about their eating, he works with the high-calorie group and has his assistant interview the low-calorie group. What is the dependent variable in this study? a. total calories consumed b. the breakfasts c. weight loss experienced in the day d. the researcher conducting the interviews 102. A social psychologist was interested in the effects of gender on attitudes toward women in leadership positions. The researcher surveyed a group of individuals, 12 of whom were men and 12 of whom were women. In this example, men is a(n) _____ variable. a. level of the independent b. independent c. dependent d. confounding

Page 20

Name:

Class:

Date:

Chap 01_5e 103. A team of researchers interested in asthma symptoms wanted to know how symptoms were affected in dry versus humid conditions. The researchers recruited 18 asthma patients to spend four weeks under two conditions: sleeping with a dehumidifier for two weeks to create a "dry" environment and sleeping with a humidifier for the remaining two weeks to create a "humid" environment. Patients were asked to rate their symptoms at regular intervals using a scale from "0 – no symptoms" to "20 – maximum asthma symptoms." The change in asthma symptoms from dry to humid conditions was 5.82, showing a reduction of symptoms in humid conditions. What was the sample in this study? a. change in responses on the scale of 5.82 on average b. four weeks, with two weeks under each condition c. 18 asthma patients d. two sleeping conditions, dry and humid 104. The United States Forest Service Wildland Fire Assessment System measures fire danger as extreme, very high, high, moderate, or low. In this system, fire danger is: a. continuous and ratio. b. continuous and interval. c. discrete and interval. d. discrete and ordinal. 105. Prof. Acosta was interested in which of two popular statistics textbooks (Statistics: It Will Change Your Life and Statistics: Bigger, Better, Stronger) was better for students. Prof. Acosta compared the two texts by assigning one text to a section of statistics taught by Prof. Agnew from 10 to 11 a.m. on Monday, Wednesday, and Friday, and the other text to a section of statistics taught by Prof. Alvarez from 7 to 10 p.m. on Wednesday evenings. At the end of the term, all students took the same comprehensive test. Students to whom Statistics: Bigger, Better, Stronger was assigned performed better on the test than did students to whom Statistics: It Will Change Your Life was assigned. Therefore, Prof. Acosta concluded that the former textbook was the better one. Which of the following is NOT a potential confounding variable in this study? a. number of classes per week b. professors teaching course c. comprehensive test scores d. time of day course is taught

Page 21

Name:

Class:

Date:

Chap 01_5e 106. A psychological study designed to assess the effects of disclosure of ingredients on the experience of taste, Lee, Frederick, and Ariely (2006) approached patrons at a local pub and asked them to taste and rate a new beer: the MIT Brew. Some participants were told about the secret ingredient in the beer (a few drops of balsamic vinegar) either before tasting (before condition) or after tasting but before rating (after condition). Other participants were not told anything regarding the secret ingredient (not told condition). Which aspect of this study is an operational definition of the dependent variable? a. sample of patrons at the local pub b. disclosure of the ingredients c. experience of taste d. participants' responses on the taste rating scale 107. Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a pre-recorded psychology lecture. At the end of the 20 minute lecture, students answered a 17 question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who hadn't texted during the lecture. What is the population in this study? a. the 99 college students who participated in the study b. all college students, those who text in class and those who don't text in class c. the number of questions answered correctly on the post-lecture quiz d. the 20-minute duration of the pre-recorded lecture 108. Controlling for _____ variables permits researchers to make statements about cause–effect relationships between variables. a. discrete b. reliable c. scale d. confounding 109. The Consideration of Future Consequences scale is intended to measure the extent to which an individual considers the future when making immediate choices. If the scale is a reliable measure, we would expect that: a. a person's score on the scale might change from day to day. b. a person's score on the scale would be relatively stable from day to day. c. people with higher scores on the scale would have a greater tendency to consider future consequences. d. people with lower scores on the scale would have a lower tendency to consider future consequences.

Page 22

Name:

Class:

Date:

Chap 01_5e 110. Why do researchers typically study samples rather than populations? a. Entire populations can be too costly to study or impossible to access. b. Entire populations are too variable to study. c. Samples are more representative than their respective populations. d. Studying a sample is more difficult than studying a population. 111. Francisco has a history of depression. As part of his self-care, he takes a depression assessment every six months. His results tend to be very consistent, except when he is in a serious depression and his results show elevated levels of depression. The fact that Francisco's results vary with his changes in mood, mirroring his depression levels, supports the _____ of the assessment. a. validity b. reliability c. continuous nature d. confounding nature 112. Inferential statistics allow a researcher to: a. summarize numerical observations for a population. b. make inferences about a sample of interest given observations taken on a larger population. c. make inferences about a population of interest given observations taken on a smaller sample. d. summarize numerical observations for a sample. 113. A preschool school teacher is interested in the association between sugar consumption and activity level in preschool children. The teacher gives 30 preschool children from Preppy Preschool Playland 0 milligrams, 25 milligrams, or 50 milligrams of sucrose (sugar) in a breakfast drink. He then observes their behavior for 30 minutes during their morning outdoor play period and codes their activity level. In this example, the independent variable is: a. 30 preschool children. b. the amount of sucrose. c. all preschool children. d. activity level. 114. An education researcher studies year in college, first through fourth year, and its relation to academic motivation. To get the most detail out of her measures, she assesses each student in both the fall and spring semesters of each their four years in school. She finds that students have increasingly higher motivation from their first to fourth year, with a trailing off in the last semester. What is the dependent variable in this study? a. year in school b. semester in school c. academic motivation d. time of year in which the assessment was completed

Page 23

Name:

Class:

Date:

Chap 01_5e 115. A behavioral economist wanted to explore the connection between the number of bathrooms in a house and the sale price of the house. She studied 1384 home sales in an economically diverse, medium-sized city and found that the average sale price went up by $63,000 for each full bath. Which statement involves a logical inferential statistic based on this research? a. Adding a bathroom to your house will cost $63,000. b. On average, adding a bathroom to your house can increase the sale price. c. Houses sell for $63,000 on average. d. Bathrooms are highly desirable features of houses in medium-sized cities. 116. _____ variables are almost always continuous. a. Ordinal b. Interval c. Nominal d. Ratio 117. When reading your college textbooks, you may sometimes find errors in them. If you track the number of errors based on the edition of the textbook, you might find that 1st editions have more errors than 3rd, 5th, and 10th editions. What type of variable is the number of errors found? a. nominal b. ordinal c. scale d. independent 118. Hair color, assessed as blonde, black, red, brown, and other, as a variable is measured on a(n) _____ scale. a. nominal b. ordinal c. interval d. ratio 119. The number of top ten songs is calculated for 30 multi-platinum recording artists. What type of variable is number of top ten songs? a. nominal b. ordinal c. interval d. ratio

Page 24

Name:

Class:

Date:

Chap 01_5e 120. Prof. Acosta was interested in which of two popular statistics textbooks (Statistics: It Will Change Your Life and Statistics: Bigger, Better, Stronger) was better for students. Prof. Acosta compared the two texts by assigning one text to a section of statistics taught by Prof. Agnew from 10 to 11 a.m. on Monday, Wednesday, and Friday, and the other text to a section of statistics taught by Prof. Alvarez from 7 to 10 p.m. on Wednesday evenings. At the end of the term, all students took the same comprehensive test. Students to whom Statistics: Bigger, Better, Stronger was assigned performed better on the test than did students to whom Statistics: It Will Change Your Life was assigned. Therefore, Prof. Acosta concluded that the former textbook was the better one. What was the dependent variable in this study? a. statistics textbooks b. professors c. comprehensive test scores d. students 121. Dr. Hanstad was interested in the effect of Valium on motor performance. He injected 25 rats in the experimental group with a small amount of Valium and 25 rats in the control group with saline solution. Following injection, he measured the rate of bar pressing by both groups of rats. On average, rats in the control group had 800 presses per hour and rats in the experimental group had 615 presses per hour. The same testing box was used for both groups of rats, but different student assistants tested the control and experimental groups. In this example, having two different student assistants test the two groups is a(n) _____ variable. a. confounding b. nominal c. independent d. dependent 122. A team of researchers interested in asthma symptoms wanted to know how symptoms were affected in dry versus humid conditions. The researchers recruited 18 asthma patients to spend four weeks under two conditions: sleeping with a dehumidifier for two weeks to create a "dry" environment and sleeping with a humidifier for the remaining two weeks to create a "humid" environment. Patients were asked to rate their symptoms at regular intervals using a scale from "0 – no symptoms" to "20 – maximum asthma symptoms." The change in asthma symptoms from dry to humid conditions was 5.82, showing a reduction of symptoms in humid conditions. Which statement involves an inferential statistic related to this research finding? a. Asthma symptoms may be lowered, on average, with humid sleeping conditions. b. You can expect your asthma symptoms to diminish if you move to a drier climate. c. Varying your sleeping conditions can affect your health. d. Asthma symptoms increase when patients sleep with humidifiers.

Page 25

Name:

Class:

Date:

Chap 01_5e 123. When a researcher reformulates a hypothesis after the data has been collected and analyzed, the researchers has engaged in: a. data transparency. b. open science. c. a severe test of the hypothesis. d. HARKing. 124. A social psychologist was interested in the effects of gender on attitudes toward women in leadership positions. The researcher surveyed a group of individuals, 12 of whom were men and 12 of whom were women. In this example, what is the dependent variable? a. the 12 men in the study b. the 12 women in the study c. gender of the participants d. participants' attitudes toward women in leadership positions 125. A preschool school teacher is interested in the association between sugar consumption and activity level in preschool children. The teacher gives 30 preschool children from Preppy Preschool Playland 0 milligrams, 25 milligrams, or 50 milligrams of sucrose (sugar) in a breakfast drink. He then observes their behavior for 30 minutes during their morning outdoor play period and codes their activity level. In this example, the population is: a. 30 preschool children. b. the amount of sucrose. c. all preschool children. d. activity level. 126. Which of the following is NOT a variable? a. students' heights b. students' scores on a statistic exam c. maximum number of points possible on a 100-point exam d. students' scores on an empathy scale Enter the appropriate word(s) to complete the statement. 127. A(n) _______ variable gets consistent results; that is, it produces the same assessment over time for an unchanging variable.

Page 26

Name:

Class:

Date:

Chap 01_5e 128. A statistic that uses sample data to make general estimates about the larger population is a(n) _______ statistic.

129. The process of _______ involves a researcher specifying their data collection plans, methods, and analyses before beginning their research study.

130. Participants experience only one level of the independent variable in a(n) _______ design.

131. A(n) _______ variable makes it impossible to determine whether the independent variable is the cause of changes in the dependent variable.

132. When two or more variables vary together, they are said to be _______.

133. _______ is the process of drawing conclusions about whether or not a particular relation between variables is supported by the data.

Page 27

Name:

Class:

Date:

Chap 01_5e 134. A variable that is manipulated to determine its effects on another variable is a(n) _______ variable.

135. Continuous observations can typically take on _______ values than discrete observations.

136. A five-star rating system for hotels is a(n) _______ variable.

137. Each participant experiences all levels of the independent variable in a(n) _______ design.

138. When an investigator changes their research hypotheses to fit the data they have collected it is known as _______.

139. A statistic that summarizes a group of numbers is a(n) _______ statistic.

140. A(n) _______ measure sets out to evaluate something and does that accurately.

Page 28

Name:

Class:

Date:

Chap 01_5e 141. Discrete values that the independent variable can take on are called the _______ of the independent variable.

142. A person's religious affiliation is a(n) _______ variable.

143. A(n) _______ is a variable that meets the criteria for an interval and ratio variable.

144. What are some advantages in using a correlational design over an experimental design?

145. Research published by Hsee and Tang (2007) described the results of a study in which 195 college students completed a happiness scale (from 1 to 7) just prior to taking their midterm exam. On this scale, 1 corresponded to very unhappy and 7 to very happy. On average, the students rated their happiness as 6.18. Identify for this study the (a) population, (b) sample, (c) dependent variable, and (d) descriptive statistic.

Page 29

Name:

Class:

Date:

Chap 01_5e 146. A preschool school teacher is interested in the relation between sugar consumption and activity level in preschool children. The teacher gives 30 preschool children from Preppy Preschool Playland 0 milligrams, 25 milligrams, or 50 milligrams of sucrose (sugar) in a breakfast drink. He then observes their behavior for 30 minutes during their morning outdoor play period and codes their activity level. Identify for this study the (a) population, (b) sample, (c) independent variable, (d) levels of the independent variable, and (e) dependent variable.

147. Identify at least one confounding variable that undermines the conclusion drawn in the following fictional study: Prof. Acosta was interested in which of two popular statistics textbooks (Statistics: It Will Change Your Life and Statistics: Bigger, Better, Stronger) was better for students. Prof. Acosta compared the two texts by assigning one text to a section of statistics taught by Prof. Agnew from 10 to 11 a.m. on Monday, Wednesday, and Friday, and the other text to a section of statistics taught by Prof. Alvarez from 7 to 10 p.m. on Wednesday evenings. At the end of the term, all students took the same comprehensive test. Students to whom Statistics: Bigger, Better, Stronger was assigned performed better on the test than did students to whom Statistics: It Will Change Your Life was assigned. Therefore, Prof. Acosta concluded that the former textbook was the better one.

148. An operational definition specifies the operations or procedures used to measure or manipulate variables. A researcher is interested in investigating the relationship between thinking ability and educational success. What are some operational definitions that could be used in this study?

Page 30

Name:

Class:

Date:

Chap 01_5e 149. In a psychological study designed to assess the effects of disclosure of ingredients on the experience of taste, Lee, Frederick, and Ariely (2006) approached patrons at a local pub and asked them to taste and rate a new beer: the MIT Brew. Some participants were told about the secret ingredient in the beer (a few drops of balsamic vinegar) either before tasting (before condition) or after tasting but before rating (after condition). Other participants were not told anything regarding the secret ingredient (not told condition). Identify for this study the (a) population, (b) sample, (c) independent variable, (d) levels of the independent variable, and (e) dependent variable.

150. What is the difference between a within-groups and a between-groups design? What are the relative advantages and disadvantages of each type of design?

151. Gabriel is interested in examining the effect of classical music on problem solving. Explain what is involved in preregistering his study and how this approach would be beneficial.

152. An article from the New York Times, published on April 24, 2007, summarized research conducted by Dr. Vallortigara, a neuroscientist at the University of Trieste, Italy. In this study, Dr. Vallortigara assessed whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. First, Dr. Vallortigara recruited 30 dogs that were family pets. While filming a dog's tail from above, he allowed the dog to view (through a slot in its cage) its owner, an unfamiliar human, a cat, and an unfamiliar dominant dog. The study found that dogs' tails wagged to the right for the owner and to the left for the unfamiliar dominant dog. Identify for this study the (a) population, (b) sample, (c) independent variable, (d) levels of the independent variable, and (e) dependent variable.

Page 31

Name:

Class:

Date:

Chap 01_5e 153. Schacter and Gross (1968) gathered data from a group of 60 male students for about one hour in the afternoon. At the end of this period of time, a clock on the wall was correct (5:30 p.m.) for 20 participants, slow (5:00 p.m.) for 20 others, and fast (6:00 p.m.) for 20 more. The actual time for all groups was 5:30 p.m., the usual dinnertime for these students. While participants filled out a final questionnaire, the experimenters provided crackers for the students to eat. The weight of the crackers each student consumed was measured. The arithmetic means were 5:00 p.m. group, 20 grams; 5:30 p.m. group, 30 grams; 6:00 p.m. group, 40 grams. Identify for this study the (a) population, (b) sample, (c) independent variable, (d) levels of the independent variable, (e) dependent variable, (f) descriptive statistic, and (g) any inferences warranted by this study.

Page 32

Name:

Class:

Date:

Chap 01_5e Answer Key 1. False 2. False 3. True 4. False 5. False 6. True 7. False 8. True 9. True 10. False 11. True 12. False 13. False 14. True 15. True 16. True 17. False 18. False 19. False 20. True 21. True 22. False 23. False 24. False 25. True 26. True Copyright Macmillan Learning. Powered by Cognero.

Page 33

Name:

Class:

Date:

Chap 01_5e 27. a 28. b 29. c 30. c 31. d 32. a 33. a 34. b 35. b 36. d 37. a 38. b 39. a 40. a 41. c 42. a 43. c 44. b 45. a 46. a 47. d 48. d 49. b 50. d 51. b 52. c 53. a 54. b Copyright Macmillan Learning. Powered by Cognero.

Page 34

Name:

Class:

Date:

Chap 01_5e 55. b 56. d 57. b 58. d 59. d 60. b 61. c 62. b 63. a 64. d 65. c 66. b 67. a 68. a 69. d 70. b 71. c 72. b 73. b 74. b 75. a 76. b 77. c 78. b 79. c 80. c 81. a 82. a Copyright Macmillan Learning. Powered by Cognero.

Page 35

Name:

Class:

Date:

Chap 01_5e 83. c 84. c 85. c 86. d 87. b 88. d 89. c 90. b 91. b 92. c 93. b 94. a 95. d 96. b 97. a 98. d 99. a 100. d 101. a 102. a 103. c 104. d 105. c 106. d 107. b 108. d 109. b 110. a Copyright Macmillan Learning. Powered by Cognero.

Page 36

Name:

Class:

Date:

Chap 01_5e 111. a 112. c 113. b 114. c 115. b 116. d 117. c 118. a 119. d 120. c 121. a 122. a 123. d 124. d 125. c 126. c 127. reliable 128. inferential 129. preregistration 130. between-groups 131. confounding 132. correlated 133. Hypothesis testing 134. independent 135. more 136. ordinal 137. within-groups

Page 37

Name:

Class:

Date:

Chap 01_5e 138. HARKing 139. descriptive 140. valid 141. levels 142. nominal 143. scale variable 144. There are situations where random assignment to groups may not be available, which makes the use of a correlational design more practical because it has fewer restrictions. In correlational research, we do not have to manipulate any variables but only measure them and look for relations (correlations) between them. Also, sometimes we cannot conduct experiments because of ethical safeguards. 145. (a) The population is all college students just about to take a midterm exam; (b) the sample is the 195 college students selected for this study; (c) the dependent variable is the student's rating on the happiness scale; and (d) the descriptive statistic is the average of the happiness ratings, which is 6.18. 146. (a) The population is all preschool children; (b) the sample is the 30 preschool children who the teacher studied; (c) the independent variable is the amount of sucrose given to the children; (d) the levels of the independent variable are 0 milligrams, 20 milligrams, and 50 milligrams; (e) the dependent variable is each child's activity level. 147. There are several possible confounding variables that students may identify. One confound is the time at which the sections of the class are taught: one in the evening and the other in the morning. A second confound is the distribution of the class sections: one meets once a week, the other three times a week. A third confound is that different instructors taught each of the courses. A fourth confound is the time of day and length of time which the class is taught at one time: mornings for one hour, evenings for three hours. 148. Operational definitions are needed for the terms thinking ability and educational success. This is because the variables do not have specific definitions. An example of operational definitions that could be used in this study can be seen in the following research question: What is the relationship between IQ scores on an intelligence scale (operational definition for thinking ability) and GPA (operational definition for educational success)? Another example: It is hypothesized that problem-solving scores on a standardized test (operational definition for thinking ability) will impact grades in an engineering course (operational definition for educational success). 149. (a) The population is people who are patrons of pubs; (b) the sample is patrons approached in this study; (c) the independent variable is whether the patron was told about the secret ingredient in the MIT Brew; (d) the levels of the independent variable are told before, told after, or not told; (e) the dependent variable is the patron's taste rating for the beer.

Page 38

Name:

Class:

Date:

Chap 01_5e 150. In a within-groups design all participants are exposed to all levels of the independent variable, but in a betweengroups design each participant experiences only one level of the independent variable. One advantage to the within-groups design is that it requires fewer participants. If all the participants can participate in all conditions of the study, fewer individuals need to be recruited. Another advantage to the within-groups design is that it allows each participant to serve as their own baseline or control, against which we can measure any change across the various levels. One disadvantage to the within-groups study is order effects. By virtue of having participated in one condition of the study, participants may change their response in subsequent conditions. Another disadvantage to the within-groups study is carryover effects. Because the participants are in more than condition, the impact of one condition may carry over into the next condition, either compounding the effect or creating a unique effect. The between-groups design does not have these problems because each participant experiences only one of the conditions. 151. Preregistration is one recommended open-science practice, and involves the researcher laying out their methods and analysis plans before conducting their research. Gabriel would specify what variables would be measured, how his data would be collected, and the particular statistical tests that he intended to use to evaluate his hypotheses. Preregistration is helpful because Gabriel can show that he did everything he intended and that his results are not the consequence of unethical data practices. This approach also allows other researchers to have more confidence in the results of Gabriel's findings. 152. (a) The population is all dogs that are family pets; (b) the sample is 30 dogs selected for this study; (c) the independent variable is the type of visual stimulus the dog was allowed to see; (d) the levels of the independent variable are owner, unfamiliar human, a cat, and an unfamiliar dominant dog; (e) the dependent variable is the direction in which the dog wagged its tail. 153. (a) The population is male students; (b) the sample is the 60 male students who participated in the study; (c) the independent variable is the time that appeared on the clock; (d) the levels of the independent variable were correct, slow, and fast; (e) the dependent variable was the amount of crackers eaten by each student in grams; (f) there are three descriptive statistics provided: the means of each group: 20 grams for the slow group, 30 grams for the correct group, and 40 grams for the fast group; (g) one could infer that male students in general eat more when they believe it to be past their dinner hour.

Page 39

Name:

Class:

Date:

Chap_02_5e Indicate whether the statement is true or false. 1. A histogram places frequency on the y-axis and variable values on the x-axis. a. True b. False 2. Floor effects can lead to positive skew in a distribution. a. True b. False 3. It is advisable to use a grouped frequency table when depicting the frequency of interval data that vary over a large range of numbers in table format. a. True b. False 4. People who report "married" as their relationship status are assumed to have no less than one marriage. The fact that the number of marriages cannot vary below one represents a ceiling effect. a. True b. False 5. In a negatively skewed distribution, the tail extends to the left. a. True b. False 6. Placing one dot plot above another allows data from two samples to be compared. a. True b. False 7. Ceiling effects can lead to negative skew in a distribution. a. True b. False 8. Floor effects can lead to negative skew in a distribution. a. True b. False 9. Ceiling effects can lead to positive skew in a distribution. a. True b. False 10. Normal distributions are nonsymmetric and inherently have no skew. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_02_5e 11. Raw scores are data that have not been modified from their original form. a. True b. False 12. Some sports have what is called a "mercy rule," that is, once the difference in scores between two teams gets to a certain level, the game is ended. In soccer, the mercy rule might end a game when one team has 15 more goals than the other team. This limit on how big the difference between points can be is an example of a ceiling effect. a. True b. False 13. In a side-by-side dot plot, data from two samples are presented so that they might be compared. a. True b. False 14. Normal distributions are symmetric and inherently have no skew. a. True b. False 15. A histogram is typically used to depict scale data. a. True b. False 16. A histogram is typically used to depict nominal data. a. True b. False 17. A dot plot offers an advantage over a histogram in that it shows the individual data points and can be easier to interpret. a. True b. False 18. When creating a grouped frequency table, most researchers recommend using between 5 and 10 intervals. a. True b. False 19. Raw data are scores that have been modified from their original form. a. True b. False

Page 2

Name:

Class:

Date:

Chap_02_5e Indicate the answer choice that best completes the statement or answers the question. This table represents the fictional scores of a set of participants who rated their level of depression on a scale from 0 to 10, with 0 indicating no feelings of depression and 10 indicating very depressed. Table: Depression Score 10 9 8 7 6 5 4 3 2 1 0

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

20. (Table: Depression) What percent of participants rated their depression as 5? a. 4.00 b. 5.00 c. 5.71 d. 18.00

Page 3

Name:

Class:

Date:

Chap_02_5e This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score 60–62 57–59 54–56 51–53 48–50 45–47 42–44

Frequency 3 9 21 18 14 10 8

21. (Table: Grouped Frequency Table) What seems to be the shape of the distribution represented in this grouped frequency table? a. symmetrical b. positively skewed c. rectangle d. negatively skewed This table represents the fictional scores of a set of participants who rated their happiness on a scale from 1 to 7, with 1 indicating very unhappy and 7 indicating very happy. Table: Happiness X 7 6 5 4 3 2 1

F 3 5 11 9 3 1 2

22. (Table: Happiness) How many participants rated their happiness as 6 or higher? a. 3 b. 5 c. 8 d. 19

Page 4

Name:

Class:

Date:

Chap_02_5e 23. _____ distributions are those in which one tail of the distribution is pulled away from the center. a. Normal b. Nominal c. Skewed d. Interval 24. A negative skew may have a tail that indicates extreme scores _____ the center of the distribution. a. around b. below c. above d. on either side of 25. Bar graphs typically provide scores for _____ data. a. nominal b. ordinal c. interval d. ratio 26. In a _____, the tail of the distribution extends to the right. a. negatively skewed distribution b. positively skewed distribution c. ceiling effect d. normal distribution 27. Professor Obuseh calculates the grades on the first exam for her statistics class. She finds that students did really well, with most students scoring 98 or higher. What type of distribution is Professor Obuseh MOST likely to have? a. normal b. positively skewed c. nominal d. negatively skewed

Page 5

Name:

Class:

Date:

Chap_02_5e This histogram represents the distribution of the number of years of education completed by twins who attended the 16th Annual Twins Day Festival in Twinsburg, Ohio, in August of 1991. Figure: Years of Education

28. (Figure: Years of Education) Based on the distribution, how many twins completed 13 years of education? a. 11 b. 12 c. 20 d. 65 29. A _____ is a data point that has not yet been transformed or analyzed. a. frequency table b. raw score c. frequency distribution d. grouped frequency distribution

Page 6

Name:

Class:

Date:

30. (Figure: Years of Education) What seems to be the shape of this distribution? a. negatively skewed b. positively skewed c. rectangle d. symmetrical

Page 7

Name:

Class:

Date:

Chap_02_5e This histogram represents the frequency of graduation rates for all U.S. colleges (data collected by U.S. News & World Report, 1995). Figure: Graduation Rates

31. (Figure: Graduation Rates) The shape of the distribution of graduation rates appears to be: a. normal. b. rectangular. c. positively skewed. d. negatively skewed. 32. A grouped frequency table is most useful when the: a. scores in the data set vary over a small range of discrete values. b. data are ordinal. c. data are measured on an interval scale and vary over a large range of continuous values. d. data are nominal. 33. In dot plot comparing two samples, : a. the two plots are placed side by side. b. different symbols or shading, such as x's and o's, are used to identify observations from the two different samples. c. the dot plot for one sample is placed directly above the dot plot for the second sample. d. there is no way to distinguish scores from one sample or the other.

Page 8

Name:

Class:

Date:

Chap_02_5e 34. Professor Obuseh calculates the grades on the first exam for her statistics class. She finds that students did really well, with most students scoring 98 or higher. What type of effect, which often corresponds to a negatively skewed distribution, is MOST likely to be influencing the shape of the distribution of scores? a. floor b. ceiling c. raw score d. interval score 35. When constructing a frequency table, the final step is to: a. Divide the total number of participants by the total number of participants in a group and then multiply by 100. b. Divide the total number of participants in a group by the total number of participants and then multiply by 100. c. Subtract the total number of participants in a group from the total number of participants and then multiply by 100. d. Add the total number of participants in all groups and divide by 100. This table shows tests scores for a cumulative final in a general education, social science course, such as introduction to psychology. Table: Test Scores Interval 90–99 80–89 70–79 60–69 50–59 40–49 30–39 20–29

Frequency 23 41 78 36 18 7 12 3

36. (Table: Test Scores) What kind of frequency distribution is this? a. frequency table b. histogram c. grouped frequency table d. frequency polygon

Page 9

Name:

Class:

Date:

Chap_02_5e 37. A statistics professor examined her students' test scores and found that they did very well. She found that out of the 27 students in the class, 19 of them got a 95 on her test. What percentage of students got a 95? a. 42.11 b. 70.37 c. 76.00 d. 95.00 38. Distributions that are negatively skewed often result from: a. a ceiling effect. b. a floor effect. c. unimodal curves. d. a symmetrical distribution. 39. Raw data are observations or data points that: a. are in their original form. b. have been manipulated in some way. c. have been plotted on a graph. d. are discarded because they appear in error.

Page 10

Name:

Class:

Date:

Chap_02_5e This table represents the fictional scores of a set of participants who rated their level of depression on a scale from 0 to 10, with 0 indicating no feelings of depression and 10 indicating very depressed. Table: Depression Score 10 9 8 7 6 5 4 3 2 1 0

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

40. (Table: Depression) What percent of participants rated their depression as 1? a. 14.00 b. 14.29 c. 15.11 d. 70.00 41. A bell-shaped curve is similar to all EXCEPT which type of distribution? a. symmetric b. normal c. unimodal d. positively skewed

Page 11

Name:

Class:

Date:

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

42. (Table: Depression) How many participants rated their depression as 1? a. 1 b. 2 c. 5 d. 11

Page 12

Name:

Class:

Date:

43. (Figure: Years of Education) Based on the distribution, what is the number of years of education that was completed by most twins? a. 12.0 b. 13.0 c. 14.0 d. 16.0 44. Which of these is NOT displayed in a frequency table? a. the frequency of observations at each variable value b. values outside of the variable's range of observed values c. all observed variable values d. outlier data that is unexpected 45. A _____ visually depicts data based on intervals rather than frequencies for specific values. a. grouped frequency table b. frequency table c. frequency polygon d. normal distribution

Page 13

Name:

Class:

Date:

Chap_02_5e 46. Dot plots offer greater transparency as compared to summary statistics in that they allow: a. summary data to be displayed visually. b. each data point in a sample to be displayed. c. values of the data to be retained for later analyses. d. comparison of participants across different measures. This table represents the fictional scores of a set of participants who rated their happiness on a scale from 1 to 7, with 1 indicating very unhappy and 7 indicating very happy. Table: Happiness X 7 6 5 4 3 2 1

F 3 5 11 9 3 1 2

47. (Table: Happiness) How many participants did not rate their happiness as either 4 or 5? a. 11 b. 14 c. 16 d. 20 48. (Table: Happiness) What percentage of participants rated their happiness as 7? a. 4.86 b. 7.00 c. 8.82 d. 33.00 49. Imagine that 680 people out of a total of 715 people surveyed reported owning a smartphone. What percentage of people surveyed own a smartphone? a. 4.90 b. 48.75 c. 95.10 d. 96.45

Page 14

Name:

Class:

Date:

Chap_02_5e 50. In a _____, the tail of the distribution extends to the left. a. negatively skewed distribution b. positively skewed distribution c. normal distribution d. floor effect 51. Imagine that 16 out of every 34 homes have a dog in the household. What percentage of homes has a dog? a. 16.00 b. 47.06 c. 52.94 d. 88.88 52. When constructing a frequency table, the first step is to: a. count the number of scores at each value and write those numbers in the frequency column. b. create two columns. c. label one column Name and another column Frequency. d. determine the highest and the lowest score. This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score 60–62 57–59 54–56 51–53 48–50 45–47 42–44

Frequency 3 9 21 18 14 10 8

53. (Table: Grouped Frequency Table) Which interval has the least common exam score? a. 42–44 b. 45–47 c. 57–59 d. 60–62

Page 15

Name:

Class:

Date:

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

54. (Table: Depression) How many participants rated their depression levels? a. 10 b. 35 c. 55 d. 100 55. A positive skew may have a tail that indicates extreme scores _____ the center of the distribution. a. around b. below c. above d. on either side of

Page 16

Name:

Class:

Date:

Chap_02_5e This table shows tests scores for a cumulative final in a general education, social science course, such as introduction to psychology. Table: Test Scores Interval 90–99 80–89 70–79 60–69 50–59 40–49 30–39 20–29

Frequency 23 41 78 36 18 7 12 3

56. (Table: Test Scores) If grades are further sorted into plus and minus letter grades, for example, the scores from 80–89 are sorted into groupings of B, B+, and B– based on order, how many people would you estimate received a B+? a. 0 b. 21 c. 41 d. This cannot be determined based on the information provided. 57. A student researcher wanted to find the tallest person in a group of 20 women. Although she found that the tallest woman was 6 feet tall, her measurement was compromised by the fact that her scale reached only 6 feet. This example BEST illustrates which concept? a. floor effect b. skewed distribution c. ceiling effect d. negative skew

Page 17

Name:

Class:

Date:

Frequency 23 41 78 36 18 7 12 3

58. (Table: Test Scores) If passing is a 60 percent or higher, what percent of the class failed this test? a. 15.39 b. 18.35 c. 19.11 d. 26.12

Page 18

Name:

Class:

Date:

Chap_02_5e This table represents the fictional scores of a set of participants who rated their happiness on a scale from 1 to 7, with 1 indicating very unhappy and 7 indicating very happy. Table: Happiness X 7 6 5 4 3 2 1

F 3 5 11 9 3 1 2

59. (Table: Happiness) Based on the frequency distribution, what can be said about the level of happiness in this sample of individuals? a. Most people are very unhappy. b. Most people are very happy. c. Most people are neither very unhappy nor very happy. d. No conclusion about happiness can be drawn. 60. (Table: Happiness) What percentage of participants rated their happiness as 5? a. 11.00 b. 16.00 c. 32.35 d. 45.45

Page 19

Name:

Class:

Date:

Chap_02_5e The following dot plot displays daily hours of television viewing for a group of 25 students enrolled in a statistics class. Figure: Television Viewing Data

61. (Figure: Television Viewing Data) Based on the data presented, what was the most commonly reported amount of daily television viewing? a. 0 b. 1 c. 2 d. 7 62. Distributions that are positively skewed often result from: a. a ceiling effect. b. a floor effect. c. unimodal curves. d. a symmetrical distribution.

Page 20

Name:

Class:

Date:

Chap_02_5e The following dot plot displays daily hours of television viewing for a group of 25 students enrolled in a statistics class. Figure: Television Viewing Data

63. (Figure: Television Viewing Data) What best describes the shape of this distribution of hours of television viewing? a. positively skewed b. negatively skewed c. normally distributed d. multi-modal

Page 21

Name:

Class:

Date:

F 3 5 11 9 3 1 2

64. (Table: Happiness) How many people participated in this study (i.e., how many people provided happiness ratings)? a. 26 b. 28 c. 34 d. 38 65. (Table: Happiness) The most frequently occurring score in this data set is: a. 4. b. 5. c. 7. d. 11. 66. (Table: Happiness) How many participants rated their happiness as 4 or lower? a. 5 b. 9 c. 10 d. 15 67. Myra observes and records the number of people who purchase breakfast at a hospital cafeteria. The cafeteria is open from 7:00 a.m. to 11:00 a.m. and employees typically eat breakfast at 9:00 a.m. What type of distribution should Myra expect to see in her data? a. normal b. positively skewed c. negatively skewed d. nonsymmetric Copyright Macmillan Learning. Powered by Cognero.

Page 22

Name:

Class:

Date:

Chap_02_5e 68. For which situation is a grouped frequency table appropriate? a. data set on the weights of 50 adolescents, age 12 to 18 b. data set on the political affiliation of the students in your statistics class c. data set on the number of siblings of 50 adolescents, age 12 to 18 d. data set on the letter grades of the students in your statistics class This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score 60–62 57–59 54–56 51–53 48–50 45–47 42–44

Frequency 3 9 21 18 14 10 8

69. (Table: Grouped Frequency Table) How many students scored below 60? a. 71 b. 74 c. 80 d. 83 70. If 3 out of 4 dentists recommend a certain kind of gum, what percentage of dentists recommend that gum, rounded to the nearest whole number? a. 25 b. 34 c. 67 d. 75

Page 23

Name:

Class:

Date:

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

71. (Table: Depression) What was the most frequently reported level of depression? a. 0 b. 2 c. 11 d. 15

Page 24

Name:

Class:

Date:

Chap_02_5e This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score 60–62 57–59 54–56 51–53 48–50 45–47 42–44

Frequency 3 9 21 18 14 10 8

72. (Table: Grouped Frequency Table) Which interval has the most common exam score? a. 45–47 b. 48–50 c. 51–53 d. 54–56 73. Histograms are typically used to depict _____, whereas bar graphs are typically used to depict _____. a. scale data; nominal data b. nominal data; interval data c. means; frequencies d. interval data; scale data

Page 25

Name:

Class:

Date:

Chap_02_5e This table and figure depict the average SAT scores for entering freshmen in the year 1995 at 36 North Carolina colleges. Table: North Carolina SAT 825 990 1054 840 600 890 780 915 915

922 1230 845 923 1030 879 757 921 848

870 1302 826 818 831 1005 1002 1071 915

1121 926 956 867 935 842 774 921 813

Figure: Histogram of SAT

74. (Figure: Histogram of SAT) Based on the frequency distribution, approximately how many participants scored 1000 or above? a. 3 b. 5 c. 8 d. 19

Page 26

Name:

Class:

Date:

Chap_02_5e 75. (Figure: Histogram of SAT) Based on the frequency distribution, what was the interval with the most common score? a. 700–799 b. 800–899 c. 900–999 d. 1000–1099 This table represents the fictional scores of a set of participants who rated their level of depression on a scale from 0 to 10, with 0 indicating no feelings of depression and 10 indicating very depressed. Table: Depression Score 10 9 8 7 6 5 4 3 2 1 0

Frequency 1 6 1 1 4 2 1 1 11 5 2

Percent 2.86 17.14 2.86 2.86 11.43 5.71 2.86 2.86 31.43 ? 5.71

76. (Table: Depression) How many participants reported their level of depression at 5 or above? a. 11 b. 15 c. 19 d. 31 77. A normal distribution is also known as a _____ distribution. a. nonsymmetrical b. symmetrical c. skewed d. negative

Page 27

Name:

Class:

Date:

Chap_02_5e 78. Histograms typically provide frequencies for _____ data. a. nominal b. ordinal c. scale d. discrete This table and figure depict the average SAT scores for entering freshmen in the year 1995 at 36 North Carolina colleges. Table: North Carolina SAT 825 990 1054 840 600 890 780 915 915

922 1230 845 923 1030 879 757 921 848

870 1302 826 818 831 1005 1002 1071 915

1121 926 956 867 935 842 774 921 813

Figure: Histogram of SAT

79. (Figure: Histogram of SAT) What seems to be the shape of the distribution represented in this histogram? a. symmetrical b. positively skewed c. rectangle d. negatively skewed Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_02_5e 80. A _____ is a visual depiction of data that shows how often each value occurred. a. frequency distribution b. frequency table c. grouped frequency table d. frequency polygon This table shows tests scores for a cumulative final in a general education, social science course, such as introduction to psychology. Table: Test Scores Interval 90–99 80–89 70–79 60–69 50–59 40–49 30–39 20–29

Frequency 23 41 78 36 18 7 12 3

81. (Table: Test Scores) Based on this table, how many people passed this test if passing is 60 percent and higher? a. 152 b. 166 c. 178 d. 189 Enter the appropriate word(s) to complete the statement. 82. When a variable cannot take on values below a certain level, this is known as a(n) _______ effect.

Page 29

Name:

Class:

Date:

Chap_02_5e 83. The _______ is obtained by dividing the number of participants in a group by the total number and multiplying by 100.

84. When a variable cannot take on values above a certain level, this is known as a(n) _______ effect.

85. A histogram shares a lot in common with a(n) _______, except that the latter displays the frequencies of raw scores as dots on a graph.

86. A dot plot displays frequency data for _______ variable(s).

87. A frequency distribution that has a tail trailing off to the left of the distribution is _______ skewed.

88. _______ plots display each individual score of a single variable.

Page 30

Name:

Class:

Date:

Chap_02_5e 89. Because humans brains are well suited to make sense of small quantities, ______ are easier to interpret than _____.

90. A frequency distribution that is bell-shaped, symmetrical, and unimodal is _______.

91. A(n) _______ shows the pattern of data by indicating how many participants had each possible score.

92. A frequency distribution that has a tail trailing off to the right of the distribution is _______ skewed.

93. A(n) _______ is a data point that has not yet been transformed or analyzed.

94. When measuring a driver's time to brake for a red light, the measure is likely to be subject to a(n) _______ effect.

95. A distribution that has a tail in a positive or negative direction indicates the _______ of the distribution.

Page 31

Name:

Class:

Date:

Chap_02_5e 96. When a variable cannot take on values _______ a certain level, this is known as a floor effect.

97. A(n) _______ table is often used to display data when those data cover a very large range of values.

98. The distribution of total sales for albums released in 2018 is likely to be _______ skewed.

99. _______ look like bar graphs but typically depict interval data.

100. When a variable cannot take on values _______ a certain level, this is known as a ceiling effect.

101. When constructing a histogram and labeling the x- and y-axis, the lowest number on each axis should ideally be _______.

Page 32

Name:

Class:

Date:

Chap_02_5e 102. This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score 60–62 57–59 54–56 51–53 48–50 45–47 42–44

Frequency 8 14 21 18 11 7 4

103. This table depicts the scores of 83 students on an exam that was worth 65 points. Table: Grouped Frequency Table Exam Score Frequency 60–62 8 57–59 14 54–56 21 51–53 18 48–50 11 45–47 7 42–44 4 How many students received a score of 49?

Page 33

Name:

Class:

Date:

Chap_02_5e 104. (Table: North Carolina SAT) This table depicts the average SAT scores for entering freshmen in the year 1995 at 36 North Carolina colleges. 825 990 1054 840 600 890 780 915 915

922 1230 845 923 1030 879 757 921 848

870 1302 826 818 831 1005 1002 1071 915

1121 926 956 867 935 842 774 921 813

Use the data provided to create a grouped frequency table for the North Carolina SAT scores.

105. The figures in this table are the salaries for each of the 30 Colorado Rockies baseball players during the 2005 baseball season. Numbers are in thousands of dollars. Table: Colorado Rockies Salaries 320 317 316 317 316 2350 317 326 319 317

328 324 650 317 12,600 366 2400 2200 6575 321

316 326 950 950 318 316 316 317 12,500 550

Describe the skew of the distribution of salaries and explain what is causing it.

Page 34

Name:

Class:

Date:

Chap_02_5e 106. If we were to look at the distribution of salaries for all full-time psychology college faculty, what shape would we expect the distribution to have? Would it be normal, negatively skewed, or positively skewed? Why?

107. Table: Frequency Table X F 7 1 6 4 5 1 4 15 3 2 2 1 1 21 Use the information in the table to determine the percentages for each score. What information do you need in order to calculate the percentages?

Page 35

Name:

Class:

Date:

Chap_02_5e 108. (Table: North Carolina SAT) This table depicts the average SAT scores for entering freshmen in the year 1995 at 36 North Carolina colleges. 825 990 1054 840 600 890 780 915 915

922 1230 845 923 1030 879 757 921 848

870 1302 826 818 831 1005 1002 1071 915

1121 926 956 867 935 842 774 921 813

Use the data set to create a histogram. Based on the histogram, describe the skew of the data.

109. How do extreme observations affect the shape of a distribution?

Page 36

Name:

Class:

Date:

Chap_02_5e 110. This table depicts the annual salary for a sample of 10 full-time psychology faculty working at a large public university in 2018 . Table: Psychology Faculty Salaries Faculty Member 1 2 3 4 5 6 7 8 9 10

Salary ($US ) 99,700 167,900 122,000 152,300 110,800 61,000 112,000 82,000 108,300 307,700

Is it possible to calculate the percentages for the 10 Psychology faculty members listed in the table without a frequency column? If so, calculate the percentages. If not, explain.

Page 37

Name:

Class:

Date:

Chap_02_5e 111. Studies of the television viewing habits of males and females consistently find that males report spending more time watching television, e.g., Robinson & Killen (2013). A psychology instructor collects data from the students in her class. Table: Television Viewing Men 5 4 5 6 4 2 3 7 0 6 5

Women 0 1 0 1 1 1 4 3 2 6 0 2

(a) Construct dot plots for these data. (b) What do the dot plots reveal about the relative distributions of these data?

Page 38

Name:

Class:

Date:

Chap_02_5e 112. This table depicts the cost of electricity in cents per kilowatt for the Northern Atlantic states during a single month of 2019. Table: Electricity Cost State Maine New Hampshire Vermont Massachusetts Rhode Island Connecticut New York New Jersey Pennsylvania

Cents per kilowatt 16.16 19.63 18.64 21.11 18.64 21.62 19.30 15.64 13.58

Describe the shape of the distribution of electricity costs in the Northern Atlantic States? Is it normal or skewed? Explain your answer.

Page 39

Name:

Class:

Date:

Chap_02_5e Answer Key 1. True 2. True 3. True 4. False 5. True 6. True 7. True 8. False 9. False 10. False 11. True 12. True 13. False 14. True 15. True 16. False 17. True 18. True 19. False 20. c 21. d 22. c 23. c 24. b 25. a 26. b Copyright Macmillan Learning. Powered by Cognero.

Page 40

Name:

Class:

Date:

Chap_02_5e 27. d 28. c 29. b 30. b 31. a 32. c 33. c 34. b 35. b 36. c 37. b 38. a 39. a 40. b 41. d 42. c 43. a 44. b 45. a 46. b 47. b 48. c 49. c 50. a 51. b 52. a 53. d 54. b Copyright Macmillan Learning. Powered by Cognero.

Page 41

Name:

Class:

Date:

Chap_02_5e 55. c 56. d 57. c 58. b 59. c 60. c 61. c 62. b 63. a 64. c 65. b 66. d 67. a 68. a 69. c 70. d 71. b 72. d 73. a 74. c 75. b 76. b 77. b 78. c 79. b 80. b 81. c 82. floor Copyright Macmillan Learning. Powered by Cognero.

Page 42

Name:

Class:

Date:

Chap_02_5e 83. percentage, percent 84. ceiling 85. dot plot 86. one, 1 87. negatively 88. Dot 89. dot plots; histograms 90. normal distribution, normal 91. frequency table 92. positively 93. raw score 94. floor 95. skewness, skew 96. below 97. grouped frequency 98. positively 99. Histograms, Histogram 100. above 101. 0, zero 102. The distribution is negatively skewed. The data rise very quickly at the higher scores and trail off to the lower values. 103. Given that this is a grouped frequency table, it is not possible to know exactly how many people received a score of 49. We do know, however, that 11 students received a score between 48 and 50.

Page 43

Name:

Class:

Date:

Chap_02_5e 104. (Table: Grouped Frequency SAT) The following table depicts one possible grouped frequency table that can be constructed from the data provided. Table: Grouped Frequency SAT SAT 1212–1313 1110–1211 1008–1109 906–1007 804–905 702–803 600–701

Frequency 2 1 3 13 13 3 1

105. The distribution of salaries is positively skewed. The salaries tend to cluster in the low to mid $300,000s, with a collection of higher salaries, including $900,000 up to $12,600,000. These salaries create the trailing off of data at the high end, which is part of a positive skew. 106. It is likely that the distribution would be positively skewed. There are a some very highly paid faculty members, which would pull the tail of the distribution out to the right. Also, a floor effect on the faculty salaries would be likely, with no college faculty members making less than a certain amount. 107. (Table: Frequency Table Answer) Before calculating the percentages for each score, we must first obtain the total number of participants. We obtain this number by adding up all of frequencies, which comes to 45. Now we can obtain the percentages for each score by dividing the total number for each group (X) by the total number of participants (45) and multiplying by 100. Table: Frequency Table Answer X 7 6 5 4 3 2 1

F 1 4 1 15 2 1 21

Percent 2.22 8.89 2.22 33.33 4.44 2.22 46.67

Page 44

Name:

Class:

Date:

Chap_02_5e 108. (Figure: Histogram of SAT Answer) A sample histogram, which was generated in SPSS, is depicted here. This distribution is positively skewed. Figure: Histogram of SAT Answer

109. Extreme observations can affect the shape of a distribution by pulling the distribution in either direction. This can result in a positively or negatively skewed distribution depending upon the nature of the extreme observation. 110. It is possible because the frequencies are already described as a total of 10 participants with each participant belonging to his own group. Since each faculty member is a single group, the corresponding frequency is 1. Since we have the total number of participants per "group" as well as the total number of participants overall, 10, it is possible to calculate the percentages for each faculty member by dividing 1 (number in group) by 10 (total number) and multiplying by 100. The result would be a percentage of 10 for each faculty member.

Page 45

Name:

Class:

Date:

Chap_02_5e 111. (a) (Table: Dot Plot)

(b) The dot plots indicate that the two groups, males and females, differ a great deal. While there is overlap between the two distributions, female participants report watching much less television than do the male participants. The data displayed in these dot plots help us visually consider whether there is an important difference between these groups or whether they are more similar than we might have expected. 112. The distribution of electricity costs in the Northern Atlantic States is negatively skewed because higher scores are clustering on the right-hand side of the distribution, pulling the tail to the left-hand side of the distribution.

Page 46

Name:

Class:

Date:

Chap_03_5e Indicate whether the statement is true or false. 1. The lie of interpolation assumes that values beyond the data points will continue indefinitely, while the lie of extrapolation assumes that some value between the data points lies on a straight line between those data points. a. True b. False 2. Graphs based on sneaky samples are representative of the population. a. True b. False 3. When creating a bar graph, the pattern that fills in each bar is a type of moiré vibration. a. True b. False 4. It is always best to use the program defaults when creating graphs using computer software programs. a. True b. False 5. A bar graph is almost always better at presenting data than a pie chart. a. True b. False 6. A duck is a type of chartjunk that takes the form of a background pattern. a. True b. False 7. If two variables are not related linearly, than they must be related nonlinearly. a. True b. False 8. Geographic information systems is an innovation in graphing that has abundant application in geography but not in other social sciences. a. True b. False 9. The lie of extrapolation assumes that values beyond the data points will continue indefinitely, while the lie of interpolation assumes that some value between the data points lies on a straight line between those data points. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_03_5e 10. The y-axis of a bar graph indicates discrete levels of a nominal variable. a. True b. False 11. A pie chart is almost always better at presenting data than a bar graph. a. True b. False 12. The x-axis of a bar graph indicates discrete levels of a nominal variable. a. True b. False 13. An inaccurate values lie tells the truth in one part of the data but visually distorts it in another place. a. True b. False 14. The y-axis on a graph should start with 0, or have cut marks to indicate that it does not go down to 0. a. True b. False 15. The purpose of a graph is to reveal and clarify relations between variables. a. True b. False 16. Multivariable graphs allow for the simultaneous depiction of several variables at once. a. True b. False 17. The lie of interpolation assumes that a trend in the data extends beyond the bounds of the measurement scale. a. True b. False 18. The Pareto chart is a type of bar graph. a. True b. False 19. A pie chart is a type of multivariable graph that allows you to plot numerous variables simultaneously. a. True b. False

Page 2

Name:

Class:

Date:

Chap_03_5e 20. A scatterplot with a line of best fit is known as a time line plot. a. True b. False 21. Data based on a biased scale typically do not convey what the researcher wants to convey. a. True b. False Indicate the answer choice that best completes the statement or answers the question. This figure depicts data reflecting college students' scores on the "Big 5" personality dimension of conscientiousness and their test grades in a section of introductory psychology. Figure: Conscientiousness and Test Performance

22. (Figure: Conscientiousness and Test Performance) Based on the graph, what type of relation exists between conscientiousness and test grade? a. linear b. nonlinear c. It is not possible to tell given the information in the graph. d. There is no relation between rating rank and earning rank. 23. What is a potential benefit of creating a Pareto chart rather than a bar graph? a. A Pareto chart is easier to create than a bar graph. b. A Pareto chart looks more sophisticated than a bar graph. c. A Pareto chart makes it easier to compare the magnitude of the dependent variable for different levels of the independent variable. d. A Pareto chart uses pictorial symbols to make it easier to compare results.

Page 3

Name:

Class:

Date:

Chap_03_5e This figure depicts data from the Schachter (1968) study on obesity and eating behavior. Figure: Taste Test

24. (Figure: Taste Test) What are the independent variables in this study? a. weight of participant and number of crackers consumed b. weight of participant and stomach condition c. stomach condition and number of crackers consumed d. obese weight and non-obese weight 25. In the _____ lie participants in a study are preselected, resulting in biased results. a. biased scale b. extrapolation c. sneaky sample d. inaccurate values

Page 4

Name:

Class:

Date:

Chap_03_5e This figure depicts data reflecting college students' scores on the "Big 5" personality dimension of conscientiousness and their test grades in a section of introductory psychology. Figure: Conscientiousness and Test Performance

26. (Figure: Conscientiousness and Test Performance) Given the construction of this graph, what is the independent variable? a. conscientiousness percentile b. test grade c. percentile rank d. There is not enough information in the graph to determine the independent variable.

Page 5

Name:

Class:

Date:

Chap_03_5e This figure depicts the relation between the level of poverty, as measured by the proportion of students qualifying for free and reduced lunch, amongst students at 11 middle schools in a medium sized midwestern city and the 9th grade core grade point average (GPA) of the students from those 11 middle schools. Figure: Poverty and GPA

27. (Figure: Poverty and GPA) Given the construction of this graph, what is the independent variable? a. poverty level b. freshman class rank c. 9th grade GPA d. number of middle schools

Page 6

Name:

Class:

Date:

Chap_03_5e This figure depicts the weekly amount of television viewing reported by 151 college undergraduates. Figure: Histogram in SPSS

28. (Figure: Histogram in SPSS) What type of variable is depicted in the graph? a. nominal b. ordinal c. scale d. Pareto 29. Pie charts represent the percent or proportion of each observed category through the: a. size of each slice of the pie. b. color of each slice of the pie. c. number of slices into which the pie is divided. d. orientation of the pie on the page.

Page 7

Name:

Class:

Date:

Chap_03_5e This graph depicts fictional data that replicate the pattern of performance observed in a study published by Nietfeld and Ender (2003). Figure: Intelligence and Teaching Efficacy

30. (Figure: Intelligence and Teaching Efficacy) What might be one way of improving the readability of the graph? a. Add some ducks to the graph. b. Trim the range represented on both axes so that it depicts only values in the actual data set. c. Use a computer program to put gridlines on the background of the graph. d. Expand the range represented on both axes so that it depicts values beyond the actual data set for grounding.

Page 8

Name:

Class:

Date:

Chap_03_5e This figure depicts the typical relation found between a person's age (in years) and his or her reaction time in a simple reaction time task (e.g., time to press a button when a light turns green). Figure: Age and Reaction Time

31. (Figure: Age and Reaction Time) Which statement best captures what the graph depicts? a. Reaction time decreases throughout childhood and adolescence, becoming fastest in a person's early 20s, but after that point, reaction time slowly increases as a person continues to age. b. There is a linear relation between age and reaction time such that reaction time increases with increased age. c. Reaction time increases throughout childhood and adolescence, becoming slowest in a person's early 20s, but after that point, reaction time slowly decreases as a person continues to age. d. There is no relation between age and reaction time. 32. Popular magazines and Internet sources often use _____ to display information, which tend to confuse rather than clarify information. a. pie charts b. line graphs c. pictorial graphs d. scatterplots

Page 9

Name:

Class:

Date:

Chap_03_5e This figure depicts changes in global temperatures from 1989 through 2018. Data represent the deviation in temperature from a 30 year average from 1951-1980. (Data were extracted from the NASA Goddard Institute for Space Studies Web site.) Figure: Global Temperature

33. (Figure: Global Temperature) Identify the type of graph depicted in the figure. a. histogram b. time series plot c. Pareto chart d. line graph

Page 10

Name:

Class:

Date:

Chap_03_5e This figure was created from 2018 U.S. News & World Report data on 4-year colleges and universities in the United States. Figure: Cost of Tuition

34. (Figure: Cost of Tuition) The independent variable in this study was: a. private colleges. b. public colleges. c. the type of college. d. the cost of tuition and fees.

Page 11

Name:

Class:

Date:

35. (Figure: Poverty and GPA) Based on this graph, what type of relation exists between middle school poverty and 9th grade GPA? a. positive b. nonlinear c. curvilinear d. negative 36. The term levels refers to: a. continuous variables. b. linear relationships. c. bar graphs. d. categories. 37. To graph the frequencies of one scale variable, use a: a. scatterplot or line graph. b. Pareto chart. c. histogram or frequency polygon. d. bar graph.

Page 12

Name:

Class:

Date:

Chap_03_5e 38. In a _____ images are used at each level of the independent variable to represent its value on the dependent variable. a. scatterplot b. time series plot c. pictorial graph d. bar graph 39. _____ allow one to connect points on a map with data points located in space, such as homes for sale. a. Pictorial graphs b. Geographic information systems (GIS) c. Interactive graphs d. Word clouds 40. A type of line graph used to depict changes in a dependent variable over time is a: a. scatterplot. b. histogram. c. time series plot. d. line graph. This figure depicts the average IQ of the population in each of the 50 states. Figure: State IQ

41. (Figure: State IQ) What kind of graph is depicted in the figure? a. bar graph b. Pareto chart c. histogram d. frequency polygon

Page 13

Name:

Class:

Date:

Chap_03_5e This figure depicts the results of a study by Shinskey and Munakata (2005) investigating the reaching behavior of seven 5-month-old infants. Figure: Reaching for Objects

42. (Figure: Reaching for Objects) The type of graph depicted in this figure is a: a. bar graph. b. Pareto chart. c. time series plot. d. line graph.

Page 14

Name:

Class:

Date:

Chap_03_5e

This figure depicts the results of a study by Ahluwalia et al. (2006) investigating the effects of two types of gum (place and two types of counseling (educational or motivational) in helping people to stop smoking. Figure: Smoking Cessation

43. (Figure: Smoking Cessation) What is the difference between the data depicted in Figure A and the data in Figure B? a. Figure A does not graph the same independent variables as Figure B. b. Figure A does not graph the same dependent variable as Figure B. c. The bars in Figure A are more accurate than those in Figure B. d. The y-axis in Figure A has been shortened, which exaggerates the differences between the groups. 44. Scatterplots are often used to assess what type of pattern the data create. For example, when the data form a pattern that flows upward and toward the right, this is considered to be a(n): a. increasing nonlinear relation. b. negative nonlinear relation. c. positive linear relation. d. negative linear relation. 45. If graphing one scale independent variable and one scale dependent variable, use a: a. scatterplot or line graph. b. Pareto chart. c. histogram or frequency polygon. d. bar graph.

Page 15

Name:

Class:

Date:

Chap_03_5e 46. To efficiently survey attitudes of alumni, a university collects data on all those who attend the annual alumni reunion on campus. What type of manipulation is this? a. biased scale lie b. sneaky sample lie c. extrapolation lie d. interpolation lie 47. What type of lie is it when the method of assessment or measure has limited responses, thus creating an imbalance in possible responses? a. inaccurate values b. extrapolation c. biased scale d. interpolation 48. At major research universities, the intense pressure to publish has led a few people to act desperately, creating fake data to impress their reviewers and colleagues. This faking may be demonstrated in the visual distortion of data in graphing. This is an example of a(n) _____ lie. a. biased sample b. interpolation c. sneaky sample d. inaccurate values 49. A graph which summarizes the frequency of different words in a text response to a question, in which the size of the word indicates the frequency of the response is an example of a: a. bubble graph. b. word plot. c. pareto chart. d. word cloud.

Page 16

Name:

Class:

Date:

50. (Figure: Age and Reaction Time) Based on the graph, what type of relation exists between age and reaction time? a. linear b. nonlinear c. It is not possible to tell, given the information in the graph. d. There is no relation between age and reaction time.

Page 17

Name:

Class:

Date:

Chap_03_5e This figure depicts the results of a study by Shinskey and Munakata (2005) investigating the reaching behavior of seven 5-month-old infants. Figure: Reaching for Objects

51. (Figure: Reaching for Objects) What is the dependent variable in this study? a. type of object b. whether the object was visible or hidden c. proportion of trials on which infants reached d. familiarity of object 52. _____ refer(s) to any unnecessary information in a chart that is distracting. a. Chartjunk b. Grids c. Ducks d. Moirè vibrations 53. Cut marks are sometimes used on an axis to indicate that the axis starts at a value other than: a. zero. b. the lowest possible score. c. the largest possible score. d. a positive number. 54. Line graphs can offer additional information to scatterplots by including a line: a. that shows predicted y scores for each x value. b. of best fit rather than the individual data points. c. that shows predicted x scores for each y value. d. representing the relation of variables over time.

Page 18

Name:

Class:

Date:

Chap_03_5e This figure depicts data from the Schachter (1968) study on obesity and eating behavior. Figure: Taste Test

55. (Figure: Taste Test) Which statement best captures what the graph depicts? a. The number of crackers consumed by obese participants did not depend on whether their stomachs were full, but non-obese participants ate more crackers when their stomachs were full. b. The number of crackers consumed by obese participants did not depend on whether their stomachs were full, but non-obese participants ate fewer crackers when their stomachs were full. c. The number of crackers consumed by obese participants was much greater when their stomachs were full, but non-obese participants ate fewer crackers when their stomachs were full. d. The number of crackers consumed does not depend on the weight of participants or stomach condition. 56. Scatterplots are often used to assess what type of pattern the data create. For example, when the data form a pattern that flows downward and toward the right, this is considered to be a(n): a. increasing nonlinear relation. b. negative nonlinear relation. c. positive linear relation. d. negative linear relation. 57. If graphing two or more nominal independent variables and one scale dependent variable, use a: a. scatterplot or line graph. b. Pareto chart. c. histogram or frequency polygon. d. bar graph.

Page 19

Name:

Class:

Date:

Chap_03_5e 58. A time plot is also known as a: a. scatterplot. b. time series plot. c. line graph. d. bar graph. This figure was created from 2018 U.S. News & World Report data on 4-year colleges and universities in the United States. Figure: Cost of Tuition

59. (Figure: Cost of Tuition) The type of graph depicted in this figure is a: a. bar graph. b. Pareto chart. c. time plot. d. line graph.

Page 20

Name:

Class:

Date:

Chap_03_5e This graph depicts fictional data that replicate the pattern of performance observed in a study published by Nietfeld and Ender (2003). Figure: Intelligence and Teaching Efficacy

60. (Figure: Intelligence and Teaching Efficacy) If the graph creator asserts with certainty that a person with a Raven's intelligence test score of 5 would have a teaching efficacy score of 15, the statement would be which kind of lie? a. biased scale b. interpolation c. inaccurate values d. extrapolation 61. (Figure: Intelligence and Teaching Efficacy) The type of graph depicted in the figure is a: a. nonlinear relation plot. b. Pareto chart modified to dot form. c. scatterplot with a line of best fit. d. histogram. 62. Which of these would be appropriate to depict on a time series plot? a. life span duration for different states in the U.S. b. age of onset for schizophrenia males versus females c. time to graduate for different college majors d. amount of ice cream consumed by month of year 63. Pictorial graphs are sometimes used when the independent variables has _____ categories. a. only a few b. a great number of c. more than three d. scale measurement

Page 21

Name:

Class:

Date:

64. (Figure: Poverty and GPA) From this graph, one knows with certainty that students from a middle school with a 30 percent poverty rate would have an average GPA of 3.0 in 9th grade. This statement is an example of a(n) _____ lie. a. extrapolation b. sneaky sample c. interpolation d. inaccurate values 65. Any background pattern on which graphs appear is a form of chartjunk called: a. background noise. b. grids. c. moirè vibrations. d. ducks. 66. A cognitive psychologist investigated the relationship between IQ and years of education. What type of graph should the researcher use to illustrate his findings? a. bar graph b. scatterplot c. line graph d. frequency polygon Copyright Macmillan Learning. Powered by Cognero.

Page 22

Name:

Class:

Date:

Chap_03_5e 67. The _____ allows researchers to make predictions from a person's value on one variable to a person's value on another variable. a. linear relationship line b. nonlinear relationship line c. line of best fit d. time plot line 68. When constructing a graph, it is better to make one's own, intentional decisions about its appearance than to rely on computer: a. defaults. b. multivariable graphs. c. programs. d. interactive graphs. 69. On a histogram the y-axis depicts counts or frequencies, while on a bar graph the y-axis depicts _____. a. levels of a nominal variable b. counts or percentages c. the independent variable d. the dependent variable 70. A bar graph summarizing the height of buildings in a major city depicts the bars as three-dimensional skyscrapers. The appearances of the bars qualify as which type of chartjunk? a. grids b. moirè vibrations c. ducks d. defaults

Page 23

Name:

Class:

Date:

Chap_03_5e This figure depicts the results of a study by Shinskey and Munakata (2005) investigating the reaching behavior of seven 5-month-old infants. Figure: Reaching for Objects

71. (Figure: Reaching for Objects) What are the independent variables in this study? a. visible object and hidden object b. type of object and whether the object was visible or hidden c. proportion of trials on which infants reached and type of object d. novel object and familiar object This figure depicts data from the Schachter (1968) study on obesity and eating behavior. Figure: Taste Test

72. (Figure: Taste Test) What is the dependent variable in this study? a. weight of participant b. stomach condition c. number of crackers consumed d. obesity

Page 24

Name:

Class:

Date:

73. (Figure: Conscientiousness and Test Performance) Given the construction of this graph, what is the dependent variable? a. conscientiousness percentile b. test grade c. percentile rank d. There is not enough information in the graph to determine the dependent variable. 74. Line graphs are used to represent the relation between two scale variables. Two common types of relations depicted in these graphs are: a. the predicted value of a dependent variable for each value of an independent variable and the change in a variable over time. b. linear relations and nonlinear relations. c. the predicted value of a dependent variable for each value of an independent variable and the line of best fit. d. time and change, as well as nonlinear relations. 75. If graphing one nominal independent variable and one scale dependent variable, use a: a. scatterplot or line graph. b. frequency polygon. c. histogram. d. bar graph.

Page 25

Name:

Class:

Date:

Chap_03_5e 76. In a pie chart, the slices represent: a. categories of the dependent variable. b. levels of the independent variable. c. values that would appear on the y-axis. d. more than bars on a bar chart represent. 77. What is a Pareto chart? a. a bar graph with the bars ordered from highest (on the left) to lowest (on the right) b. a line graph with bars depicted in three dimensions c. a bar graph in which all the bars are stacked on top of one another d. a pictorial graph that uses a symbol to represent the scale dependent variable This figure was created from 2018 U.S. News & World Report data on 4-year colleges and universities in the United States. Figure: Cost of Tuition

78. (Figure: Cost of Tuition) The dependent variable in this study was: a. private colleges. b. public colleges. c. the type of college. d. the cost of tuition and fees.

Page 26

Name:

Class:

Date:

79. (Figure: Poverty and GPA) The type of graph depicted in the figure is a: a. scatterplot. b. histogram. c. Pareto chart. d. range-frame. 80. (Figure: Poverty and GPA) Given the construction of this graph, what is the dependent variable? a. poverty level b. freshman class rank c. 9th grade GPA d. number of middle schools 81. A graph that depicts the relation between two scale variables is a: a. frequency table. b. bar graph. c. time series plot. d. scatterplot.

Page 27

Name:

Class:

Date:

Chap_03_5e 82. The interpolation lie can involve: a. using a biased scale. b. using a biased sample. c. graphing extreme data points and assuming a straight line between them. d. using labels on the graph that do not precisely reflect what data the graph actually display. 83. A psychologist was interested in measuring reaction time differences between men and women on a computerized task. What is the dependent variable? a. researcher b. computerized task c. reaction time d. gender 84. In a pictorial graph, images are used at each level of the _____ variable to represent its value on the _____ variable. a. independent; dependent b. dependent; independent c. scale; nominal d. scale; dependent 85. A graph which includes information about response variability by displaying the middle score in a dataset along with indicators of the range of the middle 50% and 95% of scores is an example of a: a. bubble graph. b. violin plot. c. GIS chart. d. word cloud. 86. A bubble graph is one that: a. is similar to a pie chart but is depicted as a bubble. b. includes captions to explain each data point. c. can display multiple variables. d. portrays data such as income and life expectancy.

Page 28

Name:

Class:

Date:

Chap_03_5e Section: These figures represent the same set of data found at nationmaster.com. Each figure represents the amount of money (in U.S. dollars) that five countries have pledged for Iraq reconstruction as of December 2005. Figure: Iraq Reconstruction

87. (Figure: Iraq Reconstruction) Which graph depicts the data more clearly and is preferred by researchers? a. The pie chart (Figure B) depicts the data more clearly and is preferred by researchers. b. The pie chart (Figure B) depicts the data more clearly, but the bar graph is preferred by researchers. c. The bar graph (Figure A) depicts the data more clearly and is preferred by researchers. d. The bar graph (Figure A) depicts the data more clearly, but the pie chart is preferred by researchers.

Page 29

Name:

Class:

Date:

Chap_03_5e This figure depicts the average IQ of the population in each of the 50 states. Figure: State IQ

88. (Figure: State IQ) What type of variable is depicted in the graph? a. nominal b. ordinal c. scale d. Pareto

Page 30

Name:

Class:

Date:

89. (Figure: Conscientiousness and Test Performance) The type of graph depicted in the figure is a: a. scatterplot. b. histogram. c. Pareto chart. d. time plot. 90. The first step in creating a scatterplot is to: a. make a mark on the graph above each study participant's score on the x- and y-axes. b. label the horizontal x-axis with the name of the independent variable and its possible values. c. label the vertical y-axis with the name of the dependent variable and its possible values. d. organize the data by participant.

Page 31

Name:

Class:

Date:

Chap_03_5e

91. (Figure: Smoking Cessation) Which Figure, A or B, would be the best figure to use when presenting the results of this study and why? a. Either graph would be perfectly acceptable for presenting the results of the study. b. Figure A would be the best because it depicts the tallest bars. c. Figure B would be the best because the y-axis depicts the actual possible values on the dependent measure. d. Neither graph would be acceptable when presenting the results of the study. 92. On a histogram the x-axis depicts the values of a scale variable, while on a bar graph the x-axis depicts the values of _____ variable. a. the dependent b. a scale c. a nominal or ordinal d. counts or percentages 93. When the data on a scatterplot form a pattern that breaks or bends to form a curve, these data are considered to be related in a(n) _____ way. a. unmeaningful b. nonlinear c. unpredictable d. linear Copyright Macmillan Learning. Powered by Cognero.

Page 32

Name:

Class:

Date:

Chap_03_5e This figure was created from 2018 U.S. News & World Report data on 4-year colleges and universities in the United States. Figure: Cost of Tuition

94. (Figure: Cost of Tuition) Which statement BEST captures what the graph depicts? a. Tuition at private universities has skyrocketed. b. Tuition at private universities is, on average, over $25,000 more than in-state tuition at public universities. c. Public universities are really inexpensive. d. The amount of tuition paid does not depend on the type of college one attends.

Page 33

Name:

Class:

Date:

95. (Figure: Iraq Reconstruction) How could the bar graph in Figure A be turned into a Pareto chart? a. Divide the value of the dependent variable by the mean of the dependent variable. b. Stack all the bars on top of each other. c. Make each bar three-dimensional. d. Reorder the bars from highest (Kuwait) to lowest (Denmark). 96. A survey of students on your university website for athletics asks, "How high do you think the football team will finish this year: 1st, 2nd, 3rd, 4th, or 5th?" An article posted a week later reports "Students predict football team to finish within the top 5 out of 12 teams in the conference." What kind of manipulation occurred here? a. an outright lie b. false face validity c. biased scale lie d. interpolation lie

Page 34

Name:

Class:

Date:

Chap_03_5e This figure depicts data from the Schachter (1968) study on obesity and eating behavior. Figure: Taste Test

97. (Figure: Taste Test) The type of graph depicted in this figure is a: a. bar graph. b. Pareto chart. c. histogram. d. line graph. This figure depicts the typical relation found between a person's age (in years) and his or her reaction time in a simple reaction time task (e.g., time to press a button when a light turns green). Figure: Age and Reaction Time

98. (Figure: Age and Reaction Time) If one were to assume with certainty that a person who is 2 years old has a reaction time of 9,000 milliseconds, which error would be made? a. sneaky sample lie b. interpolation lie c. extrapolation lie d. inaccurate values lie Copyright Macmillan Learning. Powered by Cognero.

Page 35

Name:

Class:

Date:

Chap_03_5e 99. The final step in creating a scatterplot is to: a. make a mark on the graph above each study participant's score on the x- and y-axes. b. label the horizontal x-axis with the name of the independent variable and its possible values. c. label the vertical y-axis with the name of the dependent variable and its possible values. d. organize the data by participant. 100. A social psychologist labels the dependent variable of a bar graph Marital Distress, when what was measured was the number of arguments among couples attending marriage counseling. This misnomer is an example of the _____ lie. a. interpolation b. inaccurate values c. extrapolation d. sneaky sample Enter the appropriate word(s) to complete the statement. 101. The Yerkes–Dodson curve is an example of a(n) _______ relation.

102. A line graph helps us predict _______ scores.

103. The _______ lie involves using preselected participants who are not representative of the true population.

104. In a(n) _______ graph images are used at each level of the independent variable to represent its value on the dependent variable.

Page 36

Name:

Class:

Date:

Chap_03_5e 105. A(n) _______ is a type of multivariable graph that allows you to plot numerous variables simultaneously.

106. Features of the data that have been dressed up to be something other than merely data are _______.

107. Grids, moiré vibrations, and ducks are all forms of _______.

108. When plotting the x-axis and the y-axis, if practical, it is best to start with a value of _______.

109. Not providing a range of responses on a measurement is a type of _______ lie.

110. Patterns that computer programs provide as options to fill in the bars of a graph are called _____.

111. When constructing a graph, it is better to make one's own, intentional decisions about its appearance than to rely on_______.

Page 37

Name:

Class:

Date:

Chap_03_5e 112. Unnecessary information or a feature in a graph that detracts from a viewer's ability to understand the data is _______.

113. Data can almost always be presented more clearly in a table or bar graph than in a(n) _______.

114. A line on a graph for predicting a dependent variable from an independent variable is the _______.

115. A(n) _______ plot is a type of graph that provides information about a distribution's middle score, range, and overall variability.

116. When graphing two scale variables you should use a(n) _______.

117. When graphing a single scale variable, you should use a(n) _______.

118. Interval and ratio variables are also referred to as _______ variables.

Page 38

Name:

Class:

Date:

Chap_03_5e 119. A type of bar graph in which the categories along the x-axis are ordered from the highest bar on the left to the lowest bar on the right is a(n) _______.

120. When graphing a nominal independent variable and a scale-dependent variable, use a(n) _______.

121. A bubble graph that allows one to see the relations amongst a number of different variables with the size and color of the bubbles conveying different types of information is an example of a(n) _______ graph.

122. For time-related data, a(n) _____ is more useful than a(n) _____.

123. _______ allow one to connect points on a map with data points located in space, such as homes for sale.

Page 39

Name:

Class:

Date:

Chap_03_5e Mehl and colleagues (2007) reported in the journal Science the results of an extensive study of 396 men and women, comparing the number of words uttered per day by each. Volunteer participants wore inconspicuous recording devices that recorded their daily word usage. Is there any validity to the notion that women talk more than men do? The following fictional data represent the number of words uttered by five women and five men. Table: Number of Words Uttered Women 17,214 15,325 14,022 18,643 15,800

Men 16,322 14,636 17,045 18,873 13,071

124. (Table: Number of Words Uttered) Graph and interpret these results.

125. In 1968, Schachter published an article in the journal Science reporting a series of experiments on obesity and eating behavior. Participants were led to believe that they were taking part in a "taste test." All participants were told to not eat for several hours before the start. Participants were classified on the basis of their weight (obese and non-obese) and randomly assigned to the full stomach or empty stomach condition (half the participants were given a full meal, as much as they wanted, after arriving at the "test site," and half were left hungry). The researchers recorded the number of crackers eaten by each participant during the fake "taste test." The averages for each of the groups are as follows: Non-Obese, Empty Stomach: average score = 22; Obese, Empty Stomach: average score = 17; Non-Obese, Full Stomach: average score = 15; Obese, Full Stomach: average score = 18. (a) Use these data to create the appropriate graph of the results. (b) Interpret the results of the study on the basis of the graph.

Page 40

Name:

Class:

Date:

Chap_03_5e 126. There are many techniques for misleading with graphs. Read the following example: A seventh-grade teacher wanted to examine how effective her teaching was before formal end-of-year evaluations. To do so, she devised her own scale with questions on specific lessons, her teaching style, and her effectiveness overall. She used a 6-point scale as follows: 1 = very poor, 2 = neutral, 3 = good, 4 = very good, 5 = excellent, 6 = superior. What type of lie is MOST likely being depicted in this example? Explain.

Figure: What Type of Chocolate Do People Prefer?

127. (Figure: What Type of Chocolate Do People Prefer?) What is wrong with the pie chart? Use concepts learned in this chapter as well as any other ideas you may have about potentially misleading aspects of this chart.

Page 41

Name:

Class:

Date:

Chap_03_5e The following data extracted from the Internet Movie Database (IMDB) reflect the box office earnings and ratings of movies for the weekend of July 13, 2007. Table: Movie Rankings Movie Ratatouille Sicko Knocked Up Transformers Live Free or Die Hard Harry Potter and the Order of the Phoenix 1408 Ocean's Thirteen Evan Almighty License to Wed

Earning Rank 3 9 8 2 4 1 6 10 7 5

Rating Rank 1 2 3 4.5 4.5 6 7 8 9 10

128. (Table: Movie Rankings) (a) Construct the appropriate graph for depicting these results. (b) What is the relation depicted in the graph?

Page 42

Name:

Class:

Date:

Chap_03_5e The following graph reports fictional data on students' preferences for various school subjects. Figure: Student Subject Preferences

129. (Figure: Student Subject Preferences) What can you infer from the graph? How could the layout of the graph be improved? Are there any potentially misleading aspects?

Page 43

Name:

Class:

Date:

Chap_03_5e The following graph depicts fictional data of participants' continuous performance on a memory task. The results plot participants' accuracy score on the y-axis and the time of day on the x-axis. Figure: Memory Performance

130. (Figure: Memory Performance) What type of graph is shown? What can be inferred from the graph?

131. The following fictional data are the ages at which 40 people received the black belt in karate.

(a) Use the data to construct a histogram. (b) Explain why a histogram is more appropriate for these data than a bar graph.

Page 44

Name:

Class:

Date:

Chap_03_5e The following figure appeared on the Statistics Canada website. Evaluate the figure. Figure: Absenteeism Rates Up

132. (Figure: Absenteeism Rates Up) What aspects of the figure should be changed so that it more clearly depicts the data?

133. The line of best fit permits one to make predictions for a value on the y variable from the value on the x variable. Can the line of best fit be applied to both linear and nonlinear relations? Why or why not?

Page 45

Name:

Class:

Date:

Chap_03_5e 134. To contradict the finding that depression sometimes occurs more often for some people during winter months, possibly as part of seasonal affective disorder, several researchers wanted to show that many people actually prefer the winter months over the summer months. They decide to use a biased-scale and sneaky sample to skew their findings toward their desired position, that winter is wonderful. Describe each of these methods for misleading with data and graphs and how they might be used by these researchers. Finally, explain why each is a serious problem for research.

Page 46

Name:

Class:

Date:

Chap_03_5e Answer Key 1. False 2. False 3. True 4. False 5. True 6. False 7. False 8. False 9. True 10. False 11. False 12. True 13. True 14. True 15. True 16. True 17. False 18. True 19. False 20. False 21. False 22. d 23. c 24. b 25. c 26. a Copyright Macmillan Learning. Powered by Cognero.

Page 47

Name:

Class:

Date:

Chap_03_5e 27. a 28. c 29. a 30. b 31. a 32. c 33. b 34. c 35. d 36. d 37. c 38. c 39. b 40. c 41. c 42. a 43. d 44. c 45. a 46. b 47. c 48. d 49. d 50. b 51. c 52. a 53. a 54. a Copyright Macmillan Learning. Powered by Cognero.

Page 48

Name:

Class:

Date:

Chap_03_5e 55. b 56. d 57. d 58. b 59. a 60. d 61. c 62. d 63. a 64. c 65. b 66. b 67. c 68. a 69. b 70. c 71. b 72. c 73. b 74. a 75. d 76. b 77. a 78. d 79. a 80. c 81. d 82. c Copyright Macmillan Learning. Powered by Cognero.

Page 49

Name:

Class:

Date:

Chap_03_5e 83. c 84. a 85. b 86. c 87. c 88. c 89. a 90. d 91. c 92. c 93. b 94. b 95. d 96. c 97. a 98. c 99. a 100. d 101. nonlinear 102. y, dependent 103. sneaky sample 104. pictorial 105. bubble graph 106. ducks 107. chartjunk 108. 0, zero 109. biased scale 110. Copyright Macmillan Learning. Powered by Cognero.

Page 50

Name:

Class:

Date:

Chap_03_5e 111. computer defaults, defaults 112. chartjunk 113. pie chart 114. line of best fit 115. violin 116. scatterplot or line graph 117. histogram or frequency polygon 118. scale 119. Pareto chart 120. bar graph 121. multivariable 122. time plot, time series plot; scatterplot 123. Geographic information systems (GIS) 124. (Figure: Number of Words Uttered) Following is a bar graph of the means. The bar graph indicates no difference between the number of words uttered by men and women. Figure: Number of Words Uttered

Page 51

Name:

Class:

Date:

Chap_03_5e 125. (a) (Figure: Taste Test) The bar graph of these data follows. Figure: Taste Test

(b) The graph indicates that the number of crackers consumed by obese participants did not depend on whether their stomachs were full. But non-obese participants ate fewer crackers when their stomachs were full. 126. This example illustrates the biased scale lie. Notice that there are far more positive categories than there are negative ones, which is likely to produce a bias in the direction of more favorable ratings for the teacher. 127. Some possible answers include: Because of the profound limitations of pie charts, the data could more clearly be presented in either a table or a bar chart. Second, the pie chart is suited for only a small number of groups. When the number of groups increases, the magnitude of difference in the slices is more difficult to interpret. The pie chart does not indicate the number of participants sampled in the study or the number of participants per group, which could be misleading. Further, a sneaky sample might come into play here, allowing the researchers to portray data as they wish them to appear. 128. (a) (Figure: Movie Rankings) The appropriate graph is the following scatterplot. Figure: Movie Rankings

(b) There appears to be little to no relation between ranking and earnings. Copyright Macmillan Learning. Powered by Cognero.

Page 52

Name:

Class:

Date:

Chap_03_5e 129. From this graph, it appears that most students prefer science and art to the rest of the subjects. Physical education and language arts were equally liked and were the second highest rated subjects. The graph suggests that math was the least preferred subject. An improvement to this graph might be labeling the x-axis and y-axis for clarity. Also, ordering the bars from highest to lowest by creating a Pareto chart would make the data easier to interpret. Potentially misleading aspects of this graph are that the scale has been changed to a range of 20ñ70. Whenever possible, it is best to have the lowest score in the scale at 0 and the highest at 100. 130. The graph shown is a line graph. Specifically, it is a time plot, or time series plot. The graph depicts the relationship between time of day and memory performance. From the results, it appears that participant's memory performance is better in the morning and evening hours. This graph suggests that memory performance overall appears to be affected by the time of day. 131. (a) (Figure: Histogram in SPSS) Following is a histogram produced in SPSS: Figure: Histogram in SPSS

(b) A histogram is appropriate to display the frequency observations of one scale variable, age. A bar graph is used to display levels of an ordinal or nominal independent variable on the x-axis and their values on the scale dependent variable on the y-axis. 132. There are several problems with the graph. First, the subtitle of the graph says that workdays missed rose between 1997 and 1998, but the data in the graph do not depict this finding. There are a number of ducks in the graph: The picture of the calendar is unnecessary. Exactly what is being graphed is also unclear. The top of the graph specifies "Workdays missed," but the bottom of the graph shows percentages. The graph does not indicate what these percentages refer to. Also, the scale for the percentage does not go to 0, thus making the differences seem larger than they actually are. 133. The goal of the line of best fit is to find the middle position on the points on the scatterplot formed by a linear relation. As you will learn in the later chapter on regression, the line of best fit is used in linear regression analysis to help understand the accuracy of a regression equation, but it focuses on only one type of relation, a linear relation. Because nonlinear relations can form many different types of patterns, the line of best fit would not accurately capture the relations formed by the data points and would result in either erroneous predictions or no predictions at all. Copyright Macmillan Learning. Powered by Cognero.

Page 53

Name:

Class:

Date:

Chap_03_5e 134. Some possible answers include: The biased-scale lie involves using a measurement tool that is slanted in the direction of interest. For example, asking how much someone likes winter with a scale from (1) very little, (2) a little, (3) moderately, and (4) a great deal does not allow for much dislike of winter. Because of its restricted response categories, a biased-scale does not allow true opinions of participants to be expressed. A sneaky sample is one that is preselected to contain participants who will express the opinions (or exhibit the behaviors) desired by the researchers. In this scenario, the researchers could include only avid skiers in their sample, resulting in inflated ratings of enjoyment of winter months. Any bias in your sample will result in reduced representativeness and compromise your ability to infer things about the greater population.

Page 54

Name:

Class:

Date:

Chap_04_5e Indicate whether the statement is true or false. 1. The interquartile range measures variability as the difference between the scores at the 75th and 25th percentiles. a. True b. False 2. The interquartile range measures variability as the difference between the scores at the 25th and 100th percentiles. a. True b. False 3. The median is the measure of central tendency that conveys the mathematical center of the data. a. True b. False 4. The range is more susceptible to the influence of outliers than other measures of variability. a. True b. False 5. The standard deviation is equal to the square root of the variance. a. True b. False 6. The mean is the measure of central tendency that conveys the middle score of the data. a. True b. False 7. The sum of squares is symbolized as Σ. a. True b. False 8. It is easier to distort and lie with the median than with the mean. a. True b. False 9. Numbers based on populations are called statistics. a. True b. False 10. The sum of squares is symbolized as SD. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_04_5e 11. The mean is the measure of central tendency that conveys the mathematical center of the data. a. True b. False 12. The range is less susceptible to the influence of outliers than other measures of variability. a. True b. False 13. The mean is an appropriate measure of center when considering nominal data. a. True b. False 14. The standard deviation is expressed in squared units. a. True b. False 15. Compared to the interquartile range, the range is less susceptible to the influence of outliers. a. True b. False 16. The mode is an appropriate measure of center when considering nominal data. a. True b. False 17. The most common measure of variability reported in research articles is the standard deviation. a. True b. False 18. The interquartile range measures variability as the difference between the scores at the 25th and 50th percentiles. a. True b. False 19. The median is preferred over the mean for skewed distributions. a. True b. False 20. It is easier to distort and lie with the mean than with the median. a. True b. False

Page 2

Name:

Class:

Date:

Chap_04_5e Indicate the answer choice that best completes the statement or answers the question. This table represents the fictional scores of a set of participants who rated their optimism levels on a scale from 1 to 7, with 1 indicating very unoptimistic and 7 indicating very optimistic. Table: Optimism X 1 2 3 4 5 6 7

Rating 1 7 6 7 7 6 5

21. (Table: Optimism) The variance in this data set is: a. 3.96. b. 5.57. c. 5.40. d. 6.11. 22. In an introductory psychology course, the average score on the first exam across all sections of the course was 76.93 percent. Because all students were included in the calculation, the mean is assumed to be based on population data. The test average may be considered to be a: a. measure of variability. b. statistic. c. parameter. d. mode. 23. Around the end-of-year holidays, such as Thanksgiving, a great amount of meat is purchased. To be prepared for the increased demand, a local grocery store manager wants to know the typical amount of ham sold at medium-sized stores around Thanksgiving. But even more than that, she wants to know how sales tend to vary around that typical amount because he wants to be prepared to sell as many hams as possible. Her interest in how much sales vary will require a measure of: a. central tendency. b. variability. c. parameters. d. statistics.

Page 3

Name:

Class:

Date:

Chap_04_5e 24. In an introductory psychology course, the average score on the first exam was 78 percent based on a sample of 146 students. This mean was calculated on the class grades to summarize overall performance. There are two other sections of the same class, but those students were not included in the calculation. The test average is a: a. measure of variability. b. statistic. c. parameter. d. mode. 25. _____ is affected by outliers because it takes the actual value of each data point into consideration. a. Mode b. Median c. Mean d. Center 26. Fifteen college freshmen were asked to record the number of alcoholic drinks they typically consume in a week. Here are their data: 1, 4, 6, 0, 1, 9, 0, 6, 3, 6, 8, 5, 4, 7, 2. What is the mode of this distribution? a. 2 b. 4 c. 4.5 d. 6 27. Σ(X – M) is always equal to: a. the sum of squares. b. the mean. c. 0. d. 1.0. 28. A major limitation of the range as a measure of variability is that it: a. can only be applied to nominal data. b. will exaggerate the variability if there are outliers in the data set. c. must be applied to scale data. d. is likely to underestimate the variability in a data set. 29. The second quartile, or the 50th percentile of a distribution, is also known as: a. the middle 50. b. Q1. c. the mean. d. the median.

Page 4

Name:

Class:

Date:

Chap_04_5e 30. If an extreme score is added to a data set, increasing the range of data, the standard deviation will: a. get closer to 0. b. decrease. c. increase. d. double. This table represents the fictional scores of a set of participants who rated their depression levels for the past two weeks on a scale from 1 to 4, 1 indicating No Depression and 4 indicating Depressed Most of the Time. Table: Depression X 1 2 3 4 5 6 7 8 9

Rating 4 2 2 1 3 1 3 4 1

31. (Table: Depression) The range in this data set is: a. 1.15. b. 1.32. c. 3.00. d. 4.00. 32. Fifteen college freshmen were asked to record the number of alcoholic drinks they typically consume in a week. Here are their data: 1, 4, 6, 0, 1, 9, 0, 6, 3, 6, 8, 5, 4, 7, 2. What is the variance of the number of alcoholic drinks consumed per week? a. 2.80 b. 3.26 c. 7.85 d. 12.25

Page 5

Name:

Class:

Date:

Chap_04_5e 33. If the standard deviation of some data is 0.447, what is the variance? a. 0.45 b. 0.20 c. 0.67 d. It is impossible to determine the variance based on the information provided. 34. If you knew nothing about your data except that it is a scale variable, which measure of central tendency would be the "safest bet"? a. mode b. median c. mean d. center 35. A bimodal distribution has _____ mode(s). a. one b. two c. two or more d. more than two

Page 6

Name:

Class:

Date:

Chap_04_5e This table represents the fictional scores of a set of participants who rated their depression levels for the past two weeks on a scale from 1 to 4, 1 indicating No Depression and 4 indicating Depressed Most of the Time. Table: Depression X 1 2 3 4 5 6 7 8 9

Rating 4 2 2 1 3 1 3 4 1

36. (Table: Depression) The variance in this data set is: a. 1.15. b. 1.32. c. 2.33. d. 3.00. 37. In a distribution that is skewed by a few extreme outliers, what would be the best choice for a measure of central tendency? a. median b. mode c. mean d. range 38. The median is preferred over the mean for _____ distributions. a. normal b. unimodal c. skewed d. symmetrical

Page 7

Name:

Class:

Date:

Chap_04_5e This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

39. (Table: Age) The standard deviation in this data set is: a. 6.14. b. 6.91. c. 7.44. d. 55.33. 40. Here is a set of data: 45, 52, 51, 37, 49, 55, 47, 39. Compute a deviation score for the data point 51. a. 0 b. 4.13 c. 15.00 d. 46.88 41. The interquartile range is the _____ percent of the data. a. middle 50 b. top 25 c. middle 30 d. bottom 25 42. If the variance for a sample is computed and it is found to be rather large, the numbers in the sample are: a. tightly packed at one extreme. b. tightly packed around the mean. c. spread out around the mean. d. close to the extremes of the distribution.

Page 8

Name:

Class:

Date:

Rating 4 2 2 1 3 1 3 4 1

43. (Table: Depression) The median depression rating in this data set is: a. 1.00. b. 1.15. c. 2.00. d. 2.33. 44. The concept of central tendency is best described as: a. the average. b. how the data are distributed. c. the physical middle of our data. d. the value around which the data seem to gather. 45. Six students from a psychology class reported the number of hours of television they watched per week. Here are their data: 12, 12, 11, 14, 13, 17. What is the mean number of hours of television watched per week for this sample of six students? a. 1.95 b. 6.00 c. 12.50 d. 13.17

Page 9

Name:

Class:

Date:

Chap_04_5e This table represents the fictional scores of a set of participants who rated their optimism levels on a scale from 1 to 7, with 1 indicating very unoptimistic and 7 indicating very optimistic. Table: Optimism X 1 2 3 4 5 6 7

Rating 1 7 6 7 7 6 5

46. (Table: Optimism) The range in this data set is: a. 3.96. b. 5.57. c. 6.00. d. 7.00. 47. A multimodal distribution has _____ mode(s). a. one b. two c. two or more d. more than two 48. Variance is the average of deviations around the mean expressed in _____ units. a. standardized b. absolute value c. squared d. logarithmic 49. The mean of the population is represented by the symbol _____, and the mean of the sample is represented by the symbol _____. a. M; µ b. µ; M c. µ; µ d. M; M

Page 10

Name:

Class:

Date:

Rating 1 7 6 7 7 6 5

50. (Table: Optimism) The mean in this data set is: a. 3.96. b. 5.57. c. 6.00. d. 7.00. 51. Here is a set of data: 31, 53, 27, 42, 39, 46, 37, 42. What is the mean? a. 26.00 b. 28.00 c. 39.63 d. 52.84 52. Here is a set of scores: 17, 98, 14, 18, 31, 43, 12, 14, 13, 51, 12. Which score is considered to be an outlier? a. 12 b. 14 c. 51 d. 98

Page 11

Name:

Class:

Date:

Rating 4 2 2 1 3 1 3 4 1

53. (Table: Depression) The standard deviation in this data set is: a. 1.15. b. 1.32. c. 2.33. d. 3.00.

Page 12

Name:

Class:

Date:

Chap_04_5e This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

54. (Table: Age) The median age in this data set is: a. 33.00. b. 36.67. c. 39.00. d. 40.00. 55. (Table: Age) The mean age in this data set is: a. 33.00. b. 36.67. c. 39.00. d. 40.00. 56. Fifteen college freshmen were asked to record the number of alcoholic drinks that they typically consume in a week. Here are their data: 1, 4, 6, 0, 1, 9, 0, 6, 3, 6, 8, 5, 4, 7, 2. What is the median number of alcoholic drinks consumed per week? a. 4.0 b. 4.5 c. 5.0 d. 6.0

Page 13

Name:

Class:

Date:

Chap_04_5e 57. If you wanted to know whether judges in your state are typically male or female, what measure of central tendency would you use to describe the typical judge? a. mean b. median c. mode d. range 58. In a sample of data that has several extremely low scores, creating negative skew, the mean will be _____ the median. a. less than b. greater than c. equal to d. double This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

59. (Table: Age) Which score in the distribution is considered to be an outlier? a. 33 b. 18 c. 44 d. 42 60. Numbers that describe populations are called: a. statistics. b. parameters. c. averages. d. variables. Copyright Macmillan Learning. Powered by Cognero.

Page 14

Name:

Class:

Date:

Chap_04_5e 61. If the variance of some data is 6.74, what is the standard deviation? a. 2.59 b. 3.37 c. 45.43 d. It is impossible to determine the standard deviation based on the information provided. 62. Fifteen college freshmen were asked to record the number of alcoholic drinks that they typically consume in a week. Here are their data: 1, 4, 6, 0, 1, 9, 0, 6, 3, 6, 8, 5, 4, 7, 2. What is the mean number of alcoholic drinks consumed per week? a. 4.00 b. 4.13 c. 4.77 d. 6.00 63. If the score at the 75th percentile is a 12 and the score at the 25th percentile is a 2, then the interquartile range is: a. 7. b. 8. c. 10. d. 14. 64. A standard deviation is equal to 4.50. This number means that the numbers in the sample deviate, on the average: a. 2.25 units from the mean. b. 4.50 units from each other. c. 4.50 units from the mean. d. 4.50 units from the two extreme scores.

Page 15

Name:

Class:

Date:

Chap_04_5e This histogram represents the distribution of the number of years of education completed by twins who attended the 16th Annual Twins Day Festival in Twinsburg, Ohio, in August of 1991. Figure: Years of Education

65. (Figure: Years of Education) What is the range of this distribution? a. 10 b. 12 c. 14 d. 20 66. A sample of 140 women results in an average weight calculation of 158 pounds. This average weight is a: a. population parameter. b. standardized measure. c. sum of squares. d. sample statistic.

Page 16

Name:

Class:

Date:

Rating 1 7 6 7 7 6 5

67. (Table: Optimism) The mode in this data set is: a. 3.96. b. 5.57. c. 6.00. d. 7.00. 68. In a sample of data that has several extremely high scores, creating positive skew, the mean will be _____ the median. a. less than b. greater than c. equal to d. double 69. Leah observes and records the number of people who purchase a meal at the school cafeteria during each operating hour. The cafeteria is open from 6:00 a.m. to 9:00 p.m. and students typically eat breakfast, lunch, and dinner. What type of distribution should Leah expect to see in her data? a. normal b. unimodal c. bimodal d. multimodal 70. In a sample of data that has several extremely low scores, creating negative skew, the mean: a. will be greater than the mode. b. will be greater than the median. c. and the mode will be equal. d. will be less than the median. Copyright Macmillan Learning. Powered by Cognero.

Page 17

Name:

Class:

Date:

Chap_04_5e 71. When calculating the median for an odd number of scores, which have been placed in ascending order, the median will be the: a. middle score. b. difference between the highest and lowest scores. c. average of the three scores in the middle. d. average of the two scores in the middle. 72. Fifteen college freshmen were asked to record the number of alcoholic drinks they typically consume in a week. Here are their data: 1, 4, 6, 0, 1, 9, 0, 6, 3, 6, 8, 5, 4, 7, 2. What is the range of this distribution? a. 0 to 9 b. 1 c. 8 d. 9 73. Here is a set of data: 33, 54, 25, 42, 39, 45, 38, 42, 48. What is the median? a. 25.0 b. 40.5 c. 42.0 d. 54.0 This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

74. (Table: Age) The modal age (or mode) in this data set is: a. 36.67. b. 39.00. c. 40.00. d. 42.00. Copyright Macmillan Learning. Powered by Cognero.

Page 18

Name:

Class:

Date:

Chap_04_5e 75. What is the basic formula for the standard deviation? a. SD = √SD2 b. range = Xhighest – Xlowest c. SD2 = √SD d. 76. Here is a set of data: 30, 57, 26, 42, 39, 45, 38, 42. What is the median? a. 28.0 b. 40.5 c. 42.0 d. 57.0 77. Which set of scores has the least amount of variability? a. 22, 26, 21, 23 b. 32, 72, 86, 100 c. 5, 18, 62, 78 d. 101, 110, 56, 13 This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

78. (Table: Age) The range in this data set is: a. 7. b. 8.5. c. 13. d. 26. Copyright Macmillan Learning. Powered by Cognero.

Page 19

Name:

Class:

Date:

Chap_04_5e 79. The mean is calculated by: a. determining the midpoint of the scores in the data set, such that half fall above the mean and half fall below. b. summing all of the scores in a data set and then dividing by the total number of scores. c. subtracting the lowest score from the highest score in the data set. d. determining the most common score in a data set. 80. Here is a set of data: 45, 52, 51, 37, 49, 55, 47, 39. Compute a deviation score for the data point 44. a. –4.16 b. –2.88 c. 0.62 d. 1.74 81. The _____ is the difference between the lowest and highest scores in the distribution. a. standard deviation b. range c. variance d. outlier

Page 20

Name:

Class:

Date:

Rating 4 2 2 1 3 1 3 4 1

82. (Table: Depression) The mean depression rating in this data set is: a. 1.00. b. 1.15. c. 2.00. d. 2.33.

Page 21

Name:

Class:

Date:

Chap_04_5e This table represents the fictional ages of a set of participants who participated in a research study. Table: Age X 1 2 3 4 5 6 7 8 9

Age 34 42 40 38 42 39 18 44 33

83. (Table: Age) The variance in this data set is: a. 7.44. b. 26.00. c. 55.33. d. 62.25. 84. Six students from a psychology class reported the number of hours of television they watched per week. Here are their data: 12, 12, 11, 14, 13, 17. What is the median number of hours of television watched per week for this sample of six students? a. 12.0 b. 12.5 c. 13.0 d. 17.0 85. If the standard deviation of some data is 6.19, what is the range? a. 0.93 b. 2.49 c. 38.32 d. It is impossible to determine the range based on the information provided.

Page 22

Name:

Class:

Date:

Chap_04_5e 86. Six students from a psychology class reported the number of hours of television they watch per week. Here are their data: 11, 12, 15, 14, 13, 16. What is the standard deviation of the number of hours of television watched per week for this sample of six students? a. 1.71 b. 2.92 c. 10.26 d. 17.54 87. Six students from an economics class reported the number of hours of television they watch per week. Here are their data: 18, 12, 15, 14, 13, 16. The variance of the number of hours of television watched per week for this sample of six students is: a. 1.97. b. 3.89. c. 15.13. d. 23.33. 88. At 27 years of age, Cara is 5 feet, 6 inches tall. The national average for height is 5 feet, 4 inches, so Megan is taller than average. The national average is a: a. population parameter. b. standardized measure. c. sum of squares. d. sample statistic. 89. A marketing manager for a major department store wants to know how much time people tend to spend in the store so that the audio advertisements will replay on a timed loop, allowing every customer a chance to hear each unique message. What information is this marketing specialist seeking? a. variability b. deviation scores c. central tendency d. parameter 90. Here is a set of data: 30, 55, 27, 42, 39, 45, 38, 42. What is the range? a. 27 b. 28 c. 42 d. 55

Page 23

Name:

Class:

Date:

Chap_04_5e 91. Professor Fisher calculates the grades on the first exam for her statistics class. She finds that students did either really well or really poorly. What kind of distribution does Prof. Fisher have? a. normal b. unimodal c. bimodal d. negatively skewed 92. Numbers that describe samples are called: a. statistics. b. parameters. c. averages. d. variables. 93. In a sample of data that has several extremely high scores, creating positive skew, the mean: a. will be less than the mode. b. will be greater than the median. c. and the mode will be equal. d. will be less than the median.

Page 24

Name:

Class:

Date:

Rating 4 2 2 1 3 1 3 4 1

94. (Table: Depression) The modal depression rating (or mode) in this data set is: a. 1.00. b. 1.15. c. 2.00. d. 2.33. 95. What is the formula for the interquartile range? a. Q3 – Q1 b. Q1 – Q3 c. Maximum – Minimum d. Minimum – Maximum 96. The median household income in Albemarle County, Virginia, for the year 2016 was $70,342. If the mean and mode are also computed, which statement could NOT be true? a. Half of the households had incomes below the median. b. The majority of households had incomes above the mean. c. The majority of households had incomes above the median. d. The majority of households had incomes above the mode.

Page 25

Name:

Class:

Date:

Chap_04_5e 97. A scientist analyzed the results of his data and calculated the standard deviation of scores at 6.4. What is the variance? a. 2.53 b. 6.70 c. 13.40 d. 40.96 98. The concept of variability is best described as: a. the spread of the distribution of data. b. the lack of consistency in data values. c. the average deviation of the data. d. how data tend to cluster around a center. 99. We _____ when calculating the average of deviations from the mean, because the sum of all deviation scores around the mean equals zero. a. take the absolute value of the scores b. square the scores c. divide by N d. take the square root 100. The median is the number that perfectly balances the: a. number of scores in the data set, such that half fall above the median and half fall below the median. b. deviations of the scores from the mean, such that the sum of the deviations for scores above the mean exactly equals the sum of the deviations for scores below the mean. c. two most extreme scores in the data set. d. even and the odd scores in the data set.

Page 26

Name:

Class:

Date:

Rating 1 7 6 7 7 6 5

101. (Table: Optimism) The median in this data set is: a. 3.96. b. 5.57. c. 6.00. d. 7.00. 102. (Table: Optimism) The standard deviation in this data set is: a. 1.99. b. 2.13. c. 3.96. d. 6.00. 103. A unimodal distribution has _____ mode(s). a. one b. two c. two or more d. more than two 104. A scientist analyzed the results of her data and calculated the variance of scores at 126.00. What is the standard deviation? a. 10.63 b. 11.22 c. 63.00 d. 15,876.00

Page 27

Name:

Class:

Date:

Chap_04_5e 105. The most frequently occurring score in a distribution is the: a. mean. b. median. c. mode. d. sample statistic. 106. Here is a set of data: 31, 54, 26, 45, 39, 45, 35, 42. What is the mode? a. 39.63 b. 40.50 c. 45.00 d. 54.00 This histogram represents the distribution of the number of years of education completed by twins who attended the 16th Annual Twins Day Festival in Twinsburg, Ohio, in August of 1991. Figure: Years of Education

107. (Figure: Years of Education) What is the mode of this distribution? a. 12 b. 14 c. 16 d. 20

Page 28

Name:

Class:

Date:

Chap_04_5e Enter the appropriate word(s) to complete the statement. 108. If the score at the 75th percentile is 20 and the score at the 25th percentile is 4, then the interquartile range is _______.

109. Standard deviation is computed by taking the square root of the _______.

110. The middle score of an ordered distribution is the _______.

111. The symbol for the sum of squares is _______.

112. Numbers based on a sample are called _______.

113. The _______ percentile marks the first quartile of a data set.

Page 29

Name:

Class:

Date:

Chap_04_5e 114. A distribution with one mode is called a(n) _______ distribution.

115. If an extreme score is added to a data set, increasing the range of data, the standard deviation will _______.

116. A distribution with more than two modes is called a(n) _______ distribution.

117. A distribution with two modes is called a(n) _______ distribution.

118. The most frequently occurring score in a distribution is the _______.

119. The arithmetic average of a set of data is the _______.

120. _______ is the most common measure of variability reported.

Page 30

Name:

Class:

Date:

Chap_04_5e 121. The _______ percentile marks the third quartile of a data set.

122. The second quartile, or the 50th percentile of a distribution, is also known as the _______.

123. The _______ percentile marks the median of a data set.

124. Standard deviation is computed by taking the _______ of the variance.

125. Numbers based on a _______ are called statistics.

126. By using the difference between the 25th and 75th percentiles, the _______ provides a measure of variability that is less affected by outliers.

127. Numbers based on a _______ are called parameters.

Page 31

Name:

Class:

Date:

Chap_04_5e 128. The measure of central tendency most likely to be distorted by outliers is the _______.

129. The _______ is the difference between the highest and lowest scores.

130. The variance calculated on 30 scores is equal to 27.56. The standard deviation is _______. (Answer to two decimal places.)

131. _______ is a measure that describes variability in squared units.

132. Numbers based on a population are called _______.

133. The sum of the deviation scores from the mean is equal to _______.

Page 32

Name:

Class:

Date:

Chap_04_5e This table depicts the cost of electricity in cents per kilowatt for the Northern Atlantic states during a single month of 2018. Table: Electricity Cost State Maine New Hampshire Vermont Massachusetts Rhode Island Connecticut New York New Jersey Pennsylvania

Cents per Kilowatt 16.16 19.63 18.64 21.11 18.64 21.62 19.30 15.64 13.58

134. (Table: Electricity Cost) Calculate the range, standard deviation, and variance for electricity costs in the South Atlantic States.

Page 33

Name:

Class:

Date:

Chap_04_5e The figures in the table are the salaries for each of the 30 Colorado Rockies baseball players during the 2005 baseball season. Numbers are in thousands of dollars. Table: Colorado Rockies Salaries 320 317 316 317 316 2350 317 326 319 317

328 324 650 317 12,600 366 2400 2200 6575 321

316 326 950 950 318 316 316 317 12,500 550

135. (Table: Colorado Rockies Salaries) Calculate the range and the interquartile range of the salaries for the Colorado Rockies players in the 2005 season. Also show all five values that would be included in a fivenumber summary of these salaries.

136. (Table: Colorado Rockies Salaries) Calculate the standard deviation of the salaries for the Colorado Rockies players in the 2005 season.

137. Eva is moving for her new job, which is in a large urban area. She is looking for an apartment and wants to get an idea of how much apartments cost. Which measure of central tendency would be of most use to Theresa in determining the typical cost of housing in this urban area? Explain why.

Page 34

Name:

Class:

Date:

Cents per Kilowatt 16.16 19.63 18.64 21.11 18.64 21.62 19.30 15.64 13.58

138. (Table: Electricity Cost) Calculate the mean, median, and mode for electricity costs in the Northern Atlantic States.

Page 35

Name:

Class:

Date:

Chap_04_5e Table: Frequency Table X (Score) 1 2 3 4 5 6 7

F 2 4 16 12 1 3 1

139. (Table: Frequency Table) (a) Use the information in the table to determine the mean, median, and mode for this set of scores. (b) If an extreme score, such as 25, is added to this data set, describe how the mode, median, and mean will be affected. No calculations are necessary to answer (b).

Page 36

Name:

Class:

Date:

Chap_04_5e This table depicts the annual salary for a sample of 10 full-time psychology faculty working at a large public university in 2018. Table: Psychology Faculty Salaries Faculty member 1 2 3 4 5 6 7 8 9 10

Salary ($US) 99,700 167,900 122,000 152,300 110,800 61,000 112,000 82,000 108,300 307,700

140. (Table: Psychology Faculty Salaries) Calculate the mean and median salary for the 10 psychology faculty members listed in the table.

141. (Table: Psychology Faculty Salaries) Calculate the range, standard deviation, and variance of the salaries of the 10 psychology faculty members listed in the table.

Page 37

Name:

Class:

Date:

Chap_04_5e 142. As quality care manager at a raisin manufacturing and packaging plant, you want to ensure that all the raisins you sell are of comparable quality–that they are similar in plumpness and weight. In the plant, raisins are poured into boxes until the box reaches its sale weight. To determine whether a similar number of raisins are going into each box, you sample 36 boxes and count the number of raisins in each box. The data for this is: 23, 24, 25, 25, 25, 27, 27, 27, 27, 27, 27, 27, 27, 28, 28, 29, 29, 29, 29, 29, 29, 30, 31, 32, 32, 33, 33, 33, 34, 34, 35, 35, 35, 36, 36, 38. To assess the variability in the number of raisins going into each box, calculate both the range and the interquartile range. Based on your calculations, do you find the variability in the number of raisins going into packages unacceptable? Describe the difference between these two measures of variability, especially in terms of their trustworthiness.

The figures in the table are the salaries for each of the 30 Colorado Rockies baseball players during the 2005 baseball season. Numbers are in thousands of dollars. Table: Colorado Rockies Salaries 320 317 316 317 316 2350 317 326 319 317

328 324 650 317 12,600 366 2400 2200 6575 321

316 326 950 950 318 316 316 317 12,500 550

143. (Table: Colorado Rockies Salaries) (a) Calculate the mean, median, and mode for the salaries of the Colorado Rockies players in the 2005 season. (b) Given your calculations, describe the impact of outliers, or skew within the distribution of salaries, on these measures of central tendency. Based on that consideration, which measure of center would you recommend using here?

Page 38

Name:

Class:

Date:

Salary ($US) 99,700 167,900 122,000 152,300 110,800 61,000 112,000 82,000 108,300 307,700

144. (Table: Psychology Faculty Salaries) If 307,700 is an outlier, show its impact on the mean and median by recalculating both of those measures of central tendency with and without the outlier. Explain the resulting changes you observe in those calculations.

Page 39

Name:

Class:

Date:

Chap_04_5e Answer Key 1. True 2. False 3. False 4. True 5. True 6. False 7. False 8. False 9. False 10. False 11. True 12. False 13. False 14. False 15. False 16. True 17. True 18. False 19. True 20. True 21. a 22. c 23. b 24. b 25. c 26. d Copyright Macmillan Learning. Powered by Cognero.

Page 40

Name:

Class:

Date:

Chap_04_5e 27. c 28. b 29. d 30. c 31. c 32. c 33. b 34. b 35. b 36. b 37. a 38. c 39. c 40. b 41. a 42. c 43. c 44. d 45. d 46. c 47. d 48. c 49. b 50. b 51. c 52. d 53. a 54. c Copyright Macmillan Learning. Powered by Cognero.

Page 41

Name:

Class:

Date:

Chap_04_5e 55. b 56. a 57. c 58. a 59. b 60. b 61. a 62. b 63. c 64. c 65. b 66. d 67. d 68. b 69. d 70. d 71. a 72. d 73. c 74. d 75. d 76. b 77. a 78. d 79. b 80. b 81. b 82. d Copyright Macmillan Learning. Powered by Cognero.

Page 42

Name:

Class:

Date:

Chap_04_5e 83. c 84. b 85. d 86. a 87. b 88. a 89. c 90. b 91. c 92. a 93. b 94. a 95. a 96. c 97. d 98. a 99. b 100. a 101. c 102. a 103. a 104. b 105. c 106. c 107. a 108. 16 109. variance 110. median Copyright Macmillan Learning. Powered by Cognero.

Page 43

Name:

Class:

Date:

Chap_04_5e 111. SS 112. statistics 113. 25th 114. unimodal 115. increase 116. multimodal 117. bimodal 118. mode 119. mean 120. Standard deviation 121. 75th 122. median 123. 50th 124. square root 125. sample 126. interquartile range 127. population 128. mean 129. range 130. 5.25 131. Variance 132. parameters 133. zero 134. Range = 8.04; SD = 2.49; SD2 = 6.20 135. Range = 12,284; interquartile range = 633; minimum = 316; Q1 = 317; median = 322.5; Q3 = 950; maximum = 12,600 136. SD = 3171.05

Page 44

Name:

Class:

Date:

Chap_04_5e 137. Eva would probably want to know the median cost of an apartment. In a large urban area there are likely to be both very wealthy people and very poor people. As such, the distribution might be skewed and the mean would not be a good representation of the typical apartment rent. 138. Mean = 18.26; median = 18.64; mode = 18.64 139. (a) Mean = 3.49; median = 3; mode = 3. (b) The mode would stay the same because the score 3 still occurs the most at a frequency of 16. The median would not change since the total number of scores will only increase from 39 to 40, which would still make the median be 3. The mean, which takes into account the actual values of all data points, would be pulled toward that outlier, so we can expect the value of the mean to change the most. The mean would increase since the extreme score is larger than the range of scores already used to calculate the mean in the data 140. M = 132,370; median = 111,400 141. Range = 246,700, SD = 65,319, SD2 = 4,266,544,100 142. The range is 15. The interquartile range is 6. Students' answers regarding whether they think this is unacceptable variability will vary. The range relies on the minimum and maximum scores, or numbers of raisins in a box, and is inherently vulnerable to the influence of outliers. The interquartile range provides a measure of variability that is insulated against outliers because it is based on the 25th and 75th percentiles within the data set. 143. (a) Mean = 1605.17; median = 322.5; mode = 317. (b) The mean is considerably larger than the median or mode, indicating that high scores within the data set pulled the mean up. Because there appear to be high outliers among these salaries, the median would be a better estimate of average player salary. 144. With the outlier: M = 132,370, median = 111,400; without the outlier: M = 112,889; median = 110,800. The removal of one high score, 307,700, results in both the mean and the median shifting down. The median moves from the average of the 5th and 6th score to the 5th, or from 111,400 to 110,800. The mean changes from 132,370 to 112,889 a more substantial change than that seen in the median. Notice that with the removal of the outlier, the mean becomes similar in value to the median.

Page 45

Name:

Class:

Date:

Chap_05_5e Indicate whether the statement is true or false. 1. Volunteer sampling is a form of random sampling. a. True b. False 2. In a research study, the control group is where experimenters exert more of their efforts, including the experimental manipulation. a. True b. False 3. Studies in which the researcher failed to reject the null hypothesis are less likely to be published than studies that rejected the null hypothesis. a. True b. False 4. Random samples are used more often in research than convenience samples. a. True b. False 5. Confirmation bias is the tendency to pay attention to evidence that goes against one's beliefs. a. True b. False 6. When one rejects the null hypothesis, it means that the research hypothesis is proven. a. True b. False 7. Volunteer sampling is a form of convenience sampling. a. True b. False 8. The null hypothesis can express that there is no relationship between variables or that no change is expected. a. True b. False 9. A false-negative pregnancy test is an example of a Type I error. a. True b. False

Page 1

Name:

Class:

Date:

Chap_05_5e 10. Even though it is not always possible to employ random selection, random assignment should be used whenever possible. a. True b. False 11. Probability is calculated by dividing the number of successes by the number of trials. a. True b. False 12. A false-negative pregnancy test is an example of a Type II error. a. True b. False 13. Biased samples can be very effective in providing desired results for a researcher. a. True b. False 14. Susan flips a coin 20 times and observes 8 heads. Those heads are the outcomes of his study. a. True b. False 15. Susan flips a coin 20 times and observes 8 heads. Those heads are the successes of his study. a. True b. False 16. External validity is strengthened by the use of volunteer samples. a. True b. False 17. A false-positive pregnancy test is an example of a Type II error. a. True b. False 18. A researcher who fails to reject the null hypothesis when in fact it should be rejected commits a Type II error. a. True b. False 19. Probability is calculated by dividing the number of trials by the number of successes. a. True b. False

Page 2

Name:

Class:

Date:

Chap_05_5e 20. A 5-year-old believes that the squeakiness of his shoes on the floor is related to his ability to play basketball. It is likely that there is no real association between the two and this is an example of an illusory correlation. a. True b. False 21. Convenience samples are used more often in research than random samples. a. True b. False 22. Confirmation bias is the tendency to ignore evidence that goes against one's beliefs. a. True b. False 23. Studies in which the researcher failed to reject the null hypothesis are more likely to be published than studies that rejected the null hypothesis. a. True b. False 24. A false-positive pregnancy test is an example of a Type I error. a. True b. False 25. Failing to reject the null hypothesis is the same as accepting the null hypothesis. a. True b. False 26. A researcher who fails to reject the null hypothesis when in fact it should be rejected commits a Type I error. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 27. Esther hypothesized that older adults would score higher on emotional intelligence measures compared to younger adults in her study. If the results are in support of Esther's hypothesis, she would: a. commit Type II error. b. reject the null hypothesis. c. fail to reject the null hypothesis. d. commit a Type I error.

Page 3

Name:

Class:

Date:

Chap_05_5e 28. A trial refers to each time a procedure is: a. a success. b. a failure. c. carried out. d. expected to succeed. 29. Random selection is _____ used, but random assignment is _____ used. a. always; rarely b. frequently; rarely c. rarely; frequently d. rarely; always 30. Kendall Jenner, the famous television personality and model who has served as a spokesman for the antiacne skin care product Proactiv, is an example of _____. While her claims are very impressive and persuasive, a well-designed study would provide more convincing evidence in support of this acne cure. a. random selection b. replication c. a testimonial d. personal probability 31. Why are true random samples rarely used? a. Researchers rarely have access to the entire population. b. Convenience sampling is easier and just as effective in producing a sample that can be generalized to the population of interest. c. Random sampling methods are not effective in producing a sample that can generalized to the population of interest. d. Researchers, like most people, have misconceptions of randomness. 32. Dr. Dutton designed an experimental study to assess potential differences between science students and art students on a math reasoning abilities test. Dr. Dutton found a mean difference in math performance between science and art students. On average, science students performed higher on the math reasoning test compared to the art students. Dr. Dutton's findings support which hypothesis? a. research b. null c. statistical d. experimental

Page 4

Name:

Class:

Date:

Chap_05_5e 33. In a study examining the effects of humor on memory, Schmidt (1994) had participants read either humorous sentences or non-humorous sentences and then later tested participants' memory for the sentences. Identify the null hypothesis this study. a. There is no difference in memory for humorous and non-humorous sentences. b. Humorous sentences are better remembered than non-humorous sentences. c. Participants remember, on average, 3.5 sentences. d. Participants have poor memory for all of the sentences. 34. Lucas tosses a quarter 4 times and 3 times it comes up heads. The probability of heads is: a. 0.75. b. 1.00. c. 0.25. d. 0.50. 35. A statement that postulates that there is no difference between populations is a: a. negative statement. b. null hypothesis. c. research hypothesis. d. positive statement. 36. The text claims that journals tend to prefer "exciting" findings over "boring" ones. Which of these constitutes an "exciting" finding? a. failing to reject the null hypothesis b. committing a Type I error c. It is not possible to determine based on the information provided. d. rejecting the null hypothesis 37. As a woman, Gretchen has noticed that males always use the middle armrests on airplanes. While those armrests are shared by two seats, she never seems to get to use them if she is in a center seat between two males. Gretchen can remember several cases where this has happened to her. Her selective memory for these cases is an example of: a. representativeness. b. confirmation bias. c. an illusory correlation. d. personal probability.

Page 5

Name:

Class:

Date:

Chap_05_5e 38. Successful replication of research builds a case for the generalizability of findings. In order for replications to build that strong case, it is important that they occur in: a. similar situations with similar participants so as to encourage the same findings. b. inconsistent contexts using different measures from the original study. c. a new context or with samples that have different characteristics. d. designs with experimental and control groups that used random sampling. 39. A social psychologist is interested in the eventual divorce rates of people who live together before they get married compared to those who do not cohabitate before marriage. Any difference found between these groups would be of interest to the researcher, as well as to the popular media. Which statement is an appropriate research hypothesis for this research? a. There is no difference in divorce rates between those who cohabitate and those who do not. b. There is a difference in divorce rates between those who cohabitate and those who do not. c. People who cohabitate before marriage have a divorce rate twice as high as those who do not. d. People who do not cohabitate before marriage have a divorce rate twice as high as those who cohabitate. 40. When his manager asked him the likelihood that he would be able to produce the company's annual report by next Friday, Skylar told him that he was 90 percent likely to complete it by the deadline. Skylar's estimate is: a. a personal probability. b. the expected relative-frequency probability. c. an expected outcome. d. generalizable. 41. In a series of clinical trials, a drug company wishes to assess the effectiveness of its newly developed medication for social anxiety by comparing its performance to the leading drug currently on the market and to a placebo control. What research technique or techniques would be possible and appropriate? a. random assignment but not random selection b. random selection but not random assignment c. both random selection and random assignment d. neither random selection nor random assignment

Page 6

Name:

Class:

Date:

Chap_05_5e 42. Random assignment differs from random selection because: a. random assignment deals with how participants for a study are chosen to begin with, whereas random selection deals with how participants in a sample are assigned to levels of the independent variable. b. random assignment deals with how participants in a sample are assigned to levels of the independent variable, whereas random selection deals with how participants for a study are chosen to begin with. c. random assignment is rarely achieved, but random selection is almost always achieved. d. random assignment must be performed by the experimenter who is carrying out the study, but random selection can be carried out at the analysis stage. 43. What is the correct formula for calculating a probability? a. divide number of trials by number of errors b. divide number of successes by number of trials and multiply by one hundred c. divide number of trials by number of errors and multiply by one hundred d. divide number of successes by number of trials 44. Kendra tosses a quarter 10 times and finds that 4 of the 10 times the quarter comes up heads and 6 of the 10 times the quarter comes up tails. What should Kendra conclude? a. The quarter is an unfair quarter. b. The expected relative frequency probability of getting heads when tossing a quarter is 4/10. c. It is more probable to get heads than it is to get tails when tossing a quarter. d. In the long-term the expected relative frequency would be 5/5, but with only 10 trials a different pattern has emerged. 45. Lucas tosses a quarter 4 times and 3 times it comes up heads. The proportion of heads is: a. 0.75. b. 1.00. c. 0.50. d. 0.25. 46. Percentage is simply the _____ multiplied by 100. a. proportion b. number of successes c. number of failures d. number of trials

Page 7

Name:

Class:

Date:

Chap_05_5e 47. Anthony once heard that wearing colorful ties helped release your positive energy toward others. Being the scientist that he is, he decided to put the claim to the test by running a well-designed study. He had 18 people wear colorful ties and another 18 people wear black ties. After carefully designing, collecting, and analyzing his data, he found no differences between the groups that he studied. What conclusion should he make? a. Fail to reject the null hypothesis and conclude that colorful ties do not help release your positive energy toward others. b. Fail to reject the null hypothesis and conclude that the effect of colorful ties is not measurable. c. Fail to reject the null hypothesis and conclude that, based on this study, he did not observe an effect for colorful ties. d. Reject the null hypothesis and conclude that colorful ties may benefit positive energy. 48. All of the recent children born into Hannah's family have been males, so there is a lot of pressure for her and her husband Sean to have the first granddaughter. In fact, everyone is sure that their first child will be a girl, because the family is due to have one! Unfortunately for their families, this logic is false because: a. the birth of a girl occurs only under the law of large numbers. b. the sex of their child is independent of that of any previous children born. c. the probability of a girl is less than that of a boy being born. d. an illusory correlation exists between gender and family expectations. 49. A social psychologist is interested in the eventual divorce rates of people who live together before they get married compared to those who do not cohabitate before marriage. Any difference found between these groups would be of interest to the researcher, as well as to the popular media. Which statement is an appropriate null hypothesis for this research? a. There is no difference in divorce rates between those who cohabitate and those who do not. b. There is a difference in divorce rates between those who cohabitate and those who do not. c. People who cohabitate before marriage have a divorce rate twice as high as those who do not. d. People who do not cohabitate before marriage have a divorce rate twice as high as those who cohabitate. 50. Whereas _____ are not good at generating random numbers because of consideration of previous events, _____ have no memory for previous events and can therefore be unbiased. a. researchers; students b. computers; random numbers tables c. humans; computers d. students; researchers

Page 8

Name:

Class:

Date:

Chap_05_5e 51. A(n) _____ refers to the outcome for which one is trying to determine the probability. a. success b. failure c. trial d. outcome 52. Esther hypothesized that older adults would score higher on emotional intelligence measures compared to younger adults in her study. If Esther erroneously rejected the null hypothesis, she would be committing what type of error? a. Type I error b. Type II error c. statistical error d. sampling bias 53. A social psychologist is interested in the eventual divorce rates of people who live together before they get married compared to those who do not cohabitate before marriage. The researcher is particularly interested in the couples who cohabitate to see if that leads to higher divorce rates. In a sense, the couples who cohabitate are serving as the _____ group. a. independent b. null c. experimental d. control 54. Rita is a nurse in the obstetrics ward of a hospital. She tells her friend that at her hospital more babies are born when there is a full moon. The hospital's records show no such relationship, however. Rita's belief is an example of: a. generalizability b. an illusory correlation. c. a Type I error. d. a Type II error. 55. True _____ is difficult and sometimes impossible to attain. a. random assignment b. cluster sampling c. convenience sampling d. random sampling

Page 9

Name:

Class:

Date:

Chap_05_5e 56. A statement that postulates that two populations are different from one another is a(n) _____ hypothesis. a. null b. research c. illusory d. control 57. A(n) _____ refers to each time a procedure is carried out. a. success b. failure c. trial d. outcome 58. In the absence of many trials, one cannot determine true probabilities of events. However, over the long run, and numerous trials, the expected relative-frequency probability of events is very clear and predictable. This is known as the: a. independence of trials. b. law of large numbers. c. long-run probability calculation. d. objective probability. 59. A statement that postulates that two populations are NOT different from one another is a(n) _____ hypothesis. a. null b. research c. illusory d. control 60. As part of their work in a psychology research methods class, a group of psychology students devised a survey to assess the relation between stress and health. Each member of the class administered the survey to 12 friends, and the data were then pooled. What method of sampling was used? a. random b. convenience c. representative d. population 61. The expected outcome if an experiment is repeated many, many times is the: a. underlying probability. b. reliable outcome. c. expected relative-frequency probability. d. expected outcome. Copyright Macmillan Learning. Powered by Cognero.

Page 10

Name:

Class:

Date:

Chap_05_5e 62. In a study examining the effects of humor on memory, Schmidt (1994) had participants read either humorous sentences or non-humorous sentences and later tested participants' memory for the sentences. Identify the research hypothesis for this study. a. There is no difference in memory for humorous and non-humorous sentences. b. Humorous sentences are better remembered than non-humorous sentences. c. Participants remember, on average, 3.5 sentences. d. Participants have poor memory for all of the sentences. 63. Which of these are independent events? a. chance of winning three hands of poker in a row b. drawing two cards from a deck of cards without placing either back in the deck c. opinions of two friends on the latest summer blockbuster d. sexual satisfaction of spouses who are married to each other 64. A behavioral researcher interested in the concept of preparedness sets up a booth at a local mall. Her idea is to compare men and women in terms of what they carry on their person, and to evaluate what types of events or issues they are prepared to handle based on what they are carrying with them. She hangs a sign on her booth that reads, "Research study underway; stop here to participate." Everyone who willingly participates in her study is part of a: a. convenience sample. b. random sample. c. volunteer sample. d. random assignment. 65. A sample in which every member of the population has an equal chance of being selected for inclusion in the study is a _____ sample. a. convenience b. representative c. volunteer d. random 66. The statement "There's a 75 percent chance that I will pass my history class" best illustrates which concept? a. subjective probability b. expected-relative frequency probability c. hypothesis testing d. Type I error

Page 11

Name:

Class:

Date:

Chap_05_5e 67. The text claims that journals tend to prefer "exciting" findings over "boring" ones. Which of these constitutes a "boring" finding? a. failing to reject the null hypothesis b. committing a Type I error c. It is not possible to determine based on the information provided. d. rejecting the null hypothesis 68. Random sampling is the method most likely to lead to a _____ sample. a. biased b. convenience c. self-selected d. representative 69. Dr. Becker designed an experimental study to assess potential differences between science students and art students on a math reasoning abilities test. Dr. Becker found a mean difference in math performance between science and art students. On average, art students performed higher on the math reasoning test compared to the science students. Dr. Becker's results were supported in two other studies performed in university settings. Dr. Becker concluded that, given that the results of her study have been replicated at other universities, her findings were reliable. Dr. Becker also hopes that other researchers will perform the study using other samples including participants from the community. Dr. Becker is concerned about replicating her study across various settings and using additional samples because she wishes to: a. improve the generalizability of her results. b. support her null hypothesis. c. improve her descriptive statistics. d. calculate a probability. 70. On April 16, 2007, BBC News reported the results of a study done by Dr. David Lewis of Mindlab International in the United Kingdom. Dr. Lewis found that eating dark chocolate had longer lasting excitatory effects on the body than did kissing a romantic partner. From his statement, it is evident that Dr. Lewis: a. rejected the null hypothesis. b. committed a Type I error. c. failed to reject the null hypothesis. d. committed a Type II error.

Page 12

Name:

Class:

Date:

Chap_05_5e 71. In a demonstration of a classic study on the effects of smiling on mood, an instructor had students sitting on the left side of the classroom read a series of cartoons while frowning and had students sitting on the right side of the classroom read the same series of cartoons while smiling. In this example, the instructor failed to: a. obtain a small sample size. b. randomly assign participants to conditions of the study. c. recruit a convenience sample. d. calculate the subjective probability of group membership. 72. Students who post responses on ratemyprofessor.com are what type of sample? a. cluster b. random c. convenience d. volunteer 73. Confirmation bias suggests that people will be most cynical about: a. studies containing Type I errors. b. studies containing Type II errors. c. research findings that they do not like. d. research findings that they like. 74. _____ sampling or _____ sampling is a type of convenience sampling. a. Random; volunteer b. Random; self-selective c. Volunteer; self-selective d. Volunteer; research 75. Which of these is an example of randomly assigning participants to conditions in a study? a. Micayla places the first six people to show up for her study in the experimental group and the next six people in the control group. b. Every time a participant shows up for his study, Samuel flips a coin to determine which condition to put the participant in. He predetermined that heads is the control group and tails is the experimental group. c. Felicity has identified her population of interest as all students attending the local county high school. She then proceeds to randomly choose high school students for her study by having a random number generator generate possible high school ID numbers. d. Preet gives his problem-solving task to a group of first graders in the classroom for which he is a student teacher.

Page 13

Name:

Class:

Date:

Chap_05_5e 76. Volunteer samples are: a. preferred to random samples. b. not preferred to random samples. c. provide similar results to random samples. d. allow for greater generalizability than random samples. 77. Makenzie wishes to study the effects of a mother's cocaine use during pregnancy on cognitive indicators during the first month of her infant's life. What research technique or techniques would be possible and appropriate? a. random assignment but not random selection b. random selection but not random assignment c. both random selection and random assignment d. neither random selection nor random assignment 78. A psychologist who fails to reject the null hypothesis when the null hypothesis is in fact false has: a. made a Type II error. b. made a Type I error. c. made an illusory correlation. d. fallen prey to the confirmation bias. 79. A(n) _____ is a level of the _____ that receives a manipulation. a. experimental group; independent variable b. independent variable; experimental group c. control group; independent variable d. independent variable; control group 80. If Dr. Baiu uses random assignment in her research, then she can make the reasonable assumption that: a. her experimental and control groups have similar characteristics prior to receiving the experimental treatment. b. any observed differences between her experimental and control groups are statistically significant. c. her findings can be generalized to populations she has not yet studied. d. the participants in her study are extremely similar to the people in the larger populations from which they were selected. 81. A(n) _____ refers to the result of a(n) _____. a. success; failure b. failure; trial c. trial; outcome d. outcome; trial

Page 14

Name:

Class:

Date:

Chap_05_5e 82. If a psychologist rejects the null hypothesis when the null hypothesis is in fact true, the psychologist has: a. made a Type II error. b. made a Type I error. c. made an illusory correlation. d. fallen prey to the confirmation bias. 83. The saying "a watched pot never boils" claims an association between the temperature at which water boils and spectator presence. This is an example of a(n): a. false assumption. b. expected relative-outcome probability. c. confirmation bias. d. illusory correlation. 84. The text claims that journals tend to prefer "exciting" findings over "boring" ones. Because of the drive to publish new findings that go against the status quo, the literature is assumed to contain _____ errors. a. Type I b. Type II c. null d. confirmatory 85. In a standard deck of playing cards, there is a total of 52 cards, 12 of which are face cards, such as queens, kings, and jacks. What is the formula for the expected relative-frequency probability of drawing a single face card from the deck of cards? a. 3/52 b. 12/52 c. (12 – 3)/52 d. 52/12 86. Two events are said to be independent events when: a. the probability of one event is influenced by the occurrence of the other event. b. you do not have knowledge about the occurrence of either event. c. the probability of one event is determined by the occurrence of the other event. d. the probability of one event is not influenced by the occurrence of the other event. 87. A placebo group is an example of a type of a(n) _____ group. a. experimental b. research c. control d. dependent

Page 15

Name:

Class:

Date:

Chap_05_5e 88. _____ sampling is the method most likely to lead to a representative sample. a. Random b. Convenience c. Self-selected d. Volunteer 89. A cognitive researcher has conducted a study on logical reasoning using a sample of college freshmen and is concerned that the results of the study may not accurately depict the logical reasoning of college seniors. The researcher's concern can be described as a concern about the _____ of the study results. a. replication b. accuracy c. generalizability d. design 90. A student researcher is interested in comparing reaction time differences between men and women. He obtains a sample of 46 students from a psychology student research database. Participants take part in the study, the data are later analyzed, and the researcher concludes that women have faster reaction times compared to men. What type of sampling strategy was used in this study? a. true b. population c. convenience d. random 91. On April 16, 2007, the BBC News reported the results of a study done by Dr. David Lewis of Mindlab International in the United Kingdom. Dr. Lewis found that eating dark chocolate had longer lasting excitatory effects on the body than did kissing a romantic partner. This is the first study to show such large effects of chocolate. If this study cannot be replicated, it implies that the researchers made a _____ error. a. Type II b. research design c. statistical analysis d. Type I 92. A psychological researcher's decision regarding whether to reject the null hypothesis is based on: a. the researcher's a priori theory regarding expected group differences. b. independent confirmation by researchers in other laboratories. c. the probability that group differences would be observed if there was no effect of the independent variable. d. whether the results of the study have been replicated at least once.

Page 16

Name:

Class:

Date:

Chap_05_5e 93. Milo insists that women are poorer drivers than men. To back up his claim he describes two incidents: one in which his girlfriend sideswiped a car and a second in which his mother failed to react in time to avoid hitting a squirrel on the highway. Milo's attention to this positive evidence for his belief reflects: a. generalizability. b. an illusory correlation. c. a confirmation bias. d. personal probability. 94. Esther hypothesized that older adults would score higher on emotional intelligence measures compared to younger adults in her study. If Esther commits a false-negative error, then she has committed a: a. Type I error. b. Type II error. c. statistical error. d. sampling bias. 95. When a researcher considers whether the results of a laboratory study will actually hold up as true when applied to the world outside of the laboratory, the researcher is considering the _____ of the findings. a. globalization b. external validity c. reliability d. expected relative-frequency probability 96. A student researcher is interested in comparing reaction time differences between men and women. He obtains a sample of 46 students from a psychology student research database. Participants take part in the study, the data are later analyzed, and the researcher concludes that women have faster reaction times compared to men. What is a potential limitation of the study's conclusion? a. generalizability b. plagiarism c. random sampling d. random selection

Page 17

Name:

Class:

Date:

Chap_05_5e 97. Anthony once heard that wearing colorful ties helped release your positive energy toward others. Being the scientist that he is, he decided to put the claim to the test by running a well-designed study. He had 18 people wear colorful ties and another 18 people wear black ties. After carefully designing, collecting, and analyzing his data, he found no differences between the groups that he studied. Which of these represents failing to reject the null hypothesis, something he should NOT do? a. Fail to reject the null hypothesis and conclude that colorful ties do not help release your positive energy toward others. b. Fail to reject the null hypothesis and conclude that the effect of colorful ties is not measurable. c. Fail to reject the null hypothesis and conclude that, based on this study, he did not observe an effect for colorful ties. d. Reject the null hypothesis and conclude that colorful ties may benefit positive energy. 98. _____ refers to the occurrence of events over the long run, and _____ refers to the calculation of the number of successes divided by the number of trials. a. Proportion; percentage b. Percentage; proportion c. Probability; percentage d. Probability; proportion 99. The extent to which research findings from one sample or context can be applied to other samples or contexts is called: a. replication. b. generalizability. c. sampling. d. probability. 100. When the outcome of one trial does not depend, in any way, on the outcome of previous trials, the events are said to be _____. a. dependent b. independent c. confirmatory d. confounding 101. A random-digit generator is MOST likely to be used for which type of strategy? a. convenience sampling b. volunteer sampling c. random assignment d. hypothesis testing

Page 18

Name:

Class:

Date:

Chap_05_5e 102. A tendency to pay attention to evidence that confirms one's a priori belief is called: a. an illusory correlation. b. personal probability. c. a Type I error. d. confirmation bias. 103. A social psychologist is interested in the eventual divorce rates of people who live together before they get married compared to those who do not cohabitate before marriage. The researcher is particularly interested in the couples who cohabitate to see if that leads to higher divorce rates. In a sense, the couples who do not cohabitate are serving as the _____ group. a. independent b. null c. experimental d. control 104. A duplication of scientific results in a different context or with a different sample is: a. an illusory correlation. b. plagiarism. c. replication. d. generalizability. 105. A behavioral researcher interested in the concept of preparedness sets up a booth at a local mall. Her idea is to compare men and women in terms of what they carry on their person, and to evaluate what types of events or issues they are prepared to handle based on what they are carrying with them. She has chosen a mall setting because people are readily available. In this sense, people at the mall are a: a. convenience sample. b. random sample. c. selected sample. d. random assignment. 106. Lucas tosses a quarter 4 times and 3 times it comes up heads. The percentage of heads is: a. 0.75%. b. 7.5%. c. 75%. d. 50%.

Page 19

Name:

Class:

Date:

Chap_05_5e 107. In a study examining the effects of humor on memory, Schmidt (1994) had participants read either humorous sentences or non-humorous sentences and then later tested participants' memory for the sentences. The experimental group in this experiment was: a. the group of participants assigned non-humorous sentences. b. the group of participants assigned humorous sentences. c. a third group of participants who were not asked to read any sentences. d. irrelevant because there was no experimental group in this experiment. 108. A(n) _____ is a level of the _____ that does not receive a manipulation. a. experimental group; independent variable b. independent variable; experimental group c. control group; independent variable d. independent variable; control group 109. A random numbers table or online generator is often used by researchers to create random selection or assignment because: a. use of tables and computers increases the quality of the scientific investigation. b. they guarantee equal assignment of participants across levels of the independent variable. c. in general, people are poor at judging randomness and therefore must use a machine to generate random sequences. d. their use ensures less bias in the research observations. 110. External validity is _____ by the use of volunteer samples. a. weakened b. strengthened c. divided in half d. doubled 111. A Type I error is to a _____ as a Type II error is to a _____. a. true positive; true negative b. false positive; false negative c. false negative; false positive d. true negative; false positive 112. A _____ sample is a type of convenience sample. a. random b. volunteer c. replication d. population

Page 20

Name:

Class:

Date:

Chap_05_5e 113. In a study examining the effects of humor on memory, Schmidt (1994) had participants read either humorous sentences or non-humorous sentences, and then he later tested participants' memory for the sentences. The control group in this experiment was: a. the group of participants assigned non-humorous sentences. b. the group of participants assigned humorous sentences. c. a third group of participants who were not asked to read any sentences. d. irrelevant because there was no control group in this experiment. 114. A fellow classmate failed to replicate the bystander effect, a well-established effect in social psychology. Because this student failed to find a statistically significant difference in helping behaviors as the number of bystanders increased, it is possible that the student: a. made a Type II error. b. made a Type I error. c. committed a false positive. d. committed the confounding variable error. Enter the appropriate word(s) to complete the statement. 115. When a researcher rejects the _______ hypothesis but it is in fact true, the researcher has made a Type I error.

116. When the outcome of one trial does not depend, in any way, on the outcome of previous trials, the events are said to be _______.

117. The _______ group is the group receiving the intervention or treatment of interest.

Page 21

Name:

Class:

Date:

Chap_05_5e 118. A statement that two populations are different from one another is a(n) _______ hypothesis.

119. Brennan tosses a quarter 4 times and 1 time it comes up heads. The _______ of heads is 25.

120. When a researcher rejects the null hypothesis but the null hypothesis is in fact true, the researcher has made a(n) _______ error.

121. The text claims that journals tend to prefer "exciting" findings over "boring" ones. Because of the drive to publish new findings that go against the status quo, the literature is assumed to contain _______ errors.

122. The ability to apply research findings to contexts or samples other than the one studied is called _______.

123. Brennan tosses a quarter 4 times and 1 time it comes up heads. The _______ of heads is 0.50.

124. A statement that two populations are not different from one another is a(n) _______ hypothesis.

Page 22

Name:

Class:

Date:

Chap_05_5e 125. A type of sample in which participants are selected because they are readily available is a(n) _______ sample.

126. The _______ group is the group that does not receive the intervention or treatment of interest.

127. There is a bias toward the publication of studies in which the _______ hypothesis has been rejected.

128. The duplication of the results of a research study in a different context or with a different sample is called _______.

129. Your own estimate of the likelihood that you will uphold your New Year's resolution is known as a _______.

130. The _______ bias is the tendency to pay attention to evidence that supports one's beliefs and ignores that which does not.

Page 23

Name:

Class:

Date:

Chap_05_5e 131. When a sample has similar characteristics to the population from which it was drawn, the sample is _______.

132. A _______ sample occurs when everyone in the population has the same chance of being selected.

133. Seeing an association between two events that are unrelated is a(n) _______.

134. A type of sample in which participants actively elect to participate is a(n) _______, or self-selected, sample.

135. If a researcher finds that the groups studied differed from each other more than would be expected by chance alone, the researcher _______ the null hypothesis.

136. When a researcher fails to reject the null hypothesis but the null hypothesis is false, the researcher has made a(n) _______ error.

Page 24

Name:

Class:

Date:

Chap_05_5e 137. Expected relative-frequency probability is computed as the number of ____ divided by the number of _____.

138. If a researcher finds that the groups studied did not differ from each other more than would be expected by chance alone, the researcher _______ the null hypothesis.

139. Brennan tosses a quarter 4 times and 1 time it comes up heads. The _______ of heads is 0.25.

140. The following testimonial was once posted to the website for Mother Nature Diapers. The testimonial claims that the diapers are the best daytime diaper around. Identify any sampling problems and statistical logic problems with drawing the conclusion that Mother Nature Diapers are the best daytime diaper. ìAfter trying unsuccessfully to use cloth diapers, I gave Mother Nature's a try. Unlike most 'natural' diapers, these are thin instead of chunky, and baby can move around more easily. They truly are the best daytime diaper around.î ~M.G., Greenville, CA

141. A clinical psychologist wishes to study depression levels in elderly adults living in assisted care facilities. (a) What is the population of interest? (b) Is random selection possible in this case? Explain why or why not. (c) Is random assignment possible in this case? Explain why or why not. (d) Identify a sampling strategy that the clinical psychologist might use and explain how she could employ the strategy to answer her question.

Page 25

Name:

Class:

Date:

Chap_05_5e 142. At glennharrold.com one can purchase a hypnosis CD to listen to help one lose weight. The following is one happy customer's testimonial. Identify any sampling problems and statistical logic problems with drawing the conclusion that this CD will help most people lose weight. " ASTONISHING! I bought this with some reservations, as I had tried a similar one from a well-known hypnotist with limited success. However, I noticed some effects immediately, and was encouraged to carry on. It is now 2 weeks later, and I am hooked! My self-image has improved, which is a vital part of the positive attitude needed to complete to a goal weight, especially when you have a lot to lose. Also, I am finding it easy to 'keep control of my eating habits', and I no longer think of food all day! Believe me, this is some achievement, having been a 'comfort eater' for a very long time! I cannot recommend this highly enough. It is EXTREMELY powerful, and used correctly, will change your life forever." ~ Bazz

143. A local United Way and a group of community agencies in a large urban area have decided that making sure young children are ready to enter the school system is a high priority for the community. They hire a group of researchers to assess the school readiness of individual children entering the 100 kindergarten classrooms in the area. (a) What is the population of interest? (b) Why would the researchers choose to assess a sample rather than the entire population? (c) Is random selection possible in this case? Explain why or why not.

144. Researchers were interested in whether the use of social media depended upon a person's gender. The researchers divided their sample of college freshman into two groups: men and women. They then gave both groups of students a survey to measure their usage of social media. (a) What is the likely null hypothesis for this study? (b) What is the likely research hypothesis for this study?

Page 26

Name:

Class:

Date:

Chap_05_5e 145. In a comprehensive review of studies on touch therapy and the development of preterm infants, Caulfield (2000) reported that most studies have found that the use of touch therapy improves the weight gain of preterm infants relative to control groups. If an attempt to replicate these studies fails to find an effect of touch therapy on the weight gain of preterm infants, what is the error that the researchers performing the replication may have made? Explain the risks associated with making such an error, especially when made in an area without such a strong history of consistent findings.

146. A group of 30 students enrolled in a research methods class is studying the effect of clear plastic report covers on the grade they receive on a paper. They suspect that papers with a nifty plastic cover will look more professional, which will earn them a higher grade regardless of paper content. (a) What is the likely null hypothesis for any experiment the students would conduct? (b) What is the likely research hypothesis for any experiment the students would conduct? (c) What are likely control and experimental groups for this study?

147. A developmental researcher is interested in whether touch therapy improves the weight gain of preterm infants. (a) What is the population of interest? (b) Is random selection possible in this case? Explain why or why not. (c) Is random assignment possible in this case? If so, explain how random assignment might be executed.

148. An educational researcher is interested in the incidence of cheating among college freshmen. (a) What is the population of interest? (b) Is random selection possible in this case? Explain why or why not. (c) Is random assignment possible in this case? Explain why or why not. (d) Identify a sampling strategy that this researcher might use and explain how he could employ that sampling strategy to answer his question.

Page 27

Name:

Class:

Date:

Chap_05_5e 149. Researchers were interested in whether touch therapy improves the weight gain of preterm infants and compared the weight gain over a 3-week period of infants receiving touch therapy to the weight gain of infants not receiving touch therapy. (a) What is the likely null hypothesis for this experiment? (b) What is the likely research hypothesis for this experiment?

150. A group of medical researchers is interested in identifying the health care needs of individuals in prisons in the United States. The researchers plan to distribute a health care survey to a sample of prisoners. (a) What is the population of interest? (b) Why would the researchers choose to distribute the health care survey to a sample rather than to the entire population? (c) Is random selection possible in this case? Explain why or why not.

151. What is the difference between a Type I and a Type II error? Which kind of error is more likely to be published and why? Which kind of error is more dangerous?

152. In the 1960s, James McConnell published an article entitled "Memory Transfer Through Cannibalism in Planarium," in which he documented the results of a study in which flatworms were trained to navigate a maze and then were diced up and fed to a second generation of flatworms. These cannibal flatworms learned to navigate the maze at a faster rate than the original group of naïve flatworms. McConnell, as indicated by the title of his article, made the claim that memory for the maze was gained by ingesting the flatworms that knew how to navigate the maze. Subsequent studies have failed to replicate McConnell's results. It is therefore likely that McConnell made what kind of error?

153. What is meant by a representative sample? Why is it important to obtain a representative sample?

Page 28

Name:

Class:

Date:

Chap_05_5e 154. In the first decade of the 2000s, several self-help gurus made claims about the power of positive thinking. In fact, they might argue that if you think you are going to do really well on this test, then that positive energy you put out will result in good things coming to you; believe it and it will come true. (a) Describe how confirmation bias might be at work here. (b) Relate this to personal, or subjective, probability. (c) Discuss people's shortfalls in judging probability and randomness, and describe one way statisticians ensure true randomness in their research.

Page 29

Name:

Class:

Date:

Chap_05_5e Answer Key 1. False 2. False 3. True 4. False 5. False 6. False 7. True 8. True 9. False 10. True 11. True 12. True 13. True 14. False 15. True 16. False 17. False 18. True 19. False 20. True 21. True 22. True 23. False 24. True 25. False 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_05_5e 27. b 28. c 29. c 30. c 31. a 32. a 33. a 34. a 35. b 36. d 37. b 38. c 39. b 40. a 41. a 42. b 43. d 44. d 45. a 46. a 47. c 48. b 49. a 50. c 51. a 52. a 53. c 54. b Copyright Macmillan Learning. Powered by Cognero.

Page 31

Name:

Class:

Date:

Chap_05_5e 55. d 56. b 57. c 58. b 59. a 60. b 61. c 62. b 63. a 64. c 65. d 66. a 67. a 68. d 69. a 70. a 71. b 72. d 73. c 74. c 75. b 76. b 77. d 78. a 79. a 80. a 81. d 82. b Copyright Macmillan Learning. Powered by Cognero.

Page 32

Name:

Class:

Date:

Chap_05_5e 83. d 84. a 85. b 86. d 87. c 88. a 89. c 90. c 91. d 92. c 93. c 94. b 95. b 96. a 97. a 98. d 99. b 100. b 101. c 102. d 103. d 104. c 105. a 106. c 107. b 108. c 109. c 110. a Copyright Macmillan Learning. Powered by Cognero.

Page 33

Name:

Class:

Date:

Chap_05_5e 111. b 112. b 113. a 114. a 115. null 116. independent 117. experimental 118. research 119. percentage 120. Type I 121. Type I 122. generalizability 123. probability 124. null 125. convenience 126. control 127. null 128. replication 129. personal probability 130. confirmation 131. representative 132. random 133. illusory correlation 134. volunteer 135. rejects 136. Type II 137. successes; trials

Page 34

Name:

Class:

Date:

Chap_05_5e 138. fails to reject 139. proportion 140. The sample in this testimonial (M.G.) is unlikely to be representative for several reasons. First, this is a volunteer sample, so M.G. is likely an extraordinary rather than a representative case. Second, M.G. is just a sample of one, and sample sizes of one are unlikely to be representative of a population. M.G. herself may be susceptible to confirmation bias in which, because she has invested in these diapers, she selectively attends to evidence that is consistent with their superiority over other diapers. 141. (a) The population is all elderly adults living in assisted care facilities. (b) Random selection would not be possible in this case as it would be impossible for the psychologist to access every single elderly individual living in assisted care; there is no national database of elderly adults in assisted care living facilities. (c) Random assignment would not be possible in this case because it would be unethical and illegal to randomly place some elderly adults in assisted care facilities and others in other living arrangements. (d) The psychologist may decide to use a convenience sample. For example, the psychologist could identify assisted living facilities within a 30-mile radius from her institution and recruit individuals from these local facilities for her study. 142. The sample in this testimonial (Bazz) is unlikely to be representative for several reasons. First, this is a volunteer sample, so Bazz is likely an extraordinary rather than a representative case. Second, Bazz is just a sample of one, and sample sizes of one are unlikely to be representative of a population. Bazz herself may be susceptible to confirmation bias in which she selectively remembers listening to the CD only at those times when she was losing weight. 143. (a) The population is all children about to enter kindergarten in the particular urban area. (b) It would be impractical to assess the school readiness of every single kindergarten-bound child for a number of reasons. First, it would be very expensive. Second, it would be very time consuming: all the children might not be able to be assessed in the time frame of interest. (c) Random selection would be possible in this case because a researcher could potentially get a list of all children registered for kindergarten for the upcoming school year in the area. However, that list would have to be constantly monitored and updated for changes, such as new kids moving into the area and other children moving out. 144. (a) A likely null hypothesis is that the use of social media will not differ depending on the gender of the individual. (b) A likely research hypothesis is that men and women differ in their use of social media. Another option is that women use social media more than men. 145. If the researcher fails to replicate a long-standing effect, she may have made a Type II error, in which she has failed to reject a null hypothesis that is false. The risk associated with this error is that an important finding has been missed. The important benefits of touch therapy for preterm infants might have been missed if this Type II error occurred at the outset of research in this area, when no other work had been previously performed. 146. (a) A likely null hypothesis is that papers with a plastic cover will receive similar grades to papers not in a plastic cover. (b) A likely research hypothesis is that papers in a plastic cover will, on average, receive better grades than papers not in a plastic cover. (c) The control group is likely to be the clear plastic report cover, while the experimental group or groups would be colored or patterned plastic report covers that look ìniftyî in some way.

Page 35

Name:

Class:

Date:

Chap_05_5e 147. (a) The population of interest is preterm infants. (b) Random selection is not possible in this case because it would be impossible to get information regarding all preterm infants. (c) Random assignment would be possible in this case as the researcher could first identify a sample of preterm infants and then randomly have half of them receive touch therapy and have the other half not receive touch therapy. Specifically, as preterm infants were born, the researcher might use a random numbers table to assign the infant to either the touch therapy group or control group. 148. (a) The population of interest is all college freshmen. (b) Random selection is not possible in this case because it would be impossible to get information regarding all individuals enrolled as freshmen in college classes; there is no national data warehouse that provides access to all college students. (c) Random assignment would not be possible in this case because it would be unethical to randomly assign some freshmen to cheat. Furthermore, the researcher is interested in the natural incidence of cheating and so if he were to randomly assign people to cheat he would alter the behavior that he was attempting to measure. It is also not possible to decide who gets to be a college freshman. (d) The researcher may decide to use a convenience sample and distribute a questionnaire to all freshmen enrolled in the college at which he is working. 149. (a) A likely null hypothesis is that the weight gain for infants receiving touch therapy will be no different from that of infants not receiving touch therapy. (b) A likely research hypothesis is that weight gain of infants receiving touch therapy will be greater than that of infants not receiving touch therapy. 150. (a) The population is all individuals in prisons in the United States. (b) It would be impractical to distribute the survey to every individual in prison in the United States for a number of reasons. First, it would be very expensive. Second, it would be very time consuming: by the time the surveys were distributed, collected, and the data analyzed, years might have gone by and the purpose for conducting the survey might be obsolete. (c) True random selection would be possible in this case because a researcher could identify all individuals imprisoned in the United States (the population). But true random selection may not be practical or convenient. For example, with that passage of time, even just one hour, the prison population changes. Making sure each prisoner has an equal chance of being selected would require a constant update of the population database. 151. A Type I error is when you reject the null hypothesis when you should not have (i.e., there really is no effect). A Type II error is when you fail to reject the null hypothesis, but you should have (i.e., there really is an effect, but you failed to detect it). A Type I error is more likely to be published because there is a bias toward publishing positive results. If a study has failed to find an effect, it is difficult to know why (maybe it had a small sample or was poorly designed). Which kind of error is more dangerous depends on the content area you are dealing with. If a pharmaceutical company claims that its drug cures cancer, then it would be very dangerous if we gave the drug to cancer patients in lieu of other treatment, but the drug had no effect. At the same time, if the company failed to reject the null hypothesis and claimed that the drug did not cure cancer, when in fact it did, that also would be very dangerous because we would have a cure for cancer but not be using it. In sum, either type of error could be dangerous, and which type of error is more dangerous depends very much on the particular situation. 152. If McConnell's effect cannot be replicated, he has probably made a Type I error, in which he rejected the null hypothesis when the null hypothesis was in fact true.

Page 36

Name:

Class:

Date:

Chap_05_5e 153. A representative sample is one that has characteristics that are similar to the characteristics of the overall population to which the researcher wishes to generalize. It is important to obtain a representative sample because the researcher will be taking some quantitative measurements of the sample and will then make a statement about the overall population based on those measurements. If the sample is not representative of the population, the researchers' claims about the population will be inaccurate. 154. (a) If you believe that positive thinking results in positive outcomes, then you might selectively attend to examples that confirm this belief and ignore evidence that challenges or contradicts that belief. This is the confirmation bias. (b) Your personal probability is your own estimate or opinion about the likelihood of events. It is just your opinion, but could be an example of your positive thinking. As the text points out, statisticians define probability more formally. (c) When attempting to generate random numbers, we tend to consider what number just occurred, which means our selection of each number is not independent, which is required for randomness. To assure random assignment to groups, or random sampling when used, statisticians use random numbers tables or generators.

Page 37

Name:

Class:

Date:

Chap_06_5e Indicate whether the statement is true or false. 1. A z score computed for a sample mean is called a z statistic. a. True b. False 2. Because a z score is the number of standard deviations a score is from its mean, the first step in converting a z score back to a raw score is multiplying z and the standard deviation. a. True b. False 3. The standard deviation of the z distribution is 0. a. True b. False 4. The mean of the z distribution is 1.0. a. True b. False 5. A negative z score will convert into a raw score that is above the mean of its distribution. a. True b. False 6. A z score allows one to compare scores to each other, even when they are based on different scales. a. True b. False 7. A z score is the distance a score is from the mean of its distribution, expressed in variance. a. True b. False 8. A distribution of means comprises many, many means of samples, all of the same size. a. True b. False 9. The standard deviation of a distribution of sample means is smaller than the standard deviation of the population when the sample size is 2. a. True b. False 10. A positive z score will convert into a raw score that is above the mean of its distribution. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_06_5e 11. One implication of the central limit theorem is that a distribution of means will be more variable than a distribution of scores taken from the same population. a. True b. False 12. Standard scores represent the number of standard deviations a particular score is from the median. a. True b. False 13. A z score allows assessment of the percentile of a raw score, but an equivalent assessment of a sample mean cannot be made. a. True b. False 14. One implication of the central limit theorem is that a distribution of means will be less variable than a distribution of scores taken from the same population. a. True b. False 15. Any raw score can be converted into a z score as long as you know the median and standard deviation of the distribution. a. True b. False 16. A negative z score will convert into a raw score that is below the mean of its distribution. a. True b. False 17. Even when the distribution of scores in the population is not normal, the sampling distribution of the mean will approach normality as sample size increases. a. True b. False 18. Less than 5 percent of the distribution of scores falls beyond a z score of +/–1.0. a. True b. False 19. When comparing two z scores to assess performance on an exam, one would conclude that a student with a z of –2.3 outperformed someone with a z of 1.7 because the first score is more extreme. a. True b. False

Page 2

Name:

Class:

Date:

Chap_06_5e 20. Any raw score can be converted into a z score as long as you know the mean and standard deviation of the distribution. a. True b. False 21. The standard deviation of a distribution of means will be larger than the standard deviation of a distribution of scores. a. True b. False 22. Standard error is the standard deviation of a distribution of means. a. True b. False 23. Gerber and Malhotra (2006) found a suspiciously low number of z scores just outside of the cutoff for significance, indicating that researchers might be encouraged to "beat" the cutoff for significance. a. True b. False 24. A z score is the distance a score is from the mean of its distribution, expressed in standard deviations. a. True b. False 25. Means are less extreme than individual scores because, with means, extreme observations are combined with more average observations or scores in the opposite direction. a. True b. False 26. The percentage of area under the curve for a negative z score will be negative. a. True b. False 27. The standard deviation of the z distribution is 1.0. a. True b. False 28. Standard error is the variance of a distribution of means. a. True b. False 29. At a z score of 0, your score is at the 50th percentile. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_06_5e 30. If a score has a z score of 1, then the raw score equals to the mean. a. True b. False 31. The distribution of means is less variable than the distribution of scores. a. True b. False 32. A z score allows one to compare scores to each other, but not when they are based on different scales. a. True b. False 33. The mean of the z distribution is 0. a. True b. False 34. In a distribution with a mean of 150 and a standard deviation of 20, a z score of –1.0 would convert into a raw score of 120. a. True b. False 35. A z statistic is used to refer to a distribution of scores. a. True b. False 36. If you have a z score of 0, then you have a raw score equal to the mean. a. True b. False 37. Data drawn from a normally distributed population approaches a normal distribution as sample size increases, thus making sample size important in relation to the normal curve. a. True b. False 38. A positive z score will convert into a raw score that is below the mean of its distribution. a. True b. False 39. A z statistic is used to refer to a distribution of means. a. True b. False

Page 4

Name:

Class:

Date:

Chap_06_5e 40. Any raw score can be converted into a z score as long as you know the mean and median of the distribution. a. True b. False 41. When attempting to create a distribution of means, we sample with replacement; that is, we do not put data back in the sample after we have computed the mean of those data. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 42. Gibson (1986) asked a sample of college students to complete a self-esteem scale on which the midpoint of the scale was the score 108. He found that the average self-esteem score for this sample was 135.2, well above the actual midpoint of the scale. Given that the standard deviation of self-esteem scores was 28.15, what would be the z score for a person whose self-esteem score was 104.28? a. –1.10 b. –0.85 c. 0.85 d. 1.10 43. In a normal standard curve, approximately _____ percent of scores fall within 2 standard deviations from the mean a. 34 b. 48 c. 96 d. 68 44. Repeated sampling of _____ approximates a normal curve, even when the underlying population is skewed. a. means b. standard deviations c. variance estimates d. population parameters 45. When creating a distribution of means, it is important that whatever scores are sampled to compute the means are: a. placed back into the population for additional sampling. b. separated out from the population so that they cannot be resampled. c. recorded in order to create a distribution of scores. d. balanced across the mean so that extreme scores are controlled.

Page 5

Name:

Class:

Date:

Chap_06_5e 46. According to _____, as sample size increases, the distribution of _____ assumes a normal curve. a. the standardized distribution; sample scores b. the central limit theorem; sample means c. the z-score distribution; population scores d. hypothesis testing; raw scores 47. Two students from two different high schools recently took a math test. The first student correctly answered 37 questions and the second student correctly answered 45 questions. What can be concluded from the two students' test scores? a. The second student is smarter than the first student. b. The two students did equally well on the exam. c. The two students cannot be compared because no standardization procedure was used to permit comparisons. d. The two students cannot be compared because the scores did not form a linear curve. 48. The formula for calculating a z score is: a. z = (σ – X)/µ. b. z = (µ – X)/ σ. c. z = (X – µ)/ σ. d. z = (X – σ)/µ. 49. In a normal standard curve, approximately _____ percent of scores fall within 1 standard deviation from the mean. a. 34 b. 46 c. 96 d. 68 50. Gerber and Malhotra (2006) found an irregularly high spike in published z scores: a. just greater than 1.96. b. just less than 1.96. c. around 1.00. d. out as far as 9.16. 51. A z score is a measure of: a. how far away from the mean a score is in terms of standard deviations. b. how far away from the mean a score is in terms of inches. c. the strength of the relationship between two variables. d. the strength of the relationship between a score and its mean.

Page 6

Name:

Class:

Date:

Chap_06_5e 52. The z distribution is equivalent to a distribution of _____ scores. a. mean b. linear c. z d. raw 53. Mason wanted to know his approximate score on the final exam for his statistics class. His professor hinted that his score was well above the class average. The professor announced that the mean for the class final exam was 88 with a standard deviation of 7. Given Mason's z score of 1.67, what is the raw score for his exam grade? a. 100.00 b. 99.69 c. 102.45 d. 88.17 54. According to the 2015 annual report of the American Psychological Association's on salaries in psychology. The average salary for those working in a teaching position was $71,471, with a standard deviation of $24,703. What is the z score of a professor making $85,500? a. –0.85 b. –0.57 c. 0.57 d. 0.85 55. A distribution of means would be more likely to have a(n) _____ compared to a distribution of raw scores. a. higher variance b. lower variance c. higher standard deviation d. equal standard deviation 56. z scores are useful because they: a. allow us to convert raw scores to mean scores, compare scores from different samples, and transform populations into samples. b. transform linear scores into nonlinear scores, convert nonlinear scores back into linear scores, and allow us to obtain comparisons between nonlinear and linear scores. c. give us an understanding of where a score falls in relation to the mean of its underlying population, allow comparisons to be made between scores from different distributions, and permit the transformation of z scores into percentiles. d. reduce the probability of Type I and Type II errors, allow us to compare raw scores with standard scores, and permit the transformation of raw scores into percentiles.

Page 7

Name:

Class:

Date:

Chap_06_5e 57. A distribution of scores has a mean of 20.4 with a standard deviation of 0.89. Compare a score of 21.26 with a z score of 1.2. Which statement is correct? a. The score of 21.26 is greater. b. The z score of 1.2 is greater, resulting in a raw score of 21.47. c. The z score of 1.2 is greater, resulting in a raw score of 21.29. d. The z score of 1.2 is greater, resulting in a raw score of 22.39. 58. The z distribution _____ has a mean of _____. a. always; 0 b. sometimes; 0 c. always; 1 d. sometimes; 1 59. The second step in converting a z score into a raw score is: a. adding the mean of the population to the product obtained from multiplying the z score and standard deviation. b. subtracting the mean of the population from the product obtained by multiplying the z score and standard deviation. c. dividing the mean of the population into the product obtained by multiplying the z score and standard deviation. d. multiplying the mean of the population and the product obtained from multiplying the z score and standard deviation. 60. The mean for the population is 82 with a standard deviation of 6. Given a z score of 1.45, what is the raw score? a. 73.30 b. 89.45 c. 90.70 d. 8.70 61. The first step in converting a z score into a raw score is multiplying the z score by the: a. population standard less the population mean. b. the raw score. c. population standard deviation. d. population mean.

Page 8

Name:

Class:

Date:

Chap_06_5e 62. The process of standardization involves the conversion of raw scores to _____ scores. a. linear b. standard c. normal d. nonlinear 63. The symbol for the population standard deviation is: a. µ. b. X. c. z. d. σ. 64. Two students recently took trigonometry class tests. The students are at different schools but wanted to compare their performance. The first student scored 80 on the test. Her class average was 85 with a standard deviation of 5. The second student scored 65. Her class average was 50 with a standard deviation of 10. Which student did better? a. first student because she had a higher score b. second student because she had an average score c. first student because she performed better relative to her class d. second student because she performed better relative to her class 65. The z distribution is a normal distribution of _____ scores. a. sample mean b. population mean c. standardized d. raw 66. The term _____ is used for the distribution of means in place of the term standard deviation. a. standard variance b. population variance c. mean variance d. standard error 67. The second step in calculating a z score is expressing the obtained values in: a. standard deviation units. b. linear form. c. nonlinear form. d. distribution of means.

Page 9

Name:

Class:

Date:

Chap_06_5e 68. The mean of the distribution of a set of z scores is: a. always 0. b. always 1. c. the same as the mean of the distribution of raw scores. d. the score corresponding to the 50th percentile in the raw score distribution. 69. The formula for the standard error is: a. b. c. d. 70. On the first statistics exam, the class average was 72 with a standard deviation of 6. Reid scored 84. What is his z score? a. –1.0 b. –2.0 c. 6.0 d. 2.0 71. As sample size _____, the spread of distribution of means _____. a. increases; increases b. decreases; stays the same c. increases; decreases d. decreases; decreases 72. A person with a z score of 0 would have a raw score equal to: a. the lowest score in the distribution of raw scores. b. the mean of the distribution of raw scores. c. the highest score in the distribution of raw scores. d. 0. 73. What percent of scores fall beyond 2 standard deviations away from the mean? a. 28 b. 14 c. 4 d. 2

Page 10

Name:

Class:

Date:

Chap_06_5e 74. If samples have at least _____ scores, the distribution of means will most likely approximate a normal curve. a. 30 b. 50 c. 88 d. 100 75. Which score is more extreme: 0.78 or –0.93? a. 0.78 b. –0.93 c. z scores do not allow us to assess this. d. Scores with different signs cannot be compared. 76. The formula for calculating the raw score from a z score is: a. X = z(σ) + µ. b. X = z(µ)/ σ. c. z = (σ – µ)/X. d. z = (X – σ)/µ. 77. In a normal standard curve, which percentile corresponds to a z score of –1.0? a. 34 b. 68 c. 16 d. 45 78. To compare two scores that are measured on different scales, one needs to transform the scores into: a. standard deviations. b. means. c. z scores. d. population parameters. 79. Kelly scored 40 on a standardized test of reading ability where the mean score is 50 and the standard deviation is 10. Based on this information, what is Kelly's z score? a. –2.0 b. 2.0 c. 1.0 d. –1.0

Page 11

Name:

Class:

Date:

Chap_06_5e 80. A _____ is composed of means based on samples rather than raw scores. a. distribution of means b. distribution of z scores c. standardized distribution d. percentile distribution 81. Because of _____, skewed distributions approximate normal curves when means are based on larger samples. a. kurtosis b. the central limit theorem c. hypothesis testing d. z scores 82. The distribution of means based on a sample size of 30, pulled from a population distribution with a mean of 100 and a standard deviation of 15, would have a standard error of: a. 0.50. b. 3.35. c. 2.74. d. 3.33. 83. The symbol for the standard error is: a. σM. b. σS . c. µM. d. µσ. 84. The z distribution always has a standard deviation of _____. a. 0 b. 0.5 c. 1 d. 10 85. When calculating a z score for a distribution of means, the z score is referred to as a: a. standard score. b. standardized score. c. z statistic. d. central limit theorem.

Page 12

Name:

Class:

Date:

Chap_06_5e 86. In a normal standard curve, what percentile corresponds to a z score of 1.0? a. 84 b. 68 c. 96 d. 45 87. The z distribution _____ has a standard deviation of _____. a. always; 0 b. sometimes; 0 c. always; 1 d. sometimes; 1 88. Sample means based on at least _____ scores tend to approximate a normal distribution, even when the underlying population is skewed. a. 10 b. 25 c. 30 d. 50 89. Which of these z scores from a single distribution of scores corresponds to the raw score farthest from the mean of the distribution? a. –1.4 b. –0.5 c. 0.2 d. 1.3 90. A _____ represents the number of standard deviations a particular score is from the mean average. a. z score b. standard mean c. standardization score d. skewed score 91. Since one rarely has access to an entire population, one typically calculates the mean of a sample and: a. compares that to the z distribution. b. computes a standardized score for that mean. c. compares that to a distribution of mean by calculating a z statistic. d. standardizes that using information about the center and spread of the population of scores.

Page 13

Name:

Class:

Date:

Chap_06_5e 92. If a distribution of scores has a mean of 40 and a standard deviation of 10, then a score of 70 has a z score that is _____ standard deviation(s) from the mean. a. 1 b. 2 c. 3 d. 4 93. The mean for the population is 66 with a standard deviation of 8.78. Given a z score of 2.54, what is the raw score? a. 88.30 b. 74.78 c. 73.32 d. 68.54 94. The _____ curve forms a symmetrical and mathematically defined bell-shaped pattern. a. asymmetric b. unstandardized c. normal d. nonlinear 95. The symbol for the population mean is: a. µ. b. X. c. z. d. σ. 96. As sample size _____, the mean of a distribution of means _____. a. increases; increases b. increases; stays the same c. increases; decreases d. decreases; increases 97. The z distribution always has a mean of _____. a. 0 b. 0.5 c. 1 d. 10

Page 14

Name:

Class:

Date:

Chap_06_5e 98. Alex scored 45 on his final exam. The class's average score was 50, with a standard deviation of 10. What is Alex's z score? a. –1 b. –0.5 c. 0.5 d. 1.5 99. Any raw score can be converted into a z score as long as you know the _____ and _____ of the distribution. a. median; mean b. median; standard deviation c. mean; standard deviation d. mean; range 100. The distribution of scores in a a sample, drawn from a normal population, will approach normality as: a. number of scores increases. b. number of scores decreases. c. variance increases. d. variance decreases. 101. If a distribution of scores has a mean of 50 and a standard deviation of 10, then a score of 40 has a z score that is _____ standard deviation(s) _____ the mean. a. 2; below b. 1; above c. 1; below d. 2; above 102. The first step in calculating a z score is calculating: a. a standard deviation score. b. the raw scores and subtracting from the mean of the sample. c. the difference between a particular score and the population mean. d. the difference between a particular score and the sample mean. 103. The formula for z based on the mean of a sample is: a. z = (M – µM)/µs. b. z = (X – µM)/µs. c. z = (X – µM)/ σM. d. z = (M – µM)/ σM.

Page 15

Name:

Class:

Date:

Chap_06_5e 104. Given the properties of the standard normal curve, we know that _____ percent of all scores fall below the mean and _____ percent fall above the mean. a. 68; 68 b. 50; 50 c. 34; 34 d. 48; 48 105. Findings that are in the most extreme _____ percent are considered significant and worthy of publishing. a. 1 b. 5 c. 10 d. 50 106. Tej scored 60 on his final exam. His class's average score was 55, with a standard deviation of 5. How many standard deviations is Tej's score from the mean? a. 1 standard deviation above the mean b. 1 standard deviation below the mean c. 2 standard deviations above the mean d. 2 standard deviations below the mean 107. A _____ is a distribution of z scores. a. normal curve b. z distribution c. standard linear distribution d. standardization Enter the appropriate word(s) to complete the statement. 108. Nearly all scores fall within _______ standard deviations of the mean.

109. In the z distribution, _______ percent of scores fall above the mean.

Page 16

Name:

Class:

Date:

Chap_06_5e 110. A distribution of a small set of scores is _______ likely to be normally distributed than a distribution of a large set of scores.

111. The number of _______ a particular score is from the mean is the z score.

112. A z score of –1.0 is equivalent to the _______ percentile.

113. When converting a z score into a raw score, we begin by multiplying the z score by the _______.

114. A z score of +2.0 is equivalent to the _______ percentile.

115. Two scores that are based on two different scales can be directly compared once they are converted into _______.

Page 17

Name:

Class:

Date:

Chap_06_5e 116. As you increase the size of a sample, the distribution of the sample will approach the _______ as long as the underlying population is normally distributed.

117. A distribution of means is less _______ than a distribution of scores.

118. In the z distribution, _______ percent of scores fall below the mean.

119. When converting a raw score into a z score, we begin by subtracting the _______ from the raw score.

120. A normal distribution of standardized scores is the _______ distribution.

121. With large sample sizes, the shape of the distribution of the mean will be _______.

122. The standard deviation of a distribution of means is called the _______.

Page 18

Name:

Class:

Date:

Chap_06_5e 123. Approximately 68 percent of scores fall within _______ standard deviation(s) of the mean.

124. A _______ is a z computed on a sample mean rather than a raw score.

125. A z score of –2.0 is equivalent to the _______ percentile.

126. The number of standard deviations a particular score is from the mean is the _______.

127. The primary assertion of the _______ is that a distribution of sample means approaches a normal curve as sample size increases.

128. A z score of 0 is equivalent to the _______ percentile.

129. Approximately 96 percent of scores fall within _______ standard deviation(s) of the mean.

Page 19

Name:

Class:

Date:

Chap_06_5e 130. A z score of +1.0 is equivalent to the _______ percentile.

131. Distributions of means computed from samples of at least _______ observations usually produce an approximately normal curve.

132. When creating a distribution of means, it is important to retain all observations in the distribution for future sampling; this is known as sampling with _______.

133. Findings that are in the most extreme _______ percent are considered significant and worthy of publishing.

134. A person who scored exactly at the mean of the distribution of raw scores would have a z score of _______.

135. In the z distribution, _______ percent of scores fall between the mean and a z score of 1.0, or the mean and a z score of –1.0.

Page 20

Name:

Class:

Date:

Chap_06_5e 136. A local community college reported that the average SAT score of its students is 960 with a standard deviation of 82. If a student scored 975 on the SAT, what is the student's z score? What formula must you use to solve this problem? Show your work.

137. A local community college reported that the average SAT score of its students is 960 with a standard deviation of 82. If a student had a z score of 3, what would be the student's SAT score? What formula must you use to solve this problem? Show your work.

138. Here are some fictional scores on a recent exam: 63, 72, 83, 95, 92, 68, 76, 85. Calculate a mean and a standard deviation. Calculate a z score for a raw score of 65.

139. Andrew scored 68 out of a possible 100 on his midterm chemistry examination. The distribution of the class had a mean of 59 and a standard deviation of 8.7. A friend of Andrew's who is in a different chemistry class scored 78 out of a possible 100. His class distribution had a mean of 66 and a standard deviation of 14. Relative to the performance of their classes, who did better?

140. Neil scored 67 on his calculus class final and was devastated by his failing grade. His class distribution had a mean of 54 and a standard deviation of 7. Greg was quite proud of his score of 88 for his Spanish language class. His class distribution had a mean of 97 with a standard deviation of 6. Who did better relative to their class?

Page 21

Name:

Class:

Date:

Chap_06_5e 141. A health researcher investigated the relationship between IQ and apple consumption. He concluded that people who eat at least one apple a day have higher IQs than people who do not. He felt very confident in his results because he had a substantial sample size of over 10,000 people and felt that his results were defensible. What is a potential problem with the researcher's conclusion?

142. The average score on a final exam for a statistics class was 87 with a standard deviation of 6.5. Andrew scored 87, a score in the average range. (a) What is Andrew's z score? What formula did you use, if any? (b) Describe the nature of the z distribution by providing its mean and standard deviation and explaining their values.

143. Here are some fictional exam scores: 63, 72, 83, 95, 92, 68, 76, 85. (a) Calculate the standard deviation of this sample of scores. (b) Calculate the standard error. What formula is needed to obtain the standard error? (c) Explain why the standard error and standard deviation are different.

144. Laura learned that the average GPA in her school is 3.6 with a standard deviation of 0.3. (a) If her GPA is a 3.0, what is her z score? (b) What do we know about the percentile of Laura's GPA score? Reflect on what this might tell us about Laura's school.

145. Tests such as the SAT, ACT, and GRE are referred to as standardized tests. (a) What is meant by standardized tests? (b) How are the SATs, ACTs, and GREs related to the process of standardization?

Page 22

Name:

Class:

Date:

Chap_06_5e Answer Key 1. True 2. True 3. False 4. False 5. False 6. True 7. False 8. True 9. True 10. True 11. False 12. False 13. False 14. True 15. False 16. True 17. True 18. False 19. False 20. True 21. False 22. True 23. True 24. True 25. True 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 23

Name:

Class:

Date:

Chap_06_5e 27. True 28. False 29. True 30. False 31. True 32. False 33. True 34. False 35. False 36. True 37. True 38. False 39. True 40. False 41. False 42. a 43. c 44. a 45. a 46. b 47. c 48. c 49. d 50. a 51. a 52. c 53. b 54. c Copyright Macmillan Learning. Powered by Cognero.

Page 24

Name:

Class:

Date:

Chap_06_5e 55. b 56. c 57. b 58. a 59. a 60. c 61. c 62. b 63. d 64. d 65. c 66. d 67. a 68. a 69. a 70. d 71. c 72. b 73. c 74. a 75. b 76. a 77. c 78. c 79. d 80. a 81. b 82. c Copyright Macmillan Learning. Powered by Cognero.

Page 25

Name:

Class:

Date:

Chap_06_5e 83. a 84. c 85. c 86. a 87. c 88. c 89. a 90. a 91. c 92. c 93. a 94. c 95. a 96. b 97. a 98. b 99. c 100. a 101. c 102. c 103. d 104. b 105. b 106. a 107. b 108. 3 109. 50 110. less Copyright Macmillan Learning. Powered by Cognero.

Page 26

Name:

Class:

Date:

Chap_06_5e 111. standard deviations 112. 16th 113. standard deviation 114. 98th 115. z scores 116. normal curve 117. variable 118. 50 119. mean, population mean 120. z 121. normal 122. standard error 123. 1, one 124. z statistic 125. 2nd 126. z score 127. central limit theorem 128. 50th 129. 2 130. 84th 131. 30 132. replacement 133. 5 134. 0 135. 34

Page 27

Name:

Class:

Date:

Chap_06_5e 136. In order to solve this problem, we need to solve for z using the following formula: z = (X – µ)/σ z = (975 – 960)/82 = 15/82 = 0.18 The student's z score is 0.18. 137. In order to solve this problem, we will need to solve for X using the following formula: X = z(σ) + µ X = 3(82) + 960 = 246 + 960 = 1206 The student's SAT score is 1206. 138. The mean for the scores is 79.25. The standard deviation for the scores is 10.67. The z score for score 65 is – 1.34. 139. Andrew's z score: (68 – 59)/8.7 = 9/8.7 = 1.03. His friend's z score: (78 – 66)/14 = 12/14 = 0.86. Although both scored relatively well and above the class averages, Andrew performed better on the chemistry exam relative to his friend. 140. Neil's z score: (67 – 54)/7 = 13/7 = 1.86. Greg's z score: (88 – 97)/6 = –9/6 = –1.50. Relative to the performance of their classes, Neil performed better on the exam compared to Greg. 141. Students should be encouraged to discuss possible flaws in the researcher's conclusions. One potential problem with the researcher's conclusions is that given the large sample size, the study's results could be biased. The shape of a distribution is affected by the size of the sample. Given that the researcher has such a large sample size, and that the shape of the distribution is affected by sample size, the researcher's findings could be an artifact of the size of the sample. Specifically, as sample size increases, the distribution of means against which the sample mean would be compared becomes narrower. With such a narrow distribution, a significant sample mean would be easier to obtain. As discussed in other chapters, sample sizes that are very large can actually raise suspicion about findings. 142. (a) Andrew's z score is 0. The mean for the population is 87; a z score of 0 is equivalent to the mean score. No calculations or formulas were needed given that Andrew's score is at the mean and that it is 0 standard deviations from the mean average. (b) The z distribution has a mean of zero and a standard deviation of 1.0. The z score standardization converts raw scores into the number of standard deviations they are from the mean. If a score is at the mean, it is therefore no standard deviations from the mean, and results in a z score of 0. If a score is one standard deviation from the mean, the z score is 1.0, positive or negative.

Page 28

Name:

Class:

Date:

Chap_06_5e 143. (a) The standard deviation is 10.67. (b) In order to obtain the standard error, the following formula is needed: . The sample size, N, is 8. The square root of 8 is 2.83. Dividing the standard deviation of 10.67 by the square root of the sample, 2.83, the standard error is 3.77. (c) The standard deviation of 10.67 is much larger than the standard error of 3.77. While the standard deviation describes the spread of scores, the standard error describes the spread of means based on a given sample size, in this case, 8. Distributions of means have less variability than distributions of scores, according to the central limit theorem. 144. (a) z = –2.0. (b) This GPA is very low, with only approximately 2 percent of GPAs being lower and around 98 percent being higher. This school seems to have some serious grade inflation. 145. Students should be encouraged to share different ideas about what is meant by standardized testing and could engage in a discussion of the pros and cons of standardized testing in order to promote contextual learning of concepts. (a) Standardized tests are tests that have standard administration, scoring, and interpretation procedures. (b) The SATs, ACTs, and GREs are related to the process of standardization because the standardized testing practices involve converting raw scores to z scores for comparisons to be made.

Page 29

Name:

Class:

Date:

Chap_07_5e Indicate whether the statement is true or false. 1. HARKing involves repeatedly testing hypotheses until a researcher obtains statistically significant results. a. True b. False 2. If one fails to reject the null hypothesis, it means that the results of the study are not statistically significant. a. True b. False 3. A z test result of –1.95 would allow one to reject the null hypothesis, assuming a two-tailed test with p = 0.05. a. True b. False 4. If 22 percent of scores fall between a negative z score and the mean, then 38 percent of scores fall below that z score. a. True b. False 5. A nondirectional test looks in both tails of the distribution. a. True b. False 6. A nonparametric test is NOT based on assumptions about the population. a. True b. False 7. The same logic used to find an individual's percentile rank can be applied to a mean to determine the relative standing of the mean in a distribution of means. a. True b. False 8. For a given population with a mean of 95 and a standard deviation of 15, a score of less than 80 is less likely to occur than a mean score of 80 based on 20 observations. a. True b. False 9. A z test result of 1.99 would allow one to reject the null hypothesis, assuming a two-tailed test with p = 0.05. a. True b. False

Page 1

Name:

Class:

Date:

Chap_07_5e 10. The critical values for a statistical test mark the value beyond which one can fail to reject the null hypothesis. a. True b. False 11. HARKing involves changing the hypotheses after the data has been collected. a. True b. False 12. The critical values for a statistical test mark the value beyond which one can reject the null hypothesis. a. True b. False 13. If one fails to reject the null hypothesis, it means that the results of the study are statistically significant. a. True b. False 14. Stopping data collection once your results allow you to reject the null hypothesis is on example of pHacking. a. True b. False 15. If one rejects the null hypothesis, it means that the results of the study are important. a. True b. False 16. A one-tailed test effectively doubles the p level in that one tail. a. True b. False 17. A score equal to the mean of its distribution falls at the 50th percentile. a. True b. False 18. A parametric test is NOT based on assumptions about the population. a. True b. False 19. If a result is statistically significant, then chance is said NOT to account for the outcome. a. True b. False

Page 2

Name:

Class:

Date:

Chap_07_5e 20. Most hypothesis tests are robust to violations of the assumption that the distribution of the population is approximately normal. a. True b. False 21. The process of p-Hacking involves repeatedly testing hypotheses until a researcher obtains statistically significant results. a. True b. False 22. As long as the sample size is 30 or more, one can assume that the underlying population has an approximately normal distribution. a. True b. False 23. A parametric test is based on assumptions about the population. a. True b. False 24. The z table only allows for the calculation of percentiles for negative z values. a. True b. False 25. A nonparametric test is based on assumptions about the population. a. True b. False 26. As long as the sample size is 30 or more, the results of a study can be generalized to most populations. a. True b. False 27. If 22 percent of scores fall between a negative z score and the mean, then 28 percent of scores fall below that z score. a. True b. False 28. A nondirectional test looks in neither tail of the distribution. a. True b. False 29. When determining the z value for a mean rather than a score, the value is called a z statistic. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_07_5e 30. If a result is statistically significant, then chance is said to account for the outcome. a. True b. False 31. The z table only allows for the calculation of percentiles for positive z values. a. True b. False 32. When determining the z value for a mean rather than a score, the value is called a standardized z. a. True b. False 33. The process of p-Hacking involves changing the hypotheses after the researcher has collected their data. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 34. The distribution of sample means has the same mean as the distribution of scores for the population, and the spread is: a. the same. b. greater. c. no longer calculated. d. smaller.

Page 4

Name:

Class:

Date:

Chap_07_5e Study Description: Texting During Class Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a pre-recorded psychology lecture. At the end of the 20-minute lecture, students answered a 17 question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who hadn't texted during the lecture. 35. (Study Description: Texting During Class) Which statement is the null hypothesis for this study? a. There will be no difference in the quiz grades of the two groups after both groups listen to the 20minute lecture. b. The quiz grades of the students in the texting group will be higher than that of the students in the nontexting group after both groups listen to the 20-minute lecture. c. The quiz grades of the students in the non-texting group will be higher than that of the students in the texting group after both groups listen to the 20-minute lecture. d. The quiz grades of the students in the texting group and the students in the non-texting group will differ after both groups listen to the 20-minute lecture. 36. When attempting to find a percentage associated with a z score, the first step involves a _____ to _____ transformation. a. z score; raw score b. raw score; z score c. standard error; z score d. z score; standard error 37. To determine a person's percentile, first convert the person's _____ score to a z score. a. normalized b. raw c. median d. formal 38. To convert an individual's percentile rank into a raw score, one needs to know the mean and standard: a. deviation for the population. b. error of the sampling distribution. c. deviation of the sample. d. error of the population.

Page 5

Name:

Class:

Date:

Chap_07_5e 39. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Ryan studies 15 hours per week. What percent of students are at least as extreme as Ryan, in both directions? (Round z score to two decimal places.) a. 6.68 b. 8.18 c. 13.36 d. 43.32 40. One rejects the null hypothesis only when: a. the sample mean is larger than the population mean. b. the p value associated with the test statistic is smaller than the p level chosen. c. the sample mean is smaller than the population mean. d. the p value associated with the test statistic is greater than the p level chosen. 41. A hypothesis test is said to be _____ when it produces fairly accurate results, even when some of the assumptions underlying the hypothesis test are violated. a. robust b. invincible c. reliable d. fair 42. Donald hypothesized that introducing tablet computers to the classroom would increase student learning and achievement. After analyzing his data, Donald found that usage of tablet computers actually led to declines in student test grades. When Donald submitted his research for publication, he reported that they had hypothesized that tablet computers would interfere with student learning because of their potential to distract students. In this example, Donald has: a. decreased the likelihood of a Type I error. b. practiced HARKing. c. engaged in p-Hacking. d. increased the likelihood of a Type II error. 43. The _____ hypothesis is usually the "boring" one. a. null b. research c. normalized d. standard

Page 6

Name:

Class:

Date:

Chap_07_5e 44. Mehl et al. (2007) published a study in the journal Science reporting the results of an extensive study of 396 men and women comparing the number of words uttered per day by each sex. If Mehl et al. was testing the idea that women talk more than men, what type of hypothesis test would he conduct? a. one-tailed test b. two-tailed test c. nonparametric test d. z test 45. Inferential statistical analyses that are based on a set of assumptions about the underlying population are: a. inaccurate. b. permissible only when certain assumptions about the sample are met. c. nonparametric tests. d. parametric tests. 46. When attempting to find a percentage associated with a z score, the second step involves looking up: a. a raw score on the raw score table. b. a standard score on the means table. c. a z score on the z table. d. the median to determine skewness. 47. The symbol for the null hypothesis is _____ and the symbol for the research hypothesis is _____. a. H1; H0 b. H0; H1 c. µ; M d. µ0; µ1 48. The statistical concept of p levels is often referred to as: a. betas. b. alphas. c. central tendency. d. variability. 49. When hypothesis testing, it is more common to use a _____ rather than a _____. a. two-tailed test; one-tailed test b. one-tailed test; two-tailed test c. null hypothesis; research hypothesis d. research hypothesis; null hypothesis

Page 7

Name:

Class:

Date:

Chap_07_5e 50. Because of the principle of _____, when sample sizes are at least 30, the distribution will most likely resemble a normal distribution. a. parametric statistics b. nonparametric statistics c. the central limit theorem d. robustness 51. What is the percentage of observations that fall between z scores of –1.04 and 0.51? a. 19.50 b. 35.08 c. 45.42 d. 54.58 52. If a dependent variable is nominal, the assumption that the _____ should not be made. a. variable is normally distributed b. variable is assessed using a scale measure c. participants are randomly selected d. participants are randomly assigned 53. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Treating this class as the population, what is the percentile for a student in the class who studies 8 hours a week? (Round z score to two decimal places.) a. 5.17 b. 44.04 c. 47.06 d. 97.06 54. Under what conditions is it permissible to proceed with a hypothesis test even though the assumption that participants are randomly selected is violated? a. We are cautious about generalizing the results. b. The sample size is 30 or more. c. The data are not clearly ratio. d. The data are not clearly nominal or ordinal.

Page 8

Name:

Class:

Date:

Chap_07_5e Study Description: Texting During Class Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a pre-recorded psychology lecture. At the end of the 20-minute lecture, students answered a 17 question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who hadn't texted during the lecture. 55. (Study Description: Texting During Class) Which statement is the research hypothesis for this study? a. There will be no difference in the quiz grades of the two groups after both groups listen to the 20minute lecture. b. The quiz grades of the students in the texting group will be higher than that of the students in the nontexting group after both groups listen to the 20-minute lecture. c. The quiz grades of the students in the non-texting group will be higher that of the students in the texting group after both groups listen to the 20-minute lecture. d. The quiz grades of the students in the texting group and the students in the non-texting group will differ after both groups listen to the 20-minute lecture. Study Description: Tail Wagging A New York Times article published on April 24, 2007, reported the research of Dr. Giorgio Vallortigara, a neuroscientist at the University of Trieste, Italy, and his two colleagues. The researchers asked whether a dog wags its tail in a preferred direction in response to positive stimuli and in another direction in response to negative stimuli. To answer their question, they recruited 30 dogs that were family pets. Filming each dog from above, they allowed it to view (through a slat in its cage) three positive stimuli separately, in order of descending positivity: its owner, an unfamiliar human, and a cat. All the dogs responded by wagging their tails to the right. But when the dogs were presented with an unfamiliar, aggressive dog, a negative stimulus, all dogs wagged their tails to the left. 56. (Study Description: Tail Wagging) Which statement is the research hypothesis for this study? a. A dog's tail will wag more to the left in response to positive stimuli and more to the right in response to negative stimuli. b. There will be no difference in a dog's tail wagging while viewing positive and negative stimuli. c. A dog's tail will wag differently in response to positive stimuli than to negative stimuli. d. A dog's tail will wag more to the right in response to positive stimuli and more to the left in response to negative stimuli.

Page 9

Name:

Class:

Date:

Chap_07_5e 57. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Treating this class as the population, what percent of students study more than 8 hours? (Round z score to two decimal places.) a. 97.06 b. 65.91 c. 55.96 d. 44.04 58. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Savannah is working two jobs and struggles to find time to study, averaging only 3 hours per week. What percent of students are at least as extreme as Savannah, in both directions? (Round z score to two decimal places.) a. 48.40 b. 40.82 c. 9.18 d. 18.36 59. Under what conditions is it permissible to proceed with a hypothesis test even though the assumption that the population distribution is approximately normal is violated? a. We are cautious about generalizing the results. b. The sample size is 30 or more. c. The data are not clearly ratio. d. The data are not clearly nominal or ordinal. 60. A research study which meets all the assumptions of the parametric test used to analyze the data: a. continues with the performance of the appropriate nonparametric follow-up test. b. has no bearing on the results of the research study. c. produces higher quality results than does a research study that fails to meet some of the assumptions. d. produces lower quality results than does a research study that fails to meet some of the assumptions. 61. The percentile is the: a. same thing as a percentage. b. percentage of scores falling at or below a particular raw score. c. percentage of scores falling at or above a particular raw score. d. percentage of scores falling between a particular raw score and the mean.

Page 10

Name:

Class:

Date:

Chap_07_5e 62. What are the consequences of failing to meet the assumptions of a parametric test when performing research? a. If we fail to meet the assumptions of a parametric test, then we will be unable to make any inferences regarding the population. b. If we fail to meet one assumption of a parametric test, then we cannot perform statistical analysis on the results of the study and will be limited to qualitative descriptions of the data. c. If we fail to meet even one assumption of a parametric test, then we should perform a nonparametric test instead. d. If the test is robust to the violation, it does not invalidate the research; however, it may make it more difficult to detect effects. 63. When is the assumption that the dependent variable is assessed on a scale measure NOT met? a. We are cautious about generalizing the results. b. The sample size is greater than 30. c. The data are clearly ratio. d. The data are clearly nominal or ordinal. 64. The statement "It is hypothesized that adults who increase their consumption of fruits and vegetables will score lower on depression tests compared to adults who do not change their diet of fruits and vegetables" best illustrates a: a. one-tailed test. b. two-tailed test. c. null hypothesis. d. z test. 65. If the percentage of scores falling between the mean and a z score of 0.60 is 22.57, then what is the percentage of scores falling below a z score of –0.60? a. –22.57 b. 22.57 c. 27.43 d. 72.57 66. Assume the average height for American women is 64 inches with a standard deviation of 2 inches. What percent of groups of size 25 would have mean heights of less than 62 inches? (Round z score to two decimal places.) a. 50.00 b. 0.40 c. 0.13 d. 0

Page 11

Name:

Class:

Date:

Chap_07_5e 67. If the percentage of scores falling between the mean and a z score of 0.40 is 15.54, then what is the percentage of scores falling between the mean and a z score of –0.40? a. 15.54 b. –15.54 c. 34.46 d. 65.54 68. A z score _____ the mean will always be _____. a. below; positive b. above; positive c. above; negative d. equal to; positive 69. To compare a group mean to a population, compute the: a. z score. b. z statistic. c. mean square error. d. proportionate reduction in error. 70. The statement "It is hypothesized that depressed and anxious participants will differ on reaction time measures" best illustrates a: a. one-tailed test. b. two-tailed test. c. null hypothesis. d. z test. 71. _____ requires that all members of a population have an equal chance of being selected for a study. a. Random selection b. Random assignment c. Normal distribution d. Scale variable assumption 72. A z score _____ the mean will always be _____. a. below; positive b. below; negative c. above; negative d. equal to; positive

Page 12

Name:

Class:

Date:

Chap_07_5e 73. Which of the following statements is NOT an assumption of the z test? a. The distribution of the population is approximately normal. b. The distribution of the sample is normal. c. Participants were randomly selected from the population. d. The dependent variable is assessed using a scale measure. 74. What proportion of a normal distribution corresponds to z scores less than 1.14? a. 0.3729 b. 0.1271 c. 0.5271 d. 0.8729 75. If one rejects the null hypothesis, the result is said to be: a. acceptable. b. important. c. statistically significant. d. valid. 76. HARKing refers to an experimenter: a. conducting the same study with different samples and getting the same results each time. b. changing their research hypotheses after their data has been collected. c. repeatedly analyzing data until they find a statistically significant result. d. stopping data collection as soon as the data indicates a significant difference between groups. 77. One rejects the null hypothesis when the test statistic falls in which area of the distribution? a. critical value b. critical region c. region of determination d. probability region 78. If one fails to reject the null hypothesis, the result is said to be: a. unacceptable. b. unimportant. c. not statistically significant. d. valid.

Page 13

Name:

Class:

Date:

Chap_07_5e 79. The _____ level is the probability used to determine the critical values, or cutoffs, in hypothesis testing. a. z b. M c. p d. s 80. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Treating the class as the population, what is the percentile for a student in the class who studies 11 hours a week? (Round z score to two decimal places.) a. 21.23 b. 28.77 c. 71.23 d. 78.77 81. A z score _____ the mean will always be _____. a. below; positive b. above; negative c. equal to; negative d. equal to; zero 82. The critical value(s) associated with a p level of 0.05 for a one-tailed hypothesis test using the z statistic is (are): a. 1.96. b. 1.65. c. –1.65 or 1.65. d. –1.96 and 1.96. 83. To determine a person's percentile, first convert the person's raw score to a: a. normalized score. b. z score. c. median. d. formal score. 84. The null hypothesis states that: a. the sample being studied is different from the population from which it was drawn. b. nothing exists. c. a difference exists between the populations being studied. d. there are no differences between the populations being studied.

Page 14

Name:

Class:

Date:

Chap_07_5e 85. Imagine that the average height for all women playing in Division I basketball programs is 69 inches with a standard deviation of 3 inches. The 2010–2011 women's basketball team at the University of Connecticut, with 10 players listed on the roster, had an average height of 71.2 inches. Using the z statistic, what percent of means would fall below that for these UConn Huskies? (Round z score to two decimal places.) a. 98.98 b. 94.98 c. 86.98 d. 48.98 86. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. Treating this class as the population, what percent of students study more than 13 hours a week? (Round z score to two decimal places.) a. 15.15 b. 34.85 c. 65.15 d. 84.85 87. The phrase statistically significant means that the: a. research finding is not important. b. research finding is of practical importance. c. research result was unlikely to have occurred by chance. d. research finding is of theoretical importance. 88. The typical probability adopted by researchers to determine whether a result is extreme is: a. 0.005. b. 0.01. c. 0.05. d. 0.10. 89. Heather believes there is a positive correlation between playing violent video games and aggressive behavior. What would be her null hypothesis? a. Playing violent video games increases aggressive behavior. b. Playing violent video games is not related to level of aggressive behavior. c. Playing violent video games causes aggressive behavior. d. Only aggressive people play violent video games.

Page 15

Name:

Class:

Date:

Chap_07_5e 90. A single observation can be expressed in a number of ways all of which refer to that same observation and the exact same place within the normal curve. These expressions are: a. formal, informal, and standard. b. a z score, percentile, and standard score. c. standard, transformed, and normalized. d. a raw score, z score, and percentile. 91. Researchers in the behavioral sciences typically use alpha of 0.05. How do they represent this value and what does it mean? a. A p value, which means scores in the most extreme 2.5 percent on either end of the distribution, will be evidence to reject the null hypothesis. b. A p value, which means scores 5 percent on either side of the mean, will be evidence to reject the null hypothesis. c. A z statistic, which means scores in the most extreme 2.5 percent on either end of the distribution, will be evidence to reject the null hypothesis. d. A z statistic, which means scores 5 percent on either side of the mean, will be evidence to reject the null hypothesis. Study Description: Tail Wagging A New York Times article published on April 24, 2007, reported the research of Dr. Giorgio Vallortigara, a neuroscientist at the University of Trieste, Italy, and his two colleagues. The researchers asked whether a dog wags its tail in a preferred direction in response to positive stimuli and in another direction in response to negative stimuli. To answer their question, they recruited 30 dogs that were family pets. Filming each dog from above, they allowed it to view (through a slat in its cage) three positive stimuli separately, in order of descending positivity: its owner, an unfamiliar human, and a cat. All the dogs responded by wagging their tails to the right. But when the dogs were presented with an unfamiliar, aggressive dog, a negative stimulus, all dogs wagged their tails to the left. 92. (Study Description: Tail Wagging) Which statement is the null hypothesis for this study? a. A dog will wag its tail more to the left in response to positive stimuli and more to the right in response to negative stimuli. b. A dog's tail wagging will be the same in response to positive stimuli as to negative stimuli. c. A dog will wag its tail differently in response to positive stimuli than to negative stimuli. d. A dog's tail will wag more to the right in response to positive stimuli and more to the left in response to negative stimuli.

Page 16

Name:

Class:

Date:

Chap_07_5e 93. Penina knows the z score for her research hypothesis is –2.05, and her critical values are –1.96 and 1.96 for a p value of 0.05. What should she conclude about her hypothesis given these statistics? a. She should reject the null hypothesis and report the direction of the effect because the score is in a critical region (below –1.96). b. She should reject the null hypothesis but not report the direction of the effect because she doesn't have enough information. c. She should fail to reject the null hypothesis because the score is not in a critical region (above 1.96 or below –1.96). d. She should fail to reject the null hypothesis because she doesn't have enough information. 94. The unethical practices of p-Hacking include situations where: a. multiple dependent measures are collected, but only those that reveal significant differences are reported. b. investigators steal other researchers' work. c. multiple researchers recruit participants from various locations to replicate findings. d. investigators change their hypotheses after they have analyzed their data. 95. Inferential statistical analyses that are NOT based on a set of assumptions about the underlying population are: a. inaccurate. b. permissible only when certain assumptions about the sample are met. c. nonparametric tests. d. parametric tests. 96. In a raw score to z score transformation, an associated value on the z table provides the percentage of scores: a. between the mean and the z score. b. below the mean. c. above the mean. d. between the upper and lower limits. 97. Assume the average height for American women is 64 inches with a standard deviation of 2 inches. What percent of groups of size 25 would have mean heights of more than 64.6 inches? (Round z score to two decimal places.) a. 43.32 b. 38.21 c. 13.36 d. 6.68

Page 17

Name:

Class:

Date:

Chap_07_5e 98. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. The professor tells Brock he scored at the 76th percentile. How many hours does he study per week? (Round z score to two decimal places.) a. 11.63 hours b. 9.32 hours c. 11.84 hours d. 15.14 hours 99. When calculating a z test, one compares data from the sample to a: a. sample distribution. b. population distribution. c. distribution determined by the null hypothesis. d. distribution determined by the research hypothesis. 100. The _____ hypothesis is usually the "exciting" one. a. null b. research c. normalized d. standard 101. The research hypothesis states that: a. the sample being studied is different from the population from which it was drawn. b. nothing exists. c. the null hypothesis is not correct. d. there are no differences between the populations being studied. 102. Which statement is NOT an assumption of parametric hypothesis tests? a. The population is normally distributed. b. Participants are selected randomly from the population. c. The dependent variable is assessed using a scale measure. d. The sample size is 30 or more. 103. If the assumptions of parametric testing are not met, researchers must: a. disregard this information and proceed with interpreting results. b. make a decision to continue with parametric or nonparametric tests. c. continue with nonparametric tests. d. get a larger sample size.

Page 18

Name:

Class:

Date:

Chap_07_5e 104. The statement, "It is hypothesized that participants in a divided attention condition will perform in a significantly different manner than participants in a control group on a series of memory tasks" best illustrates a: a. one-tailed test. b. two-tailed test. c. null hypothesis. d. z test. 105. Which statement regarding the denominator of the equation for the z score and that of the z statistic is true? a. There is no difference between the denominators of the two equations. In both cases, we divide by the standard error. b. There is no difference between the denominators of the two equations. In both cases, we divide by the standard deviation. c. When computing a z score, we divide by the population standard deviation, but when computing a z statistic, we divide by the standard error of the sampling distribution. d. When computing a z statistic, we divide by the population standard deviation, but when computing a z score, we divide by the standard error of the sampling distribution. 106. The area in the tails of the comparison distribution in which the null hypothesis can be rejected is called the _____. a. critical value b. cutoff c. p level d. critical region 107. The critical value(s) associated with a p level of 0.05 for a two-tailed hypothesis test using the z statistic is (are): a. 1.96. b. 1.65. c. –1.65 and 1.65. d. –1.96 and 1.96. 108. Which of the following is NOT an example of p-Hacking? a. collecting multiple dependent measures and only reporting those measures which are associated with significant effects b. deciding to remove certain extreme or unusual scores after initial data analysis has begun c. conducting the same study with different samples in order to find the same results d. analyzing data while the study is ongoing and stopping data collection once significant results are obtained

Page 19

Name:

Class:

Date:

Chap_07_5e 109. In one psychology course, students reported studying an average of 8.62 hours a week, with a standard deviation of 4.24. The professor tells Anna she scored at the 46th percentile. How many hours does she study per week? (Round z score to two decimal places.) a. 0.10 hours b. 6.85 hours c. 7.42 hours d. 8.20 hours 110. Jan is investigating how smaller class sizes improve academic outcomes of elementary and middle school students. Her first analysis of the data fails to show any significant differences in standardized test grades between students in small versus large classes. Jan then explores her data looking for differences amongst different subgroups of students and examines different measures of academic performance. After this analytic effort, she publishes a paper reporting that small class sizes benefit the math test performance of fourth grade girls. In this example, Jan has : a. fallen prey to confirmation bias. b. engaged in p-Hacking. c. increased the likelihood of a Type II error. d. decreased the likelihood of a Type I error. Enter the appropriate word(s) to complete the statement. 111. The p level is also known as _______.

112. A z score _______ the mean will always be negative.

113. Tests that are reliable even when statistical assumptions underlying the population are not met are referred to as _______.

Page 20

Name:

Class:

Date:

Chap_07_5e 114. _______ percent of scores in a normal distribution are greater than a z score of 0.31.

115. _______ percent of scores in a normal distribution are below a z score of 0.93.

116. A z score above the mean will always be _______.

117. The _______ level is the probability used to determine the critical values, or cutoffs, in hypothesis testing.

118. The "boring" hypothesis is called the _______ hypothesis.

119. _______ percent of scores in a normal distribution are below a z score of –1.33.

120. _______ tests are statistical tests that require assumptions about the population to be met.

Page 21

Name:

Class:

Date:

Chap_07_5e 121. A z score _______ the mean will always be positive.

122. A z score at the mean will always be _______.

123. _______ percent of scores in a normal distribution are as extreme as a z score of 1.62.

124. Values of a test statistic beyond which one rejects the null hypothesis are called _______.

125. A hypothesis test in which the research hypothesis is directional is a _______ test.

126. _______ percent of scores in a normal distribution are as extreme as a z score of –2.02.

127. The "exciting" hypothesis is called the _______ hypothesis.

Page 22

Name:

Class:

Date:

Chap_07_5e 128. _______ tests are statistical tests that do not require assumptions about the population to be met.

129. The _______ is the area in the tails of the comparison distribution in which the null hypothesis can be rejected.

130. A _______-tailed test is the more common type of test conducted and is considered to be the conservative choice.

131. Changing the research hypothesis after the data has been collected is called _______.

132. A z score _______ the mean will always be 0.

133. If the data differ from what one would expect if chance was the only thing operating, the finding is called _______.

Page 23

Name:

Class:

Date:

Chap_07_5e 134. _______ percent of scores in a normal distribution are greater than a z score of –0.33.

135. A z score below the mean will always be _______.

136. A hypothesis test in which the research hypothesis specifies that there will be a difference but does not specify the direction of that difference is a _______ test.

137. For a normal distribution with a given mean less than 13, a score of 13 will result in a _______ z score than a mean of 13 based on a sample from the distribution.

138. When multiple tests are conducted in order to find a result that is statistically significant it is called _______.

139. According to the 2015 annual report of the American Psychological Association's on salaries in psychology. The average salary for those working in a teaching position was $71,471, with a standard deviation of $24,703. A group of 10 faculty members teaching at large research universities had an average salary of $112,370, with a standard deviation of $65,319. Did the average salary of these faculty members differ from that of all teachers of psychology as a whole?

Page 24

Name:

Class:

Date:

Chap_07_5e 140. It is known that in 2018 the population mean for the Evidence-Based Reading and Writing section (ERW) of the SAT is 536, with a standard deviation of 102. A sample of 400 students taking the SAT whose family income was between $60,000 and $80,000 had an ERW score of 550. (a) Perform a one-tailed hypothesis test to determine whether this group scored significantly higher on average than the population. (b) Assume that rather than a sample of 400, the sample size was only 20. Perform the same hypothesis test again with this new sample size. Does the conclusion differ from what was found in (a)?

141. The pressure to obtain statistically significant results and publish papers has led some researchers to engage in questionable statistical practices known a p-Hacking. What are some of the ways a researcher might engage in p-Hacking?

142. The average age of licensed drivers in a particular county is μ = 41.6, σ = 12, and the distribution is approximately normal. A county police officer was interested in whether the average age of drivers receiving speeding tickets differed from the average age of the driving population. She obtained a sample of N = 16 drivers receiving speeding tickets. The average age for this sample was M = 34.4. Perform the six steps of hypothesis testing necessary to determine whether this group differs from the population of drivers in the county.

143. According to the 2015 annual report of the American Psychological Association's on salaries in psychology. The average salary for those working in a teaching position was $71,471, with a standard deviation of $24,703. Assuming that these data are normally distributed, what was the salary for a faculty member in the 85th percentile? (Round z score to two decimal places.)

Page 25

Name:

Class:

Date:

Chap_07_5e 144. The critical cutoff for a one-tailed z test is either 1.65 or –1.65, depending on the direction of interest. The critical cutoffs for a two-tailed test are 1.96 and –1.96. Imagine a researcher conducts a one-tailed test and finds a significant test result at z = 1.82. (a) Explain why a two-tailed test is preferred to a one-tailed test. (b) Explain your concerns about this researcher's significant finding.

145. The Stanford Education Data Archive (SEDA) includes a wealth of data regarding education outcomes on students in school districts across the United States and includes measures of academic achievement at different grade levels. This achievement data indicates how many years ahead or behind a school district's students are performing relative the expected level of performance at that grade. Examining sixth grade achievement across more than 70,000 school districts, the students in the average district are achieving at a 6.10 grade level with a standard deviation of 1.21 grade levels. Assuming that these data are normally distributed, what was the grade level achievement of a district in the 59th percentile? (Round z score to two decimal places.)

146. According to the 2015 annual report of the American Psychological Association's on salaries in psychology. The average salary for those working in a teaching position was $71,471, with a standard deviation of $24,703. A group of 18 new faculty members had an average salary of $61,074, with a standard deviation of $10,306. Did the average salary of these new faculty members differ from that of all teachers of psychology as a whole?

147. A set of researchers interested in the health of professional hockey players randomly selected a sample of 40 hockey players from the National Hockey League and asked them to report the number of work-related injuries they incurred over the past month. Would it be appropriate for the researchers to use a parametric hypothesis test? Evaluate each of the assumptions for parametric tests to explain your answer.

Page 26

Name:

Class:

Date:

Chap_07_5e 148. It is known that in 2018 the population mean for the Evidence-Based Reading and Writing section (ERW) of the SAT is 536, with a standard deviation of 102. A sample of 100 students taking the SAT whose family income was less than $20,000 had an ERW score of 480. (a) Perform a two-tailed hypothesis test to determine whether the group whose family income was less than $20,000 had a significantly different score on average than the population. (b) Assume that rather than a sample of 100, the sample size was only 10. Perform the same hypothesis test again with this new sample size. Does the conclusion differ from what was found in (a)?

149. The average age of licensed drivers in a particular county is μ = 41.6, σ = 12, and the distribution is approximately normal. A county police officer was interested in whether the average age of drivers receiving parking tickets differed from the average age of the driving population. She obtained a sample of N = 25 drivers receiving parking tickets. The average age of these drivers was M = 40.2. Perform the six steps of hypothesis testing necessary to determine whether this group differs from the population of drivers in the county.

150. The Stanford Education Data Archive (SEDA) includes a wealth of data regarding education outcomes on students in school districts across the United States and includes measures of academic achievement at different grade levels. This achievement data indicates how many years ahead or behind a school district's students are performing relative the expected level of performance at that grade. Examining sixth grade achievement across more than 70,000 school districts, the students in the average district are achieving at a 6.10 grade level with a standard deviation of 1.21 grade levels. Assuming that these data are normally distributed, what was the grade level achievement of a district in the 33rd percentile? (Round z score to two decimal places.)

Page 27

Name:

Class:

Date:

Chap_07_5e 151. Researchers collected data on the ages of people who died due to cardiac arrest over a 1-year period. The distribution of the ages is normal, and the mean age was 65, with a standard deviation of 5. (a) What percentage of people who died were between the ages of 61 and 71? (b) What percentage of people who died were older than 71?

152. Cromley and Azevedo (2007) were interested in the reading-comprehension performance of ninth graders. Students' reading comprehension was assessed with a test in which population mean performance is 33, with a standard deviation of 10. (a) One of the researchers' goals was to identify the bottom 30 percent of students so that they could determine the characteristics of students who struggled with reading. What cutoff score would Cromley and Azevedo use to identify the bottom 30 percent of readers? (b) Let's say we want to identify the best 10 percent of readers to put into an advanced class. What cutoff score on the readingcomprehension test would we use? (Round z score to two decimal places.)

153. Data from 2018 finds that the average age at which men get married in California is normally distributed with a mean of μ = 30.3, σ = 2.5. (a) What percentage of the married men in California were married between the ages of 31 and 35? (b) What percentage of the married men in California were married before they were 25?

154. A New York Times article published on April 24, 2007, reported the research of Dr. Giorgio Vallortigara, a neuroscientist at the University of Trieste, Italy, and his colleagues. In this study, the researchers investigated whether a dog's tail wags in a preferred direction in response to positive as opposed to negative stimuli. The researchers answered this question by recruiting 30 dogs that were family pets. Filming the dog's tail from above, they allowed each dog to view (through a slat in its cage) its owner, an unfamiliar human, a cat (positive stimuli), and an unfamiliar, dominant dog (a negative stimulus). Would it be appropriate for the researchers to use a parametric hypothesis test? Evaluate each of the assumptions for parametric tests to explain your answer.

Page 28

Name:

Class:

Date:

Chap_07_5e Answer Key 1. False 2. True 3. False 4. False 5. True 6. True 7. True 8. False 9. True 10. False 11. True 12. True 13. False 14. True 15. False 16. True 17. True 18. False 19. True 20. True 21. True 22. False 23. True 24. False 25. False 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_07_5e 27. True 28. False 29. True 30. False 31. False 32. False 33. False 34. d 35. a 36. b 37. b 38. a 39. c 40. b 41. a 42. b 43. a 44. a 45. d 46. c 47. b 48. b 49. a 50. c 51. d 52. b 53. b 54. a Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_07_5e 55. d 56. c 57. c 58. d 59. b 60. c 61. b 62. d 63. d 64. a 65. c 66. d 67. a 68. b 69. b 70. b 71. a 72. b 73. b 74. d 75. c 76. b 77. b 78. c 79. c 80. c 81. d 82. c Copyright Macmillan Learning. Powered by Cognero.

Page 31

Name:

Class:

Date:

Chap_07_5e 83. b 84. d 85. a 86. a 87. c 88. c 89. b 90. d 91. a 92. b 93. a 94. a 95. c 96. a 97. d 98. a 99. c 100. b 101. a 102. d 103. b 104. b 105. c 106. d 107. d 108. c 109. d 110. b Copyright Macmillan Learning. Powered by Cognero.

Page 32

Name:

Class:

Date:

Chap_07_5e 111. alpha 112. below 113. robust 114. 37.83 115. 82.38 116. positive 117. p 118. null 119. 9.18 120. Parametric 121. above 122. zero, 0 123. 10.52 124. critical values, or cutoffs 125. one-tailed 126. 4.34 127. research 128. Nonparametric 129. critical region 130. two 131. HARKing 132. at, equal to 133. statistically significant 134. 62.93 135. negative 136. two-tailed 137. smaller

Page 33

Name:

Class:

Date:

Chap_07_5e 138. p-Hacking 139. The calculated z statistic, z = 5.24, exceeds the cutoff values of –1.96 or 1.96 (two-tailed hypothesis test, using a p level of 0.05). Therefore, we reject the null hypothesis. We have evidence that the average salary for faculty teaching at large research universities differs from that of all teachers of psychology in 2015. 140. (a) The calculated z statistic, z = 2.75, exceeds the the cutoff value of 1.65 (one-tailed hypothesis test, using a p level of 0.05). Therefore, we reject the null hypothesis. The ERW SAT score of students with a family income between $60,000 and $80,000 was significantly greater than that of all students who took the SAT. (b) If the sample size is only 20, the standard error increases from 5.10 to 22.81. If we use that sample size of 20 to calculate the z statistic, the resulting z = 0.61, and the critical value of 1.65 is not exceeded. We would fail to reject the null hypothesis. This conclusion differs from that reached when the sample size was 400. One of the reasons is that with a large sample size, the spread of the sampling distribution of the mean is very small, and it is thus easier to obtain an extreme test statistic. 141. There are a number of ways in which a researcher might engage in p-Hacking including: analyzing the data as it is collected and stopping data collection once a significant effect is obtained; removing certain scores, such as outliers, after the data has been analyzed; and collecting data on multiple dependent measures and only reporting the results for the variables where significant differences emerged. 142. Step 1: The populations to be compared are (1) all drivers in the county and (2) drivers receiving speeding tickets. The distribution is the sampling distribution. The first two assumptions of the z test are met: The dependent variable, age, is assessed on a scale measure. The entire population has been identified, and the sample has been randomly selected from it. It is difficult to know whether the assumption that the population is normally distributed is met. We do not know the distribution and the sample size is less than 30. Step 2: Null hypothesis: The average age of drivers receiving speeding tickets is not different from that of the population of drivers: H0: µ1 = µ2. Research hypothesis: The average age of drivers receiving speeding tickets differs from that of the population of drivers: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a z distribution, with a mean of 41.6 and a standard deviation of 12. Step 4: The critical values, using a p level of 0.05 and a two-tailed test, are –1.96 and 1.96. Step 5: z = –2.40. Step 6: The calculated test statistic exceeds the critical value. Therefore, we reject the null hypothesis. On average, the age of drivers receiving speeding tickets is significantly lower than the age of all drivers in the county. 143. x = $71,471 + 1.04($24,703) = $71,471 + $25,691.12 = $97,162.12 144. (a) A two-tailed test allows a researcher to detect a difference in two directions, an increase or a decrease, for example. A one-tailed test only allows the researcher to look for differences in the expected direction. If an unexpected result occurs in the opposite direction, we cannot switch the direction of the hypothesis after the fact, thus causing us to miss the findings. (b) This researcher's z statistic falls between the critical cutoffs for a onetailed and two-tailed test. As a one-tailed test, this result is significant, but if run as a two-tailed test, the result is no longer significant. This seems like a result that is not very strong, and it could represent a Type I error. 145. x = 6.10 + .23(1.21) = 6.10 + .2783 = 6.38

Page 34

Name:

Class:

Date:

Chap_07_5e 146. The calculated z statistic, z = –1.79, does not exceed the cutoff values of –1.96 or 1.96 (two-tailed hypothesis test, using a p level of 0.05). Therefore, we fail to reject the null hypothesis. We do not have evidence that the average salary for the newly hired faculty members differed from that of all teachers of psychology in 2015. 147. It would be appropriate for the researchers to perform a parametric test. The assumption that the dependent variable is assessed using a scale measure is met. Having players record the number of times they have workrelated injuries is actually a discrete, ratio variable, so it definitely meets the assumption of equal intervals. Given that the entire population can be identified (the National Hockey League), and given that the researchers said they randomly selected from this population, the assumption of random selection has been met. It is difficult to know whether the distribution of injuries in the population is normal, but given that the researchers have a sample size greater than 30, the sampling distribution of the mean is probably normal. 148. (a) The calculated z statistic, z = –5.49, exceeds the cutoff value of –1.96. Therefore, we reject the null hypothesis. The ERW SAT score of students with a family income of less than $20,000 was significantly lower than that of all students who took the SAT. (b) If the sample size is only 10, the standard error changes from 10.20 to 32.26. If we use the sample size of 10 to calculate the z statistic, z = –1.74, and the critical value of – 1.96 is not exceeded, we would fail to reject the null hypothesis. This conclusion differs from that reached when the sample size was 100. One of the reasons is that with a large sample size, the spread of the sampling distribution of the mean is very small, and it is thus easier to obtain an extreme test statistic. 149. Step 1: The populations to be compared are (1) all drivers in the county and (2) drivers receiving parking tickets. The distribution is the sampling distribution. The first two assumptions of the z test are met: the dependent variable, age, is assessed on a scale measure. The entire population has been identified, and the sample has been randomly selected from it. It is difficult to know whether the assumption that the population is normally distributed is met. We do not know the distribution and the sample size is less than 30. Step 2: Null hypothesis: The average age of drivers who receive parking tickets is not different from that of the population of drivers: H0: µ1 = µ2. Research hypothesis: The average age of drivers who receive parking tickets differs from the average age of the population of drivers: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a z distribution, with a mean of 41.6 and a standard deviation of 12. Step 4: The critical values, using a p level of 0.05 and a two-tailed test, are –1.96 and 1.96. Step 5: z = –0.58. Step 6: The calculated test statistic does not exceed the critical value. Therefore, we fail to reject the null hypothesis. We have no evidence that the average age of drivers receiving parking tickets is different from the average age of drivers in the population. 150. x = 6.10 – .44(1.21) = 6.10 – .5324 = 5.57 151. (a) 67.30 percent; (b) 11.51 percent. 152. (a) Students receiving a score of 27.80 or lower would be considered the bottom 30 percent. (b) Students receiving a score of 45.80 or higher would be considered the top 10 percent. 153. (a) 35.96 percent; (b) 1.70 percent.

Page 35

Name:

Class:

Date:

Chap_07_5e 154. From the study description, it would not be appropriate for Dr. Vallortigara and his colleagues to use a parametric test. First, the assumption that the dependent variable is assessed using a scale measure is violated. The dependent variable is the direction in which the dog wags its tail. This is a nominal variable. Second, it is highly unlikely that the sample of dogs was randomly selected from the population of all dogs that are family pets, as it would be impossible to identify the entire population to begin with. It is unclear what kind of distribution to expect with tail wagging, but the sampling distribution of the mean may be normally distributed, given that there are 30 dogs in the sample.

Page 36

Name:

Class:

Date:

Chap_08_5e Indicate whether the statement is true or false. 1. Two-tailed tests have more statistical power than one-tailed tests. a. True b. False 2. Meta-analysis includes an assessment of null effects, often called a file drawer analysis. a. True b. False 3. The amount of overlap between two distributions can be decreased if their means are closer together. a. True b. False 4. An interval estimate is based on a population parameter and provides a range of plausible values for the sample statistic. a. True b. False 5. As alpha increases, the likelihood of H0 being true also increases. a. True b. False 6. A statistically significant difference between two groups indicates that all or most participants in one group are different from all or most of the participants in the other group. a. True b. False 7. If the population mean falls within the confidence interval, it is not plausible that the sample comes from the null hypothesized population. a. True b. False 8. Increasing the number of scores in a sample from 16 to 36 decreases the width of a confidence interval to estimate the population mean. (Assume that all other factors are held constant.) a. True b. False 9. If the population mean falls within the 95% confidence interval, you should fail to reject the null hypothesis. a. True b. False

Page 1

Name:

Class:

Date:

Chap_08_5e 10. The amount of overlap between two distributions can be decreased if their means are further apart. a. True b. False 11. An interval estimate offers an advantage over a point estimate because it estimates the population parameter using a range of values rather than a single number. a. True b. False 12. If a real difference exists between a sample and the comparison distribution, that difference is necessarily important and meaningful. a. True b. False 13. As Cohen's d increases from 0.2 to 0.8, the overlap between distributions decreases. a. True b. False 14. Effect size indicates the size of the difference between the sample and the null hypothesized value, which is unaffected by sample size. a. True b. False 15. If the population mean falls outside the 95% confidence interval, you should fail to reject the null hypothesis. a. True b. False 16. Statistical power is the probability that a researcher will not make a Type II error. a. True b. False 17. The sample mean is centered at the middle of the confidence interval. a. True b. False 18. Statistical power is a measure of the ability to reject the null hypothesis when it is false. a. True b. False 19. Statistical power is a measure of the ability to retain the null hypothesis when it is true. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 2

Name:

Class:

Date:

Chap_08_5e 20. Exaggerating the levels of an independent variable will increase statistical power. a. True b. False 21. Statistical power calculations help researchers know how many participants are needed in their study. a. True b. False 22. An interval estimate is based on a sample statistic and provides a range of plausible values for the population parameter. a. True b. False 23. If the population mean falls within the 95% confidence interval, you should reject the null hypothesis. a. True b. False 24. As Cohen's d increases from 0.2 to 0.8, the overlap between distributions increases. a. True b. False 25. The amount of overlap between two distributions can be decreased if the amount of variation within each population is reduced. a. True b. False 26. If the population mean falls within the confidence interval, it is plausible that the sample comes from the null hypothesized population. a. True b. False 27. As sample size increases, the standard error decreases, resulting in a larger test statistic. a. True b. False 28. As sample size increases, the standard error decreases, resulting in a smaller test statistic. a. True b. False 29. As N increases, the width of the confidence interval decreases. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_08_5e 30. As N increases, the width of the confidence interval increases. a. True b. False 31. Both effect size and the value of the test statistic are unaffected by changes in the sample size (holding other factors constant). a. True b. False 32. Increasing the number of scores in a sample from 16 to 36 increases the width of a confidence interval to estimate the population mean. (Assume that all other factors are held constant.) a. True b. False 33. A point estimate offers an advantage over an interval estimate because it estimates the population parameter using a range of values rather than a single number. a. True b. False 34. Increasing N to increase statistical power is considered a suspect practice by statisticians. a. True b. False 35. The population mean is centered at the middle of the confidence interval. a. True b. False 36. Effect size indicates the size of the difference between the population and the null hypothesized value. a. True b. False 37. Cohen's d indicates the difference between two means in terms of variance, not standard error. a. True b. False

Page 4

Name:

Class:

Date:

Chap_08_5e Indicate the answer choice that best completes the statement or answers the question. 38. If the expected direction of an effect is correct, then using a one-tailed hypothesis test instead of a two-tailed hypothesis test: a. increases power. b. decreases power. c. makes power 1.0. d. makes power 0. 39. Which of the following does NOT increase statistical power? a. increasing alpha b. using a one-tailed hypothesis c. increasing sample size d. reducing differences between levels of the independent variable 40. The minimum acceptable level of estimated power for a study is: a. 0.95. b. 0.80. c. 0.65. d. 0.50. 41. Statistical power is: a. the strength of the research study. b. the ability to find important results. c. a combination of the distance between group means and distribution variability. d. the percentage of the comparison distribution that falls beyond the critical cutoff. 42. The larger the effect size is, the: a. smaller the test statistic is. b. smaller the sample size is. c. more two distributions overlap. d. less two distributions overlap. 43. When considering the results from an opinion poll, where several verbal expressions are rated for their level of annoyance, what is particularly useful about margins of error is that: a. they tell us how accurate our data are. b. they directly pinpoint a single value to estimate the data. c. the accuracy of the point estimate increases with the addition of the margin of error. d. we can figure out more than one interval estimate for the same poll to see if they overlap.

Page 5

Name:

Class:

Date:

Chap_08_5e 44. Statistical power is calculated as 0.93. This means that if the null hypothesis is _____, there is a _____ chance of rejecting the null hypothesis. a. false; 93% b. false; 7% c. true; 93% d. true; 7% 45. The ability to reject the null hypothesis given that the null hypothesis is false is: a. a Type II error. b. statistical power. c. a false alarm. d. a Type I error. 46. Which of the following does NOT increase statistical power? a. increasing alpha b. using a two-tailed hypothesis c. increasing sample size d. exaggerating differences between levels of the independent variable 47. It is known that the population mean for the Evidence-Based Reading and Writing (ERW) section of the SAT is 536, with a standard deviation of 100. In 2018, a sample of 400 students whose family income was between $60,000 and $80,000 had an average ERW SAT score of 550. The point estimate of the mean for this group is _____ and the 95% confidence interval for this group is _____. a. 536; [540.2, 559.8] b. 536; [526.2, 545.8] c. 550; [540.2, 559.8] d. 550; [526.2, 545.8]

Page 6

Name:

Class:

Date:

Chap_08_5e 48. A researcher interested in the effects of humor on memory randomly assigns 18 participants to either the humor group or the no-humor group. The humor group reads humorous sentences and the no-humor group reads non-humorous sentences. On a later memory test, the researcher finds that in terms of the direction of the means, the humor group had better memory than the no-humor group, but this effect was not significant (p = 0.06). What should this researcher do? a. She can attempt to increase her statistical power by using a two-tailed hypothesis test rather than a one-tailed hypothesis test. b. She can abandon the study of humor on memory because, given her results, it is obvious that humor has no effect on memory. c. Because the effect looks as though it is barely missing significance, she can just treat it as though the effect exists and communicate this exciting effect to her colleagues. d. Because the effect looks as though it is barely missing significance and her sample size is fairly small, she can increase her sample size to increase her statistical power to detect the effect. 49. A high degree of overlap between two distributions of approximately 95% is likely to result in a(n) _____ effect size. a. small b. medium c. large d. unconventional 50. An overlap between two distributions of approximately 55% is likely to result in a(n) _____ effect size. a. small b. medium c. large d. unconventional 51. Which of the following does NOT increase statistical power? a. increasing alpha b. using a one-tailed hypothesis c. decreasing sample size d. exaggerating differences between levels of the independent variable

Page 7

Name:

Class:

Date:

Chap_08_5e 52. A behavioral neuroscientist is testing the effects of adrenaline on memory using a group of 16 rats. The researcher is unsure about how much adrenaline might produce an effect on memory. One group of rats will be injected with placebo saline (0 micrograms of adrenaline); the other group will be injected with a dose of adrenaline. When deciding between a 3-microgram dose or an 6-microgram dose (both of which are safe), the researcher opts to use the 6-microgram dose. The researcher has: a. made a Type II error. b. exaggerated the difference between the levels of the independent variable, thereby increasing statistical power. c. exaggerated the difference between the levels of the independent variable, thereby decreasing statistical power. d. given the rats an overdose of adrenaline. 53. According to Cohen's convention, a d value of _____ is a medium effect size. a. 0.2 b. 0.5 c. 0.8 d. 1.2 54. An overlap between two distributions of approximately 39% is likely to result in a(n) _____ effect size. a. small b. medium c. large d. unconventional 55. If there is less than a(n) _____% chance of rejecting the null hypothesis when it is false, there is insufficient power. a. 50 b. 60 c. 80 d. 90 56. To remove the adjustment for the influence of sample size, Cohen's d uses the _____ rather than the _____ as part of its formula. a. standard error; standard deviation b. standard deviation; standard error c. raw scores; standard scores d. variance; raw scores

Page 8

Name:

Class:

Date:

Chap_08_5e 57. A researcher performs a meta-analysis and finds that the mean d = 0.11, and the 95% confidence interval around this mean is [–0.04, 0.26]. What could the researcher conclude? a. All future studies of this effect will find effect sizes somewhere between 0.34 and 0.56. b. Averaging across all of the literature, there really is no effect. c. There is a strong effect, but the direction of the effect is unclear. d. Averaging across all of the literature, there is a strong effect, and this effect is statistically significant. 58. Before hypothesis testing and at the beginning of a study, a researcher is advised to conduct _____ because it _____. a. a power analysis; tells the researcher the number of participants needed for trustworthy results b. an effect size estimate; tells the researcher the number of participants needed for trustworthy results c. a statistical significance; determines how meaningful the study results will be d. alpha testing; determines how meaningful the study results will be 59. Effect sizes are affected by _____ and _____. a. large standard deviations; large standard errors b. standard deviations; variability of population distributions c. standard errors; variability of population distributions d. mean differences; variability of population distributions 60. Alpha refers to: a. statistical power. b. the probability of making a Type II error. c. the probability of making a Type I error. d. effect size. 61. Which of the following does NOT increase statistical power? a. decreasing alpha b. using a one-tailed hypothesis c. increasing sample size d. exaggerating differences between levels of the independent variable 62. It becomes progressively easier to declare statistical significance as the _____ increases. a. standard error b. value of the critical cutoff c. sample size d. number of items on the instrument

Page 9

Name:

Class:

Date:

Chap_08_5e 63. When alpha increases, both _____ and _____ increase. a. standard error; power b. power; probability of a Type I error c. power; probability of a Type II error d. probability of a Type I error; probability of a Type II error 64. Cohen's d is: a. a method for calculating confidence intervals for the z test. b. the difference between the sample means divided by the standard error. c. the standardized difference between group means. d. a measure of statistical power. 65. Following a meta-analysis, the researcher might decide to perform a(n) _____ to determine the number of null results that would have to exist to overturn any statistically significant effect found in the meta-analysis. a. power analysis b. file drawer analysis c. effect-size analysis d. hypothesis test 66. One of the roles of the researcher performing a meta-analysis is to: a. determine how many studies were never published and find those studies. b. throw out statistical outliers from the analysis. c. decide on the criteria for the inclusion of studies in the analysis. d. convince the reader of the existence of the effect of interest. 67. The range of raw scores contained in an 90% confidence interval will be _____ the range of raw scores contained in a 95% confidence interval. a. larger than b. smaller than c. the same size as d. smaller than or the same size as 68. As sample size increases, the: a. population mean increases. b. standard error increases. c. size of the test statistic increases. d. size of the test statistic decreases.

Page 10

Name:

Class:

Date:

Chap_08_5e 69. What falls within the 95% confidence interval? a. all the means we would expect to obtain 5% of the time when repeatedly sampling from a population b. all the means we would expect to obtain 95% of the time when repeatedly sampling from a population c. the true population mean d. the null hypothesis population mean 70. The practical use of statistical power is that it informs researchers: a. whether they will find significant results. b. how many participants are needed to conduct a study with findings they can trust. c. whether they will find important results. d. what effect size they can expect to find in conducting their study. 71. If the sample mean is 57.2, with an upper limit to the confidence interval of 62.74, what is the lower limit? a. 5.54 b. 51.66 c. 53.46 d. 57.20 72. The statement "The findings based on a sample of 200 participants were statistically significant, providing evidence for the research hypothesis" would be strengthened by: a. using convenience sampling. b. hypothesis testing. c. measuring effect sizes. d. sampling college students. 73. According to a "how to stop bullying" Web site, 15% of students reported experiencing bullying one to three times within the most recent month. Assume the standard deviation is 4.5% of students. Robert collects data from 150 students at a medium-sized school in Iowa and finds that only 12% reported this rate of bullying. What is his 95% confidence interval? a. [7.5, 16.5] b. [11.28, 12.72] c. [11.39, 12.61] d. [14.28, 15.72] 74. Before conducting a power analysis, a researcher should know the desired: a. alpha level. b. gender distribution. c. standard error. d. variance.

Page 11

Name:

Class:

Date:

Chap_08_5e 75. The statistical convention for the minimal acceptable power is: a. 0.95. b. 0.90. c. 0.80. d. 0.75. 76. If an effect is significant but the effect size for the difference between the two means is small (according to Cohen's conventions), about how much overlap will there be between the two distributions? a. 99% b. 85% c. 50% d. 15% 77. Imagine that a study of memory and aging finds that younger participants correctly recall 58% of studied words and older participants correctly recall 41% of studied words; the size of this effect is Cohen's d = 0.48. According to Cohen's conventions for interpreting d, this effect is: a. small. b. medium. c. large. d. so small as to be considered virtually no effect. 78. Using a one-tailed test allows us to increase statistical power over a two-tailed test because it: a. divides the alpha into two tails but focuses on only one of those tails. b. divides the alpha into two tails, which reduces the size of each. c. puts the entire alpha into one tail, increasing the chances of rejecting the null hypothesis. d. puts the entire alpha into one tail, decreasing the chances of rejecting the null hypothesis. 79. When Cohen's d is large (based on Cohen's conventions), the amount of overlap between the two distributions being compared is _____%. a. 75 b. 53 c. 15 d. 0 80. If the mean difference between levels of the independent variable is exaggerated, statistical power: a. increases. b. decreases. c. stays the same. d. gets closer to 0.

Page 12

Name:

Class:

Date:

Chap_08_5e 81. Assume for a given study that the null hypothesis asserts the expected value of a phenomenon is 50. A research study results in a 95% confidence interval reported as [48.76, 55.24]. What decision should be made based on this confidence interval? a. Reject the null hypothesis. b. Fail to reject the null hypothesis. c. Retain the null hypothesis. d. Perform a hypothesis test before making a decision. 82. According to Cohen's convention, a d value of _____ is a small effect size. a. 0.2 b. 0.5 c. 0.8 d. 1.2 83. An overlap between two distributions of approximately 99% is likely to result in a(n) _____ effect size. a. small b. medium c. large d. unconventional 84. Recent research published by Frumin and colleagues (2011) in the journal Science addresses whether females' tears have an effect on males. Imagine that exposure to tears lowered self-rated sexual arousal by 1.37 points, with a margin of error of 0.22 point. The interval estimate is: a. 1.37. b. ±0.22. c. [1.15, 1.59]. d. 0.22. 85. Increasing sample size: a. decreases the likelihood that we will reject the null hypothesis. b. increases the likelihood that we will reject the null hypothesis. c. has no effect on the likelihood that we will reject the null hypothesis. d. makes it more likely that we will make a Type II error. 86. Statistical power is a measure of the ability to reject the null hypothesis when: a. it is true. b. it is false. c. there are no significant differences. d. the sample size cannot be increased.

Page 13

Name:

Class:

Date:

Chap_08_5e 87. Why are effect sizes rather than test statistics used when comparing study results? a. Effect sizes, unlike test statistics, are not affected by sample size and thus ensure a fair comparison. b. It is easier to average effect size than it is to average test statistics. c. Effect sizes are based on standard error, while test statistics are based on standard deviation. d. Effect sizes, unlike test statistics, account for sample size, thereby ensuring an accurate comparison. 88. As sample size increases, the: a. standard error decreases. b. test statistic decreases. c. standard error increases. d. standard deviation increases. 89. Cohen's d is one measure of: a. statistical significance. b. effect size. c. clinical significance. d. sample characteristics. 90. According to Cohen's convention, a d value of _____ is a large effect size. a. 0.2 b. 0.5 c. 0.8 d. 1.2 91. A confidence interval is a(n) _____ that includes the population mean after repeatedly sampling. a. point estimate b. interval estimate c. probability d. hypothesis 92. A researcher performs a meta-analysis and finds that the mean d = 0.45, and the 95% confidence interval around this mean is [0.34, 0.56]. What can the researcher conclude? a. All future studies of this effect will find effect sizes somewhere between 0.34 and 0.56. b. Averaging across all of the literature, there really is no effect. c. There is a medium effect, but the direction of the effect is unclear. d. Averaging across all of the literature, there is a medium effect, and this effect is statistically significant.

Page 14

Name:

Class:

Date:

Chap_08_5e 93. According to the textbook, a(n) _____ test has more statistical power; however, a(n) _____ is more conservative. a. one-tailed; two-tailed b. two-tailed; one-tailed c. one-tailed; effect-size d. effect-size; hypothesis 94. Measures of effect size: a. are unaffected by sample size. b. increase as sample size increases. c. decrease as the difference between population means increases. d. do not rely on sample means. 95. A degree of overlap between two distributions of approximately 50% is likely to result in a(n) _____ effect size. a. small b. medium c. large d. unconventional 96. Meta-analysis involves: a. finding all studies published on a topic, contacting the authors of the studies to request their original data, and then analyzing all the obtained data in one large analysis of variance. b. finding all studies published on a topic, calculating the effect size for each of those studies, and averaging the effect sizes together to find the average size of the effect across all studies. c. averaging all the test statistics from every possible study on a given topic. d. attempting to recreate the experimental conditions of every published study on a given topic. 97. Increasing sample size does NOT: a. increase statistical power. b. decrease standard error. c. increase the magnitude of the test statistic. d. decrease statistical power.

Page 15

Name:

Class:

Date:

Chap_08_5e 98. Recent research published by Frumin and colleagues (2011) in the journal Science addresses whether females' tears have an effect on males. Imagine that exposure to tears lowered self-rated sexual arousal by 1.37 points, with a margin of error of 0.22 point. The point estimate is _____, while the interval estimate is _____. a. 0.32; [1.15, 1.59] b. 1.37; [1.15, 1.59] c. 0.32; [–1.15, 1.59] d. 1.37; [–1.15, 1.59] 99. Effect size assesses the degree to which two: a. populations overlap. b. populations do not overlap. c. samples overlap. d. samples do not overlap. 100. An article in the journal Applied Nutritional Investigation reported the results of a comparison between a low-calorie soy-protein diet and a low-calorie traditional-protein diet (Liao, 2007). Twelve obese participants were randomly assigned to each diet. At the end of the diet period, those on the soy diet lost an average of 2.3% of their body fat (SD = 0.55), while those on the traditional diet lost an average of 1.22% of their body fat (SD = 0.50). If the sample size of this study is increased, the value of the test statistic would _____ and the effect size would _____. a. increase; remain the same b. decrease; remain the same c. decrease; increase d. increase; decrease 101. Mehl et al. (2007) published a study in the journal Science reporting the results of an extensive study of 396 men and women comparing the number of words uttered per day by each sex. They found that, on average, women uttered 16,215 words a day and men uttered 15,669 words a day. The effect size calculated on the basis of these findings is Cohen's d = 0.07. According to Cohen's conventions for interpreting d, this effect is: a. small. b. medium. c. large. d. so small as to be considered virtually no effect.

Page 16

Name:

Class:

Date:

Chap_08_5e 102. Effect sizes rely on comparison of a distribution of _____ rather than on a distribution of _____ and, therefore, are unaffected by sample size. a. means; scores b. scores; means c. errors; residuals d. residuals; errors 103. Assume for a given study that the null hypothesis asserts the expected value of a phenomenon is 15. A research study results in a 95% confidence interval reported as [12.14, 14.86]. What decision should be made based on this confidence interval? a. Reject the null hypothesis. b. Fail to reject the null hypothesis. c. Retain the null hypothesis. d. Perform a hypothesis test before making a decision. 104. Imagine that a study of memory and aging finds that younger participants correctly recall 58% of studied words and older participants correctly recall 41% of studied words; the size of this effect is Cohen's d = 0.48. This effect size indicates that the memory performance of: a. older participants is approximately half a standard deviation above that of younger participants. b. older participants is approximately half a standard deviation below that of younger participants. c. younger participants is approximately half a standard deviation below that of older participants. d. younger participants is significantly lower than that of older participants. 105. Confidence in a point estimate _____, whereas confidence in an interval estimate _____. a. is very high; is very low b. is based on the alpha level used; is based on the margin of error calculated c. cannot be articulated; is directly related to the size of the interval constructed d. is very low; is very high 106. If an effect is significant but the effect size for the difference between the two means is medium (according to Cohen's conventions), about how much overlap will there be between the two distributions? a. 99% b. 85% c. 67% d. 53%

Page 17

Name:

Class:

Date:

Chap_08_5e 107. Hypothesis testing tells: a. what the size of an effect is. b. which group differences are of practical importance. c. whether two distributions overlap at all. d. what results are significant, but no details regarding significant results. 108. Mehl et al. (2007) published a study in the journal Science reporting the results of an extensive study of 396 men and women comparing the number of words uttered per day by each sex. They found that, on average, women uttered 16,215 words a day and men uttered 15,669 words a day. The effect size calculated on the basis of these findings is Cohen's d = 0.07. This effect size indicates that the: a. number of words uttered by the men and the women differed significantly. b. women uttered a significantly greater number of words in a day than did the men in this study. c. means of the men and women overlap by only 7%. d. means of the men and women are not even one-tenth of 1 standard deviation apart. 109. Martha is a cognitive psychologist who is studying reading times for stories that contain either consistent or inconsistent information. She runs 32 people through her study and concludes that reading times slow when coherence breaks occur in a story. Specifically, she concludes reading times slow by 8.7 milliseconds on average. Martha's prediction is a(n): a. interval estimate. b. standard deviation. c. point estimate. d. sigma score. 110. Recent research published by Frumin and colleagues (2011) in the journal Science addresses whether females' tears have an effect on males. Imagine that exposure to tears lowered self-rated sexual arousal by 1.37 points, with a margin of error of 0.22 point. The point estimate is: a. 1.37. b. 1.37 ± 0.22. c. [1.15, 1.59]. d. 0.22. 111. The sample mean is _____ the confidence interval. a. at the beginning of b. at the end of c. in the center of d. excluded from

Page 18

Name:

Class:

Date:

Chap_08_5e 112. The formula for Cohen's d substitutes the _____ symbol for the _____ symbol used in the denominator of the z statistic formula. a. σM; µ b. σ; σM c. µM; µ d. µ; σM 113. As sample size increases, the test statistic increases because the: a. difference between the means increases. b. overlap between distributions increases. c. distance between distributions decreases. d. standard error decreases. 114. What falls within a 90% confidence interval? a. all the means we would expect to obtain 10% of the time when repeatedly sampling from a population b. all the means we would expect to obtain 90% of the time when repeatedly sampling from a population c. the true population mean d. the null hypothesis population mean Enter the appropriate word(s) to complete the statement. 115. A(n) ____-tailed test has more statistical power than a(n) _____-tailed test.

116. A sample statistic using just one number to estimate a population parameter is a(n) _______ estimate.

117. Increasing sample size leads to an increase in the test statistic because of its impact on the _______.

Page 19

Name:

Class:

Date:

Chap_08_5e 118. Cohen's d measures the difference between means in terms of _______.

119. It becomes progressively easier to declare statistical significance as the _______ is increased.

120. A larger effect size indicates that there is _______ overlap of two distributions.

121. A confidence interval listed as [39.54, 64.46] is centered at a mean of _______.

122. As alpha increases, statistical power _______.

123. As standard deviation decreases, statistical power _______.

124. To enhance the possibility of finding significant differences in the data, researchers may consider exaggerating the levels of the _______ variable.

Page 20

Name:

Class:

Date:

Chap_08_5e 125. The benefit of calculating _______ is that researchers know how many participants or observations they need to conduct an adequate study.

126. According to Cohen's conventions, a d value of _______ indicates a large effect size.

127. As sample size decreases, statistical power _______.

128. The most practical way to increase statistical power is to increase _______ in a study.

129. According to Cohen's conventions, a d value of _______ indicates a medium effect size.

130. The confidence interval is centered at the mean of the _______.

131. A smaller effect size indicates that there is _______ overlap of two distributions.

Page 21

Name:

Class:

Date:

Chap_08_5e 132. A study that calculates the mean effect size from the individual effect sizes of many studies is called a(n) _______.

133. Statistical _______ refers to the probability of successfully rejecting the null hypothesis.

134. A range of sample statistics used to estimate a population parameter is a(n) _______.

135. As alpha decreases, statistical power _______.

136. As sample size increases, statistical power _______.

137. Statistical power _______ when the mean difference between levels of the independent variable is exaggerated.

138. As standard deviation increases, statistical power _______.

Page 22

Name:

Class:

Date:

Chap_08_5e 139. To compute a confidence interval when the population mean and standard deviation are known, use a(n) _______ distribution.

140. Following a meta-analysis, the researcher might decide to perform a(n) _______ to determine the number of null results that would have to exist to overturn any statistically significant effect found in the meta-analysis.

141. A two-tailed test has _______ statistical power than a one-tailed test.

142. According to Cohen's conventions, a d value of _______ indicates a small effect size.

143. Statistical power refers to the probability of successfully rejecting the _______.

144. The average age for licensed drivers in a county is µ = 41.6, σ = 12, and the distribution is approximately normal. A county police officer was interested in whether the average age of drivers receiving speeding tickets differed from the average age of the driving population. She obtained a sample of N = 16 drivers with speeding tickets. The average age for this sample was M = 34.4 and alpha = 0.05. (a) Calculate the effect size for this study. (b) Using the effect size, calculate statistical power for this study. (c) Does this statistical power meet the minimum criterion for statistical power to detect an effect?

Page 23

Name:

Class:

Date:

Chap_08_5e 145. The average age for licensed drivers in a county is µ = 41.6, σ = 12, and the distribution is approximately normal. A county police officer was interested in whether the average age of drivers receiving speeding tickets differed from the average age of the driving population. She obtained a sample of N = 24 drivers receiving speeding tickets. The average age for this sample was M = 32.6. (a) Calculate the 95% confidence interval for this sample mean. (b) Based on this confidence interval, make a decision to accept or reject the null hypothesis.

146. An article in the journal Applied Nutritional Investigation reported the results of a comparison of a lowcalorie soy-protein diet and a low-calorie traditional-protein diet (Liao, 2007). Twelve obese participants were randomly assigned to each diet. At the end of the diet period, those on the soy diet lost on average 2.3% of their body fat (SD = 0.55), while those on the traditional diet lost an average of 1.22% of their body fat (SD = 0.50). Treat those on the traditional diet as your population, and calculate the point estimate and the 95% confidence interval for a test comparing those participants on the soy diet to this known population. Does the confidence interval include 0? What can we conclude from that?

147. Mehl et al. (2007) published a study in the journal Science reporting the results of an extensive study of men and women comparing the number of words uttered per day by each sex. Volunteer participants wore inconspicuous recording devices that recorded their daily word usage. On average, women (N = 198) uttered 16,201 words per day (SD = 1779.45), and men uttered 15,993 words per day (SD = 2224.61). Treating the men's data as the population parameters, do the following: (a) Calculate the point estimate for women's words uttered per day. (b) Calculate the 95% confidence interval around the point estimate. (c) Make a decision regarding the null hypothesis on the basis of this confidence interval. (d) To see the effect of sample size on a test statistic, recalculate the 95% confidence interval assuming a study of 598 women. Reassess the status of the null hypothesis with this new sample size.

148. How is the calculation of confidence intervals superior to null hypothesis testing? Compare and contrast the information provided by both statistical techniques.

Page 24

Name:

Class:

Date:

Chap_08_5e 149. Describe meta-analysis, including the information required to compute a meta-analysis from previously published research studies. Also, discuss the role of a file drawer analysis in assessing the true strength of your estimate of effect size based on meta-analysis.

150. According to the American Psychological Association's 2015 annual report on salaries in psychology, the average salary for those working in a teaching position was $71,471, with a standard deviation of $24,703. Treating the psychology professors who responded to the survey as the population, assume that you asked eight professors at your institution what their annual salary was, and found that the average salary was $93,250 with a standard deviation of $65,319. (a) Construct an 80% confidence interval for this sample mean. (b) Construct a 95% confidence interval for this sample mean. (c) Based on these two confidence intervals, if you had performed a two-tailed hypothesis test with a p level of 0.20, would you have found that the new full professors at your school earn more, on average, than the population of new full professors? If you had performed the same test with a two-tailed p level of 0.05, would you have made another decision regarding the null hypothesis?

151. A researcher conducted a power analysis before beginning her study. When she used the software program G*power, she found that she would need 100 participants to have adequate power at 0.80. The researcher reasoned that she would run 1000 participants instead of 100 participants so that she could have even more statistical power. What is wrong with the researcher's reasoning? What factors should she consider in her reasoning?

152. Many companies that manufacture lightbulbs advertise their 60-watt bulbs as having an average life of 1000 hours. A cynical consumer bought 30 bulbs and burned them until they failed. He found that they burned for an average of M = 1183 hours, with a standard deviation of σ = 229.06. Calculate the effect size for the difference between the sample and the population. What does this effect size mean?

Page 25

Name:

Class:

Date:

Chap_08_5e 153. When analyzing effect size, why does effect size increase as distributions become further apart?

154. Is it possible to obtain an effect size of 1.00 using Cohen's statistic? Explain your answer.

Page 26

Name:

Class:

Date:

Chap_08_5e Answer Key 1. False 2. False 3. False 4. False 5. False 6. False 7. False 8. True 9. True 10. True 11. True 12. False 13. True 14. True 15. False 16. True 17. True 18. True 19. False 20. True 21. True 22. True 23. False 24. False 25. True 26. True Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_08_5e 27. True 28. False 29. True 30. False 31. False 32. False 33. False 34. False 35. False 36. False 37. False 38. a 39. d 40. b 41. d 42. d 43. d 44. a 45. b 46. b 47. c 48. d 49. a 50. c 51. c 52. b 53. b 54. c Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_08_5e 55. c 56. b 57. b 58. a 59. d 60. c 61. a 62. c 63. b 64. c 65. b 66. c 67. b 68. c 69. b 70. b 71. b 72. c 73. b 74. a 75. c 76. b 77. b 78. c 79. b 80. a 81. b 82. a Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_08_5e 83. a 84. c 85. b 86. b 87. a 88. a 89. b 90. c 91. b 92. d 93. a 94. a 95. c 96. b 97. d 98. b 99. b 100. a 101. d 102. b 103. a 104. b 105. c 106. c 107. d 108. d 109. c 110. a Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_08_5e 111. c 112. b 113. d 114. b 115. one; two 116. point estimate 117. standard error 118. standard deviation 119. sample size 120. less 121. 52 122. increases 123. increases 124. independent 125. statistical power 126. 0.8 127. decreases 128. participants 129. 0.5 130. sample 131. more 132. meta-analysis 133. power 134. interval estimate 135. decreases 136. increases 137. increases

Page 31

Name:

Class:

Date:

Chap_08_5e 138. decreases 139. z 140. file drawer analysis 141. less 142. 0.2 143. null hypothesis 144. (a) Cohen's d =

0.60. (b) Statistical power is 0.6700. (c) No. The criterion is 0.80.

145. (a) The 95% confidence interval is [27.80, 37.40] (using the z cutoff of –1.96 and 1.96). (b) Based on this confidence interval, we would reject the null hypothesis at a two-tailed p of 0.05. The average age of the drivers getting speeding tickets is significantly lower than that of the whole population of drivers. 146. The point estimate is 2.3% body fat loss and the confidence interval is 1.96(0.50/3.464) ± 2.3% or 0.283 ± 2.3%, listed as [2.017, 2.583]. Note that 3.464 is the square root of 12. This confidence interval does not include 1.22% body fat loss, the null hypothesized value, which means that 1.22% is not a likely body fat loss amount when repeatedly sampling from these populations. The null hypothesis states that no change (zero change) is expected, and in this case, zero change would mean body fat loss of 1.22%. In other words, the soy diet produces a significantly greater loss in body fat percentage than does the traditional diet. 147. (a) The point estimate for women is 16,201 words uttered per day. (b) The 95% confidence interval is (1.96) (158.096) ± 16,201 or [15,891.13, 16,510.87]. (c) Given that the confidence interval includes the number of words uttered per day by men, we would fail to reject the null hypothesis. We do not have evidence that the number of words uttered by men and women differs. (d) The 95% confidence interval is now (1.96)(90.971) ± 16,201 or [16,022.7, 16,379.3]. Now the rate for men, 15,993, does not fall within our confidence interval, indicating that the sample of women does not overlap with the national data on men. We would reject the null hypothesis. 148. Both statistical techniques allow us to determine whether it is likely that two sample means are drawn from the same population. In a null hypothesis test, if the calculated statistic is more extreme than the critical value for the statistic, then we reject the null hypothesis. When calculating confidence intervals based on two groups, for example, if the confidence interval does not include 0, we can basically reject the null hypothesis. The confidence interval provides more information, however, because it also offers us a sense of how large the range of expected values is for a given group. The confidence interval begins to hint at how different the two means are.

Page 32

Name:

Class:

Date:

Chap_08_5e 149. Meta-analysis takes the effect size values from many studies conducted about the same topic and calculates a mean of those effect sizes. The studies you collect and analyze must all fit your predetermined criteria for inclusion. To be included, the study must also report effect size or include sufficient summary statistics so that effect size can be computed. Those summary statistics include the mean, hypothesized mean, and the standard deviation (see the equation for Cohen's d). Once you have computed the average of the effect size measures you have gathered, you still do not know if that average is inflated and higher than perhaps it should be. Remember, there is a bias in the research literature to publish significant effects, rather than the often important null effects that researchers compute. A file drawer analysis, or a statistical calculation of how many null results would be needed to reduce the average effect size calculated as part of your meta-analysis to a nonsignificant finding, can help assess that degree of inflation. 150. (a) The 80% confidence interval is [81,983.36, 104,516.64]. To build this confidence interval, a z cutoff of 1.29 is used, so as not to exceed 10% in each tail. (b) The 95% confidence interval is [76,131.69, 110,368.31]. (c) In this case, regardless of whether the p level is 0.20 or the more conservative 0.05, the null hypothesis should not have been rejected. 151. The researcher did not account for the fact that increasing her sample size would directly impact her power. Power analysis is affected by the size of the sample. Large sample sizes also influence the test statistic, making it easier to reject the null hypothesis. Increasing her sample to 1000 might result in less meaningful results because the results might be significant only as an artifact of the large sample size. 152. The effect size is Cohen's d = 0.80. This effect size indicates that the life of the purchased bulbs is eight-tenths of a standard deviation greater than the manufacturer's claims for bulb life. 153. Effect size compares the amount of difference between two distributions—that is, the distance between the two distribution means. The further the distribution means are from one another, the greater the effect size. When the distributions are further apart, the differences between the two distributions are greater, which increases the effect size statistic. This increase allows us to quantify the relative impact of the statistically significant results. 154. Yes, it is possible to obtain an effect size equal to or greater than 1.00 with Cohen's statistic. Cohen provided rough guidelines for small, medium, and large effect sizes, rather than hard-and-fast rules. Cohen estimated that a large effect size is approximately 0.8, but it is possible to get an effect size as large as 1.0.

Page 33

Name:

Class:

Date:

Chap_09_5e Indicate whether the statement is true or false. 1. The z test uses estimated standard error, whereas the t test uses actual standard error. a. True b. False 2. As degrees of freedom decrease, the critical cutoff for the t test decreases. a. True b. False 3. The standard deviation of the sample is used to estimate the standard deviation of the population. a. True b. False 4. Effect size measures have a different meaning for the t test than they do for the z test. a. True b. False 5. As degrees of freedom decrease, the critical cutoff for the t test increases. a. True b. False 6. The appropriate APA format for the presentation of t test results appears as follows: t = 4.27, df = 29, p < 0.05. a. True b. False 7. When computing the confidence interval for a t test, the cutoffs for the 95% interval are always the same. a. True b. False 8. For smaller samples, the t distributions are skinnier and taller than the z distributions. a. True b. False 9. Replication involves conducting the same study with the same people multiple times. a. True b. False 10. The appropriate APA format for the presentation of t test results appears as follows: t(29) = 4.27, p < 0.05. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_09_5e 11. For smaller samples, the t distributions are wider and flatter than the z distributions. a. True b. False 12. When computing standard error for the t test, you divide s by the square root of N. a. True b. False 13. Replication can help determine the circumstances in which a finding holds true or not. a. True b. False 14. The standard deviation of the population is used to estimate the standard deviation of the sample. a. True b. False 15. If a t statistic is found to be negative, then an error has been made in its calculation. a. True b. False 16. As degrees of freedom increase, the critical cutoff for the t test decreases. a. True b. False 17. For a two-tailed t test, there are two critical cutoff values. a. True b. False 18. The single-sample t test compares the mean of a sample to a population for which the mean and standard deviation are known. a. True b. False 19. The standard error is always smaller than the standard deviation because a distribution of means is less variable than a distribution of scores. a. True b. False 20. The t test is used when the mean and the standard deviation of the population are known. a. True b. False

Page 2

Name:

Class:

Date:

Chap_09_5e 21. A sample of 21 with a standard deviation of 82.41 results in a standard error calculated as 18.42. a. True b. False 22. When computing standard error for the t test, you divide s by the square root of N – 1. a. True b. False 23. A t distribution with 3 degrees of freedom has fewer extreme scores than a t distribution with 20 degrees of freedom. a. True b. False 24. The crowdsourcing movement in research has led to open science. a. True b. False 25. There is one t distribution for all possible sample sizes. a. True b. False 26. For a two-tailed t test, there is only one critical cutoff value. a. True b. False 27. There is one t distribution for each possible sample size. a. True b. False 28. The corrected standard deviation formula serves to increase the estimate of variability. a. True b. False 29. The single-sample t test compares the mean of a sample to a population for which the mean is known but the standard deviation is not known. a. True b. False 30. The t test is a parametric statistical test that allows you to generalize what was learned about one sample to a larger population. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_09_5e 31. Degrees of freedom for the single-sample t test is calculated as N – 2. a. True b. False 32. A sample of 21 with a standard deviation of 82.41 results in a standard error calculated as 17.98. a. True b. False 33. A confidence interval for a t test is centered at the mean of the sample. a. True b. False 34. Replication is not warranted when a finding is new or unexpected. a. True b. False 35. For a one-tailed t test, there is only one critical cutoff value. a. True b. False 36. Effect size measures have the same meaning for the t test as they do for the z test. a. True b. False 37. The more observations made, the more confident the researcher can be in the estimate of the population. a. True b. False 38. If a t statistic is found to be positive, then something has been done right in its calculation. a. True b. False 39. As degrees of freedom increase, the critical cutoff for the t test increases. a. True b. False 40. The symbol s represents the standard deviation of the population. a. True b. False

Page 4

Name:

Class:

Date:

Chap_09_5e 41. The t statistic indicates the distance of a sample mean from a population mean in terms of estimated standard error. a. True b. False 42. Replication involves conducting the same study with different people. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 43. H1: µ1 ≠ µ2 is used to represent the: a. confidence interval. b. research hypothesis. c. null hypothesis. d. standard error. 44. As the sample size becomes smaller, the t distributions become: a. narrower. b. wider. c. more reliable. d. more accurate. 45. With very few degrees of freedom, the test statistic: a. becomes more reliable. b. should be a z test rather than a t test. c. needs to be more extreme to reject the null hypothesis. d. needs to be less extreme to reject the null hypothesis. 46. A t statistic is _______ a z statistic, making it _______. a. not as extreme as; less conservative b. not as extreme as; more conservative c. more extreme than; less conservative d. more extreme than; more conservative 47. The third step in conducting the single-sample t test is: a. identifying a Type I error. b. stating the null and research hypotheses. c. determining the critical values or cutoffs. d. determining the characteristics of the comparison distribution. Copyright Macmillan Learning. Powered by Cognero.

Page 5

Name:

Class:

Date:

Chap_09_5e 48. The second step in conducting the single-sample t test includes: a. identifying a Type I error. b. identifying a Type II error. c. checking that assumptions have been met. d. stating the null and research hypotheses. 49. A researcher collects 36 data points that yield a mean of 8.664 and a standard deviation (based on N – 1) of 2.238. If the researcher is comparing the sample to a population mean of 9.5, what is the 95% confidence interval? a. [7.91, 9.42] b. [8.03, 9.29] c. [7.65, 9.68] d. [8.74, 10.26] 50. When the population mean is known but the population standard deviation is not known, which statistic is used to compare a sample to the population? a. µ b. z c. F d. t 51. If we conduct the same study with different samples and get the same results each time, it is: a. more likely that we have an error in our results. b. more likely that the results are accurate. c. less likely that the results are accurate. d. less likely that we can reject the null hypothesis. 52. The t statistic for a single-sample t test indicates the: a. number of standard deviations an individual score is from the sample mean. b. number of standard deviations a sample is from the population mean. c. distance of two sample means from a single population mean. d. distance of a sample mean from the population mean in terms of estimated standard error.

Page 6

Name:

Class:

Date:

Chap_09_5e 53. What is the formula for the single-sample t statistic? a.

54. The second step in conducting the single-sample t test involves _____ and _____. a. identifying a Type I error; checking that assumptions have been met b. identifying a Type II error; stating the null hypothesis c. stating the null hypothesis; stating the research hypothesis d. stating the null hypothesis; identifying effect size 55. A negatively skewed distribution would most likely violate which assumption? a. normality b. dependent variable is scale c. random selection d. random assignment 56. It is known that the population mean on the math portion of the SAT is 527, with a standard deviation of 107. Assume that the average math SAT score for freshmen entering your college is 550, with a standard deviation of 110. What statistical analysis is used to answer this question: Does the math performance of students entering your college differ from that of all individuals taking the SAT?? a. single-sample t test b. z test c. standard deviation analysis d. dependent-samples t test

Page 7

Name:

Class:

Date:

Chap_09_5e 57. When performing a single-sample t test, an effect size of 0.84 would be interpreted as a _____ effect. a. small b. medium c. large d. negligible 58. The crowdsourcing movement in research is: a. closed science. b. new science. c. open science. d. dirty science. 59. Many companies that manufacture lightbulbs advertise their 60-watt bulbs as having an average life of 1000 hours. A cynical consumer bought 30 bulbs and burned them until they failed. He found that they burned for an average of M = 1233 hours, with a standard deviation of s = 232.06 hours. What statistical test would this consumer use to determine whether the average burn time of lightbulbs differs significantly from that advertised? a. single-sample t test b. z test c. standard deviation analysis d. dependent-samples t test 60. At a sample size of infinity, the t distribution: a. is unreliable. b. has a standard deviation of 0. c. is identical to the z distribution. d. has a standard deviation of 0.50. 61. If studies fail to replicate: a. it is unlikely the original study is accurate. b. science is a failure itself. c. it can help us better understand the context under which findings might exist. d. it confuses our understanding of the original study. 62. The correct formula for effect size using Cohen's d for a single-sample t test is: a. (M – µ)/s + σ. b. (M – µ)/s. c. (X – µ)/s. d. (σ – µ)/s.

Page 8

Name:

Class:

Date:

Chap_09_5e 63. A newspaper article reported that the typical American family spent an average of $86.79 for Halloween candy and costumes last year. A sample of N = 16 families this year reported spending a mean of M = $90, with s = $21. What statistical test would be used to determine whether these data indicate a significant change in holiday spending? a. single-sample t test b. z test c. effect-size test d. paired-samples t test 64. The fifth step in conducting the single-sample t test is: a. identifying a Type I error. b. determining the critical values or cutoffs. c. calculating the test statistic. d. determining the characteristics of the comparison distribution. 65. As the sample size becomes larger, the t distributions look ________ the z distribution. a. less like b. more like c. different from d. taller than 66. A single-sample t test is conducted on a sample of 25 people who were selected from a large population estimated at 2500 people. The critical cutoffs for a two-tailed test at a p level of 0.05 would be: a. –1.711 and 1.711. b. –1.980 and 1.980. c. –2.060 and 2.060. d. –2.064 and 2.064. 67. According to the National Student Clearinghouse Research Center, the average time to complete a 4-year bachelor degree was actually 5.1 years in 2016. You collect data on the 65 psychology students who started school during the same semester as you, finding an average time to complete at 4.7 years with a standard deviation of 0.6 year. What would be the effect size associated with a single-sample t test? a. –5.37 b. –0.67 c. 0.40 d. 0.92

Page 9

Name:

Class:

Date:

Chap_09_5e 68. The single-sample t test compares a sample mean to a population mean when: a. the sample standard deviation is unknown. b. no comparison can be made based on the variability in either group. c. the population standard deviation is unknown. d. a within-groups design is employed. 69. A researcher conducts a single-sample t test and finds statistical significance at the 0.01 level. The effect size is then calculated and found to be 0.05. What might be concluded about the findings? a. The findings are statistically significant, but the effect size indicates that they may not be practically significant. A very large sample may have been studied, driving up the test statistic value. b. The findings seem flawed because without a substantial effect size, it is not possible to find statistical significance. c. The findings are both statistically and practically significant in this case, as the effect size indicates a medium effect and the 0.01 level of significance is rather impressive. d. These findings are exciting because statistical significance was found with a very small effect, indicating that the results are real and the chance of a Type I error are low. 70. Which report of statistical results is in appropriate APA format? a. t(25) = 1.7, fail to reject null b. t = 1.7, df = 25, reject null c. t(25) = 1.7, p > 0.05 d. t = 1.7, df = 25, p > 0.05 71. The symbol s is a _____ letter that describes a _____ statistic. a. Latin; sample b. Greek; sample c. Latin; population d. Greek; population 72. A researcher collects 25 data points that yield a mean of 8.164 and a standard deviation (based on N – 1) of 2.467. What is the standard error for the distribution of means? a. 0.099. b. 0.302 c. 0.493 d. 0.504

Page 10

Name:

Class:

Date:

Chap_09_5e 73. A researcher collects 36 data points that yield a mean of 8.664 and a standard deviation (based on N – 1) of 2.238. If the researcher is comparing the sample to a population mean of 9.5, what is the value of the test statistic? a. –0.374 b. –0.836 c. –2.210 d. –2. 241 74. When scientists call a hypothesis test conservative, they mean that it: a. is difficult to reject the null hypothesis (i.e., the status quo). b. leads to results that Republicans will find favorable. c. is very easy to reject the null hypothesis (i.e., the status quo). d. leads to results that Democrats will find favorable. 75. Why do we divide by N – 1 rather than by N when estimating a population standard deviation from the sample standard deviation? a. Because the sample standard deviation is likely to be an imprecise estimate, we allow the error of the estimate (the standard deviation) to be larger by dividing the sum of squares by a smaller number. b. The sample standard deviation is a superior estimate of the variability in the population than is the population standard deviation. c. We typically have to throw out at least one data point in any given study, so the N – 1 allows us to account for that. d. Because the population is always smaller than the sample, we must divide by a smaller number. 76. When multiple researchers recruit participants from various locations to replicate findings, this practice is called: a. crowdsourcing science. b. stealing other researchers' work. c. independent science. d. new research. 77. If the standard deviation for a population, as estimated from a sample, is s = 7.43, then the standard error for a sample size of N = 22 is sM =: a. 0.34. b. 1.58. c. 1.62. d. 2.96.

Page 11

Name:

Class:

Date:

Chap_09_5e 78. If the standard deviation for a population, as estimated from a sample, is s = 4.8, then the standard error for a sample size of N = 16 is sM =: a. 0.30. b. 0.83. c. 1.20. d. 1.24. 79. The first step in conducting the single-sample t test includes: a. identifying a Type I error. b. identifying a Type II error. c. checking that assumptions have been met. d. stating the null hypothesis. 80. The average salary for all 27 Arizona Diamondbacks (Major League Baseball) players in the 2012 baseball season was $2,653,029, with a standard deviation of $2,587,139. The population, all National League baseball players in 2012, earned an average salary of $3,213,479, with a standard deviation of $1,451,308. Does the average salary of the Diamondbacks players differ from that of all players in the National League? What statistical analysis is used to answer this question? a. single-sample t test b. z test c. standard deviation analysis d. dependent-samples t test 81. A researcher collects 36 data points that yield a mean of 8.664 and a standard deviation (based on N – 1) of 2.238. If the researcher is comparing the sample to a population mean of 9.2 using a single-sample t test, what would be the effect size? a. –0.54 b. –0.24 c. 0.24 d. 0.37 82. The fourth step in conducting the single-sample t test is: a. stating the null and research hypotheses. b. determining the critical values or cutoffs. c. calculating the test statistic. d. determining the characteristics of the comparison distribution.

Page 12

Name:

Class:

Date:

Chap_09_5e 83. A team of researchers conducted a single-sample t test and found that the 95% confidence interval around their sample mean includes the value of the population mean. Based on this, they should conclude that: a. the null hypothesis should be rejected. b. their data are insufficient to assess the validity of the null hypothesis. c. they failed to find a significant difference between the sample and population means. d. the research hypothesis is supported. 84. The final step in conducting the single-sample t test is: a. making a decision. b. calculating the test statistic. c. determining the critical values. d. identify the characteristics of the comparison distribution. 85. The standard deviation of a distribution of scores in a sample is known as _____, while the standard deviation of a distribution of means is referred to as _____. a. sM; s b. s; σ c. s; sM d. σM; σ 86. The difference between the denominator of the z statistic and that of the single-sample t statistic is that in a: a. z statistic we divide by the actual population standard error (sM), but in a t statistic we divide by the estimated standard error (σM). b. z statistic we divide by the actual population standard error (σM), but in a t statistic we divide by the estimated standard error (sM). c. t statistic we divide by the actual population standard error (sM), but in a z statistic we divide by the estimated standard error (σM). d. t statistic we divide by the actual population standard error (σM), but in a z statistic we divide by the estimated standard error (sM). 87. The formula for the degrees of freedom for the single-sample t test is: a. N. b. dfX + dfY. c. N – 1. d. (N – 1)(N – 1).

Page 13

Name:

Class:

Date:

Chap_09_5e 88. What is the correct formula for calculating the standard deviation of a sample when you are NOT trying to estimate the population standard deviation? a.

89. The correct formula for the upper bound of a confidence interval for a single-sample t test is: a. t(sM) + Msample. b. –t(sM) – Msample. c. t(sM) – Msample. d. –t(sM) + Msample. 90. According to the National Student Clearinghouse Research Center, the average time to complete a 4-year bachelor degree was actually 5.1 years in 2016. You collect data on the 65 psychology students who started school during the same semester as you, finding an average time to complete at 4.7 years with a standard deviation of 0.6 year. What is your 95% confidence interval? a. [4.73, 4.77] b. [3.85, 4.15] c. [3.50, 5.90] d. [4.55, 4.85] 91. Of the statistical results shown here, which would lead us to reject the null hypothesis? a. t(5) = 2.51, p = 0.06 b. t(5) = 2.02, p = 0.10 c. t(15) = 2.23, p < 0.05 d. t(15) = 2.12, p > 0.05

Page 14

Name:

Class:

Date:

Chap_09_5e 92. Of the statistical results shown here, which would lead us to fail to reject the null hypothesis? a. t(5) = 2.95, p = 0.04 b. t(15) = 2.95, p = 0.01 c. t(15) = 2.20, p < 0.05 d. t(11) = 2.20, p > 0.05 93. As the sample size becomes larger, the t distributions become: a. narrower. b. wider. c. broader. d. less accurate. 94. The symbol representing a standard deviation calculated by using a sample to estimate the population standard deviation is: a. sd. b. σ. c. s. d. SD. 95. When performing a single-sample t test, an effect size of 0.22 would be interpreted as a _____ effect. a. small b. medium c. large d. negligible 96. When performing a single-sample t test, an effect size of 0.51 would be interpreted as a _____ effect. a. small b. medium c. large d. negligible 97. The correct formula for the lower bound of a confidence interval for a single-sample t test is: a. t(sM) + Msample. b. –t(sM) – Msample. c. t(sM) – Msample. d. –t(sM) + Msample.

Page 15

Name:

Class:

Date:

Chap_09_5e 98. For the single-sample t test, the confidence interval is centered on the: a. sample mean. b. population mean. c. difference between the sample and population means. d. standard error of the distribution of means. 99. What is the correct formula for using the sample standard deviation to estimate the population standard deviation? a.

100. The numerator (top portion) of the ratio for calculating all t statistics contains: a. a difference between means. b. a variance estimate. c. the degrees of freedom. d. the sample mean. 101. H0: µ1 = µ2 is used to represent the: a. confidence interval. b. research hypothesis. c. null hypothesis. d. standard error.

Page 16

Name:

Class:

Date:

Chap_09_5e 102. The following data were collected from a veterinarian's office over the course of a day to assess the average weight of dogs (in pounds): 10, 5, 17, 22, 50, 32, 38, 29. What is the standard deviation for these data, first without the correction and then with the correction used when estimating a population? a. 13.00; 13.90 b. 13.90; 13.00 c. 13.95; 14.91 d. 14.91; 13.95 Enter the appropriate word(s) to complete the statement. 103. For a sample of 32 people, with a standard deviation of 4.87, the standard error is _______.

104. _______ is the measure of effect size commonly used for the t statistic.

105. The number of scores that are free to vary when estimating a population parameter from a sample is the _______.

106. The critical values of the t statistic for a two-tailed test with df = 154 and a p level of 0.01 are _______.

107. There are many t distributions, one for each possible _______.

Page 17

Name:

Class:

Date:

Chap_09_5e 108. A t test is used when one knows the population _____ but not the _____.

109. The t statistic indicates the distance of a sample mean from a(n) _______ in terms of estimated standard error units.

110. The sample mean should fall at/in the _______ of the confidence interval.

111. The symbol for the estimated standard error is _____.

112. The two groups compared in the single-sample t test are the sample and the _______.

113. The _______ end of the confidence interval is calculated as –t(sM) + Msample.

114. The t test compares _______; therefore, the denominator must include an estimate of variability among means called the standard error.

Page 18

Name:

Class:

Date:

Chap_09_5e 115. A hypothesis test for comparing data from a sample to a population when the population mean is known but the population standard deviation is unknown is the _______ test.

116. When the t statistic _______ the critical cutoffs, the researcher should reject the null hypothesis.

117. The t statistic indicates the distance of a sample mean from a population mean in terms of estimated _______.

118. Conducting the same study with different samples to determine how likely the results are accurate is called _______.

119. The critical value of the t statistic for a one-tailed test with df = 16 and a p level of 0.05 is _______.

120. When the t statistic _______ the critical cutoffs, the researcher should fail to reject the null hypothesis.

Page 19

Name:

Class:

Date:

Chap_09_5e 121. When using a sample to estimate variability in the population, the researcher assumes that the sample will _______estimate the population variability, thus requiring a "correction."

122. A Cohen's d equal to _______ is considered a small effect for a single-sample t test.

123. The t test tells how confident one can be that the sample differs from a larger _______.

124. A researcher reports the results of a single-sample t test as t(24) = 2.13. There were _______ participants in the researcher's sample.

125. The t test compares means; therefore, the denominator must include an estimate of variability among means called the _______.

126. A Cohen's d equal to _______ is considered a large effect for a single-sample t test.

Page 20

Name:

Class:

Date:

Chap_09_5e 127. If one needs to know the critical values for the z distribution and a t table is available, the critical values can be looked up on the t table using _______ degrees of freedom.

128. When multiple researchers recruit participants from various locations to participate in a study it is called _______.

129. When the t statistic exceeds the critical cutoffs, the researcher should _______ the null hypothesis.

130. When comparing the formulas for the z and t statistics, the _______ of the formula is where a difference is observed.

131. The critical values of the t statistic for a two-tailed test with df = 13 and a p level of 0.05 are _______.

132. A Cohen's d equal to _______ is considered a medium effect for a single-sample t test.

Page 21

Name:

Class:

Date:

Chap_09_5e 133. When the t statistic does not exceed the critical cutoffs, the researcher should _______ the null hypothesis.

134. The critical cutoffs for a two-tailed test with df of 22 and a p level of 0.05 are _______.

135. A correction of _______ is used when calculating the standard deviation estimate for the population.

136. Effect size assesses the difference between the sample and population means in terms of _______ units.

137. The _______ end of the confidence interval is calculated as t(sM) + Msample.

138. Earlier you learned that parametric tests are statistical analyses based on a set of assumptions about the population and that nonparametric tests are statistical analyses that are not based on a set of assumptions about the population. Is the single-sample t test parametric or nonparametric? Explain your answer.

Page 22

Name:

Class:

Date:

Chap_09_5e 139. (Table: TV and Aggressiveness) In a fictional study, researchers examined the influence of a television program on children's aggressiveness. The number of aggressive responses was measured during an observation period after viewing the television program. Imagine that the known national average for number of aggressive responses typically performed by children who do not watch television is 6.847. (a) Perform the six steps of hypothesis testing using the data in the table and a one-tailed test to determine if there is an increase in the number of aggressive behaviors in children after having viewed the television program. (b) Now consider the same data using a nondirectional test. Does your decision about the hypothesis change? Explain why changing the directionality of your test can change your decision. (c) Compute the effect size and interpret its meaning. Relate its meaning to the issue of directional versus nondirectional tests. Table: TV and Aggressiveness Participant 1 2 3 4 5 6 7

Aggression after viewing the TV show 9 4 11 12 14 7 12

140. An all-women's college is interested in whether it places more females in male-dominated careers (e.g., engineering, physical science) than is reflected in the national data for career placement. According to some statistics from the National Center for Educational Statistics, only approximately 22% of people in engineering and physical science jobs were females in the 1990s (see Bona, Kelly, & Jung, 2011, who published about this topic in the Psi Chi journal, if you are interested in this topic). For this problem, assume that figure has remained constant over time. You examine your alumni data, which simply include annual averages over the past 25 years (N = 25). You find that, on average, 23.7% of graduates have been placed in such occupations, with a standard deviation of 6.1%. (a) Compute the t statistic to assess your hypothesis as a two-tailed test with p of 0.05. (b) Compute the 95% confidence interval to assess your hypothesis. (c) Compute the effect size for this analysis and explain what additional information it provides.

Page 23

Name:

Class:

Date:

Chap_09_5e 141. As quality control manager at a raisin manufacturing and packaging plant, you want to ensure that all the boxes of raisins you sell are comparable, with 30 raisins in each box. In the plant, raisins are poured into boxes until the box reaches its sale weight. To determine whether a similar number of raisins are poured into each box, you randomly sample 36 boxes ready to leave the plant and count the number of raisins in each. You find the mean number of raisins in each box to be 29.86, with s = 3.87. Perform the six steps of hypothesis testing to determine whether the average number of raisins per box differs from the expected 30.

142. Are there circumstances when we do not want to reject the null hypothesis, or is it always the case that statistical hypothesis testing is aimed at rejecting the null hypothesis? Explain your answer.

143. Many companies that manufacture lightbulbs advertise their 60-watt bulbs as having an average burn time of 1000 hours. A cynical consumer bought 30 bulbs and burned them until they failed. He found that they burned for an average of M = 1183 hours, with a standard deviation of s = 229.06 hours. Perform all six steps of hypothesis testing to determine whether the burn time of lightbulbs differs from that claimed by companies manufacturing them.

144. Many companies that manufacture lightbulbs advertise their 60-watt bulbs as having an average burn time of 1000 hours. A cynical consumer bought 30 bulbs and burned them until they failed. He found that they burned for an average of M = 1138 hours, with a standard deviation of s = 221.04 hours. Calculate a 95% confidence interval to determine whether the burn time of lightbulbs differs from that claimed by companies manufacturing them.

Page 24

Name:

Class:

Date:

Chap_09_5e 145. Using the following information, calculate an effect size using Cohen's d for a single-sample t test: population mean = 13.5, sample mean = 15, s = 6.57.

146. A quality control manager at a raisin manufacturing and packaging plant wants to ensure that all the boxes of raisins sold are comparable, with 30 raisins in each box. In the plant, raisins are poured into boxes until the box reaches its sale weight. To determine whether a similar number of raisins are poured into each box, you randomly sample 36 boxes ready to leave the plant and count the number of raisins in each. You find the mean number of raisins in each box to be 29.86, with s = 3.87. Calculate a 95% confidence interval to assess whether the number of raisins is different from the expected 30 count.

147. A researcher calculated a standard error for the t statistic. The standard deviation was 1.29 and the sample size was 18. The researcher performed the following operations to convert the standard deviation into a standard error: 1.29/ 17 = .31. Are the researcher's operations and answer correct? Explain your answer.

Page 25

Name:

Class:

Date:

Chap_09_5e 148. (Table: Infant Attention) A researcher is interested in whether infants' attention to their mothers' voices increases in the first week of life. Assume an established baseline exists showing that infants attend to their mothers on average 5.97 seconds on their first day after birth. The researcher selects 15 full-term infants in normal health who experienced uncomplicated deliveries and tests the number of seconds the infants oriented in the direction of their mother's voice on day 7 after delivery. The fictional data follow. Table: Infant Attention Day 7 (seconds) 7 7 6 8 8 8 8 8 6 7 7 7 7 8 6 Perform all six steps of hypothesis testing on these data using a nondirectional hypothesis test.

Page 26

Name:

Class:

Date:

Chap_09_5e Answer Key 1. False 2. False 3. True 4. False 5. True 6. False 7. False 8. False 9. False 10. True 11. True 12. True 13. True 14. False 15. False 16. True 17. True 18. False 19. True 20. False 21. False 22. False 23. False 24. True 25. False 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_09_5e 27. True 28. True 29. True 30. True 31. False 32. True 33. True 34. False 35. True 36. True 37. True 38. False 39. False 40. False 41. True 42. True 43. b 44. b 45. c 46. b 47. d 48. d 49. a 50. d 51. b 52. d 53. a 54. c Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_09_5e 55. a 56. b 57. c 58. c 59. a 60. c 61. c 62. b 63. a 64. c 65. b 66. d 67. b 68. c 69. a 70. c 71. a 72. c 73. d 74. a 75. a 76. a 77. b 78. c 79. c 80. b 81. b 82. b Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_09_5e 83. c 84. a 85. c 86. b 87. c 88. c 89. a 90. d 91. c 92. d 93. a 94. c 95. a 96. b 97. d 98. a 99. b 100. a 101. c 102. c 103. 0.861 or 0.86 104. Cohen's d 105. degrees of freedom 106. -2.617 and 2.617 107. sample size 108. mean; standard deviation 109. population mean 110. middle Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_09_5e 111. 112. population 113. lower 114. means 115. single-sample t 116. exceeds 117. standard error 118. replication 119. 1.746 or -1.746 120. does not exceed 121. under 122. 0.2 123. population 124. 25 125. standard error 126. 0.80 127. infinity 128. crowdsourcing 129. reject 130. denominator 131. -2.161 and 2.161 132. 0.5 133. fail to reject 134. -2.074 and 2.074 135. N - 1 136. standard deviation 137. upper

Page 31

Name:

Class:

Date:

Chap_09_5e 138. The single-sample t test is considered a parametric statistical test. Parametric statistics require that certain assumptions about the underlying population be met. Three assumptions must be met for the single-sample t test to be a parametric test: (1) The dependent variable is scale; (2) there is random selection of participants; and (3) the population is normally distributed.

Page 32

Name:

Class:

Date:

Chap_09_5e 139. (a) Step 1: The populations to be compared are (1) all children who have not seen the television program and (2) all children who have seen the television program. The comparison distribution will be a distribution of means, and the hypothesis test will be a single-sample t test. One of the assumptions for the t test is met: The dependent variable, number of aggressive behaviors, is measured on an interval scale. We do not know whether this dependent variable is normally distributed, and the sample size is less than 30. Thus, the sampling distribution may not be normal. Finally, it is unlikely that the children were randomly selected. Therefore, we must be cautious in generalizing from the results. Step 2: Null hypothesis: The average number of aggressive behaviors demonstrated by children who do not watch TV is the same as or greater than that demonstrated by children who watch the television program: H0: µ1 ≥ µ2. Research hypothesis: The average number of aggressive behaviors demonstrated by children who did not watch TV is less than the average number of aggressive behaviors demonstrated by children who watch the television program: H1: µ1 < µ2. Stated another way, the average number of aggressive behaviors will be greater after viewing the program. Step 3: The comparison distribution is a t distribution with a mean of 6.847 and sM = 1.30. Step 4: The critical value using a p level of 0.05, a one-tailed hypothesis test, and the t distribution with df = 6 is 1.943. Step 5: t = (9.857 – 6.847)/1.30 = 2.315 Step 6: The calculated t value exceeds the critical value. Therefore, we reject the null hypothesis. Aggressive behavior was significantly higher among children who watched the violent TV program (M = 9.857, SD = 3.436) compared to an established measure from children who do not watch TV, t(6) = 2.315, p < 0.05. (b) Now we consider the same data using a nondirectional test. For the nondirectional test, your hypotheses change to H0: µ1 = µ2 and H1: µ1 ≠ µ2. The critical value of t also changes, which now would be – 2.447 and +2.447, rather than just +1.943. This results in a change in your decision about your hypotheses. The t statistic of 2.315 no longer exceeds the critical cutoff, so we fail to reject the null hypothesis and conclude that we don't know that aggressive behavior was affected by the violent TV program. As the textbook states, nondirectional tests are preferred to one-tailed tests. When performing a one-tailed test, you pool alpha all in one end of the distribution, which reduces your critical cutoff and prevents you from finding unexpected effects. It is concerning to get excited about a result that is significant with a one-tailed test but not with a two-tailed test. (c) The effect size is d = (9.857 – 6.847)/3.436 = 0.876. This effect size exceeds the threshold for strong effects (0.80). With a larger sample size, perhaps this effect would reach statistical significance with a two-tailed test. The fact that the calculated t statistic fell between the cutoffs for the one-tailed and two-tailed tests indicates that there may be a real effect present, and the sample size is rather small in this study. Copyright Macmillan Learning. Powered by Cognero.

Page 33

Name:

Class:

Date:

Chap_09_5e 140. (a) t = (23.7 – 22)/1.22 = 1.39 The t critical value with 24 df is ±2.064. Our t statistic does not exceed the cutoff, so we would fail to reject our null hypothesis of no difference between our school and the national norm. (b) Mlower = –t (sM) + Msample = –2.0643(1.22) + 23.7 = 21.18 Mupper = t (sM) + Msample = 2.064(1.22) + 23.7 = 26.22 The population value of 22% falls within our confidence interval, indicating that our sample mean and the population mean could come from the same distribution. We would fail to reject the null hypothesis again. (c) d = (23.7 – 22)/6.1 = 0.28 This is a somewhat small effect, showing us that the school is only 0.28 standard deviation different from what was expected. While this is higher, it is really not a very big difference that would allow us to make any claims about the school creating greater job placement for women in male-dominated professions. 141. Step 1: The populations to be compared are (1) the boxes of raisins actually produced at this plant and (2) the boxes of raisins supposed to be produced at this plant. The comparison distribution will be a distribution of means, and the hypothesis test will be a single-sample t test. The assumptions for the t test are met: The dependent variable, number of raisins per box, is a scale variable. We do not know whether the number of raisins per box is normally distributed, but the sample size is larger than 30. Thus, the sampling distribution of the mean should be normally distributed. Finally, the participant boxes were randomly selected. Step 2: Null hypothesis: The average number of raisins do not differ from the number of raisins that are supposed to be in each box: H0: µ1 = µ2. Research hypothesis: The average number of raisins in each box differs from the number of raisins that are supposed to be in each box: H1: µ1 ≠ µ2 . Step 3: The comparison distribution is a t distribution with a mean of 30 and

Step 4: The critical value using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 35 is –2.030 and 2.030. Step 5:

Step 6: The calculated t value does not exceed the critical value. Therefore, you fail to reject the null hypothesis. There is no evidence that the number of raisins per box actually produced differs from the number that is supposed to be produced: t(35) = –0.217, p > 0.05.

Page 34

Name:

Class:

Date:

Chap_09_5e 142. Encourage students to actively discuss their ideas. The null hypothesis for t tests indicates that there are no statistically significant differences between groups. Sometimes, however, researchers want to prove that there are no differences between groups. Hence, researchers may want to find evidence in favor of the null hypothesis. For example, a pharmaceutical researcher may want to prove that patients do not gain weight after taking a specific medication or an instructor may want to know that student test performance is unaffected by whether the students use an e-book or a physical textbook. In these situations, it is important that the study have sufficient power to detect effects should there, in fact, be a difference so that the researcher can be confident that the failure to reject the null hypothesis is not simply a Type II error. 143. Step 1: The populations to be compared are (1) the 60-watt bulbs available for purchase and (2) the ideal 60-watt bulb, as claimed by the manufacturers. The comparison distribution will be a distribution of means, and the hypothesis test will be a single-sample t test. The assumptions for the t test are met: The dependent variable, hours of burn time, is a scale variable. We do not know whether the number of hours of life is normally distributed, but the sample size is larger than 30. Thus, the sampling distribution of the mean should be normally distributed. Finally, it is unlikely that the bulbs were randomly selected, as it would be impossible for a consumer to identify the entire population of lightbulbs. Therefore, we must be cautious in generalizing from the results. Step 2: Null hypothesis: The average burn-time hours of the purchased lightbulb does not differ from the average burn-time hours of the ideal lightbulb, as advertised by the manufacturers: H0: µ1 = µ2. Research hypothesis: The average burn-time hours of the purchased lightbulb differs from that of the ideal bulb, as claimed by the manufacturers: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a t distribution with a mean of 1000 and sM = 41.82. Step 4: The critical values using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 29 are –2.045 and 2.045. Step 5: t = (1183 – 1000)/41.82 = 4.38 Step 6: The calculated t value exceeds the critical value. Therefore, reject the null hypothesis. The average number of hours a lightbulb burns (M = 1183, SD = 229.06) is actually greater than that claimed by manufacturers: t(29) = 4.38, p < 0.05. 144. The t cutoff for a two-tailed test, p level of 0.05, and 29 degrees of freedom is ±2.045. sM = 40.36 Mlower = – t (sM) + Msample = –2.045(40.36) + 1138 = 1055.46 Mupper = t (sM) + Msample = 2.045(40.36) + 1138 = 1220.54 The 95% confidence interval is [1055.46, 1220.54]. The industry standard of 1000 hours does not fall within this range, indicating that our data represent a new finding in lightbulb life. We would reject the null hypothesis and conclude that lightbulbs last significantly longer than expected, by about 138 hours on average (SD = 221.04). Copyright Macmillan Learning. Powered by Cognero.

Page 35

Name:

Class:

Date:

Chap_09_5e 145. To calculate effect size for a single-sample t test for Cohen's d, use the following formula: Cohen's d = (M – µ)/s Cohen's d = (15 – 13.5)/6.57 = 0.23 For a sample mean of 15, a population mean of 13.5, and an estimate of population standard deviation of 4.57, the effect size using Cohen's d is 0.23, which is a weak, or small, effect. 146. The t cutoff for a two-tailed test, p level of 0.05, and 35 degrees of freedom is ±2.030.

Mlower = – t (sM) + Msample = –2.030(0.645) + 29.86 = 28.55 Mupper = t (sM) + Msample = 2.030(0.645) + 29.86 = 31.17 The 95% confidence interval is [28.55, 31.17]. The desired standard of 30 raisins per box falls within this range, indicating that the data do not reveal a significant difference from what is expected. We would fail to reject the null hypothesis in this case. We cannot say that the number of raisins is significantly different from 30. 147. The researcher's operations are not correct. The formula for converting a standard deviation into a standard error for the t statistic is sM = s/ . The numerator of the formula contains the sample standard deviation, which the researcher correctly identified as 1.29. The denominator of the formula is the square root of the sample size. The researcher appears to have made an error in the denominator of the formula. The denominator should be , which translates to , the total sample size. The researcher appears to have incorrectly used the formula . The correct operation and answer is 1.29/ 18 = .30.

Page 36

Name:

Class:

Date:

Chap_09_5e 148. Step 1: The populations to be compared are (1) infants during their seventh day of life and (2) infants on their first day of life, as established by baseline data. The comparison distribution will be a distribution of means, and the hypothesis test will be a single-sample t test. The data meet the assumptions for the t test: The dependent variable, seconds of directed attention, is a scale variable. We do not know whether the seconds of attention are normally distributed, and the sample size is smaller than 30, so we should be careful in considering our findings. Finally, it is unlikely that the infants were randomly selected; therefore, we must be cautious in generalizing from the results. Step 2: Null hypothesis: The average number of seconds infants orient in the direction of their mother's voice on the seventh day of life does not differ from the baseline established on the first day of life: H0: µ1 = µ2. Research hypothesis: The average number of seconds infants orient in the direction of their mother's voice on the seventh day of life differs from the baseline established on the first day of life: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a t distribution with a mean of 5.97 seconds and sM = 0.20. Step 4: The critical values using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 14 are –2.145 and 2.145. Step 5: t = (7.2 – 5.97)/0.20 = 6.15 Step 6: The calculated t value exceeds the critical value. Therefore, we reject the null hypothesis. The average number of seconds infants attend to their mother's voice (M = 7.2 seconds, SD = 0.775) is actually greater than that claimed for the first day of life: t(14) = 6.15, p < 0.05.

Page 37

Name:

Class:

Date:

Chap_10_5e Indicate whether the statement is true or false. 1. The paired-samples t test allows you to assess how performance differs for one group of people over time. a. True b. False 2. The effect size calculations for the paired-samples t test are interpreted in the same manner as those for the single-sample t test. a. True b. False 3. Counterbalancing eliminates the confounding variables of within-groups designs. a. True b. False 4. For the paired-samples t test, one of the assumptions is that the variance of scores at time 1 is similar to the variance of scores at time 2, known as homogeneity of variance. a. True b. False 5. We would expect children's vocabularies to increase over time, with, say, 200 new words per year established as the standard. The null hypothesis for a paired-samples t test that assesses 150 children at age 5 and again at age 6 would assume an average mean difference of zero words. a. True b. False 6. The confidence interval for a paired-samples t test is centered on a hypothesized mean difference of zero. a. True b. False 7. The paired-samples t test is also known as the independent-samples t test. a. True b. False 8. The sample mean difference should fall at the center of the confidence interval for a paired-samples t test. a. True b. False 9. The paired-samples t test is also known as the dependent-samples t test. a. True b. False

Page 1

Name:

Class:

Date:

Chap_10_5e 10. For the paired-samples t test, confidence interval calculations do not add any additional information. a. True b. False 11. We would expect children's vocabularies to increase over time, with, say, 200 new words per year established as the standard. The null hypothesis for a paired-samples t test that assesses 150 children at age 5 and again at age 6 would assume an average mean difference of 200 words. a. True b. False 12. If the null hypothesis is true for a paired-samples t test, the mean of the distribution of mean differences is typically 0. a. True b. False 13. Counterbalancing minimizes order effects due to the presentation of different levels of the dependent variable. a. True b. False 14. The paired-samples t test allows you to assess how performance differs between two groups of people over time. a. True b. False 15. The equation for the paired-samples t test is the same as that for the single-samples t test except that the data are now difference scores. a. True b. False 16. Counterbalancing minimizes order effects due to the presentation of different levels of the independent variable. a. True b. False

Page 2

Name:

Class:

Date:

Chap_10_5e Indicate the answer choice that best completes the statement or answers the question. 17. A clinical researcher was interested in determining whether her interventions were effective in minimizing depression symptoms among female participants. To assess the effectiveness of the treatment program, she administered a depression inventory prior to treatment and after treatment. She hypothesized that depression scores would be lower at time 2 compared to time 1. She then compared the mean differences between the two time points and found that the treatment was effective. The researcher's hypothesis is: a. one-tailed. b. two-tailed. c. not supported. d. supported by the null hypothesis. 18. Assume the following for a paired-samples t test: N = 18, Mdifference = 11.19, s = 22.7. What is the 95% confidence interval for a two-tailed test? a. [9.08, 13.30] b. [8.53, 13.85] c. [–0.10, 22.48] d. [–10.76, 10.76] 19. The null hypothesis for a paired-samples t test is: a. H0: µ1 = µ2. b. H0: µ1 ≠ µ2. c. H1: µ1 = µ2. d. H1: µ1 ≠ µ2. 20. A study by Bettmann (2007) published in the Journal of the American Psychoanalytic Association assessed whether the attachment relationships between adolescents and their parents changed as a result of a residential wilderness treatment experience for the adolescents. Participants completed the Adolescent Attachment Questionnaire at the start and at the end of the wilderness program. Which statistical test should be used to determine whether the wilderness treatment affected attachment relationships? a. single-sample t test b. z test c. effect size test d. paired-samples t test 21. What is the mean of the difference scores for the following difference scores: 2.7, 6.5, 3.8, 7.8, 10, 4.1? a. 6.98 b. 5.82 c. 5.30 d. 34.90 Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_10_5e 22. Assume the following for a paired-samples t test: N = 16, Mdifference = 15.19, s = 23.2. What is the t statistic? a. 0.65 b. 2.54 c. 2.62 d. 10.48 23. What is the 95% confidence interval for a two-tailed test for the following difference scores: –1, –24, +20, – 28, +4? a. [–14.77, –3.17] b. [–24.92, 13.32] c. [–28.86, 17.26] d. [–30.70, 19.10] 24. In a paired-samples t test, the comparison distribution is a distribution of: a. mean difference scores. b. raw score differences. c. scores. d. means. 25. The critical cutoff(s) for a one-tailed, paired-samples t test with 17 participants at a p level of 0.05 is (are): a. –1.746 or 1.746. b. –1.740 or 1.740. c. –2.120 and 2.120. d. –2.110 and 2.110. 26. An education researcher is worried that performance on a statistics aptitude test will improve simply because of repeated exposure to the instrument, which will cloud his ability to assess the impact of two unique educational interventions he wants to study (Program A and Program B). Which of these design options includes counterbalancing? a. He could vary the order of the educational interventions such that half of the participants experience Program A first and the other half experience Program B first. b. He could recruit different participants for each educational program. c. Participants could complete Program A and then wait several months before completing Program B. d. The weaker of the two programs could be administered first so that improvement would still be measurable.

Page 4

Name:

Class:

Date:

Chap_10_5e 27. The American Psychological Association encourages researchers to report _____ for paired-samples t tests. a. effect sizes only b. confidence intervals only c. both effect sizes and confidence intervals d. either effect sizes or confidence intervals 28. For the following data, what is the standard error of the difference scores? Before

After 160 125 142 187

152 123 137 172

a. 9.64 b. 4.82 c. 2.78 d. 2.41 29. What is the standard error for the paired-samples t test for the following difference scores: –1, –24, +20, – 28, +4? a. –2.59 b. 4.01 c. 8.02 d. 8.97 30. What is the standard error of the difference scores for a paired-samples t test for the following difference scores: 2.7, 6.5, 3.8, 7.8, 10, 4.1? a. 38.63 b. 2.78 c. 1.24 d. 1.13 31. Another name for a paired-samples t test is the _____ t test. a. single-sample b. dependent-samples c. independent-samples d. no-sample

Page 5

Name:

Class:

Date:

Chap_10_5e 32. Assume the following for a paired-samples t test: N = 16, Mdifference = 121.4, s = 461.7. What is the 90% confidence interval for a two-tailed test? a. [–80.94, 323.74] b. [–124.69, 367.49] c. [–80.13, 322.93] d. [70.81, 171.99] 33. The correct formula for effect size using Cohen's d for a paired-samples t test is: a. (M – µ)/s + σ. b. (M – µ)/s. c. (X – µ)/s. d. (σ – µ)/s. 34. The critical cutoffs for a two-tailed, paired-samples t test with eight participants at a p level of 0.01 are: a. –2.998 and 2.998. b. –2.365 and 2.365. c. –3.500 and 3.500. d. –3.356 and 3.356. 35. What would be the decision for a two-tailed paired-samples t test where the confidence interval was determined to be [–3.45, –0.91]? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 36. When conducting a paired-samples t test, you can assess the research hypothesis and get a range of mean differences that could be expected in the future by using: a. a hypothesis test. b. an effect size measure. c. post hoc tests. d. a confidence interval. 37. Assume the following for a paired-samples t test: N = 22, Mdifference = 357.82, s = 289.53. What is the size of the effect? a. small b. medium c. large d. no effect Copyright Macmillan Learning. Powered by Cognero.

Page 6

Name:

Class:

Date:

Chap_10_5e 38. A researcher studies 40 volunteer senior citizens from a small retirement community and asks them about the amount of caffeine (in milligrams) they ingest before and after lunch each day. The null hypothesis for this paired-samples study could be: a. more caffeine is ingested before lunch. b. more caffeine is ingested after lunch. c. there is a difference between the amounts of caffeine ingested before and after lunch. d. there is no difference between the amounts of caffeine ingested before and after lunch. 39. In a paired-samples t test, the null hypothesis posits that the mean of the comparison distribution is: a. the same as the mean of the population. b. 1. c. 0. d. –1. 40. A clinical researcher was interested in determining whether her interventions were effective in minimizing depression symptoms among female participants. To assess the effectiveness of the treatment program, she administered a depression inventory prior to treatment and after treatment. She hypothesized that depression scores would be lower at time 2 compared to time 1. She then compared the mean differences between the two time points and found that the treatment was effective. The dependent variable in this study is: a. depression scores. b. gender. c. age. d. time. 41. For the following data, what is the paired-samples t test statistic? Before 160 125 142 187

After 152 123 137 172

a. 7.50 b. 3.75 c. 2.69 d. 2.33

Page 7

Name:

Class:

Date:

Chap_10_5e 42. For the following data, what is the mean of the difference scores? Before

After 160 125 142 187

152 123 137 172

a. 7.50 b. 6.00 c. 5.57 d. 10.00 43. The formula for the degrees of freedom for the paired-samples t test is: a. N. b. dfX + dfY. c. N – 1. d. (N – 1)(N – 1). 44. When conducting a paired-samples t test, the sample mean difference is compared to: a. a distribution of mean differences. b. sample means. c. differences between means. d. the t distribution. 45. Assume the following for a paired-samples t test: N = 18, Mdifference = 14.17, s = 22.9. What is the size of the effect? a. small b. medium c. large d. no effect 46. A researcher studies 40 volunteer senior citizens from a small retirement community and asks them about the amount of caffeine (in milligrams) they ingest before and after lunch each day, a phenomenon assumed to be normally distributed. Which assumption for the paired-samples t test is NOT met in this research design? a. The dependent variable is a scale variable. b. Participants are randomly selected. c. The population is normally distributed. d. All the assumptions are met. Copyright Macmillan Learning. Powered by Cognero.

Page 8

Name:

Class:

Date:

Chap_10_5e 47. A social scientist investigates whether the extent to which people care about keeping their house clean and neat changes if they are given new things in that home. She follows eight families who were selected to receive home makeovers, assessing their cleanliness before and after the makeover. Given the confidence interval [–1.26, 2.95], what decision about the hypothesis should the researcher make? a. Reject the null hypothesis because the confidence interval includes the null hypothesized value. b. Reject the null hypothesis and conclude that cleanliness did not change as a result of the manipulation. c. Fail to reject the null hypothesis because 0 falls in the confidence interval. d. Fail to reject the null hypothesis because the confidence interval has positive and negative values. 48. When conducting a paired-samples t test, you can assess the practical importance of the obtained results by calculating: a. a hypothesis test. b. an effect size measure. c. post hoc tests. d. a confidence interval. 49. What would be the decision for a two-tailed paired-samples t test where the confidence interval was determined to be [2.04, 7.15]? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 50. What is the formula for the paired-samples t test? a. b.

c. d. 51. What would be the decision for a two-tailed paired-samples t test where the t statistic is –2.67, with cutoffs of ±2.145? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis Copyright Macmillan Learning. Powered by Cognero.

Page 9

Name:

Class:

Date:

Chap_10_5e 52. A researcher studies 40 volunteer senior citizens from a small retirement community and asks them about the amount of caffeine (in milligrams) they ingest before and after lunch each day. Two measures are taken from each participant, for a total of 80 data points. How many degrees of freedom does this paired-samples study have? a. 80 b. 79 c. 40 d. 39 53. The denominator (bottom portion) of the ratio for calculating the paired-samples t statistic is the: a. pooled variance. b. estimated standard error of the distribution of mean difference scores. c. estimated standard error of the sampling distribution of the mean. d. estimated standard deviation. 54. A clinical researcher was interested in determining whether her interventions were effective in minimizing depression symptoms among female participants. To assess the effectiveness of the treatment program, she administered a depression inventory prior to treatment and after treatment. She hypothesized that depression scores would be lower at time 2 compared to time 1. She then compared the mean differences between the two time points and found that the treatment was effective. The independent variable in this study is: a. depression scores. b. gender. c. age. d. time. 55. Twenty-five people participate in a weight-loss program for 3 months. Their weights after the 3 months are compared to their starting weights. What kind of mean difference might be expected if the null hypothesis is true for a paired-samples t test? a. a loss of pounds, on average, showing the effectiveness of the program b. a negative mean difference, indicating significant weight loss c. no change in weight or weight gain, indicating that the program does not work d. weight loss, indicating that the program does not work 56. Assume the following for a paired-samples t test: N = 17, Mdifference = 467.72, s = 264.50. What is the effect size statistic? a. 0.66 b. 0.83 c. 1.77 d. 7.29

Page 10

Name:

Class:

Date:

Chap_10_5e 57. The possible threat posed by participants' familiarity with the variables in a within-groups design when they experience those variables for the second time is known as: a. order effects. b. counterbalancing. c. repeated effects. d. stimulus repeating. 58. A paired-samples test is reported as t(28) = 1.83. What decision about this test should be made, assuming a two-tailed hypothesis test with a p level of 0.05? a. fail to reject the research hypothesis b. fail to reject the null hypothesis c. reject the null hypothesis d. reject the research hypothesis 59. What would be the decision for the following paired-samples t test: t(34) = 3.81, p < 0.05? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 60. What is the paired-samples t test statistic for the following difference scores: 2.7, 6.5, 3.8, 7.8, 10, 4.1? a. 5.82 b. 5.15 c. 4.69 d. 2.09 61. What decision should be made for a two-tailed paired-samples t test where the confidence interval was determined to be [–3.45, 2.73]? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 62. The formula for calculating effect sizes for the paired-samples t test replaces the _____ symbol used in the formula for the t statistic with the _____ symbol. a. s; sM b. sM; s c. µ; µM d. µM; µ Copyright Macmillan Learning. Powered by Cognero.

Page 11

Name:

Class:

Date:

Chap_10_5e 63. The comparison distribution in a paired-samples t test is a distribution of: a. mean difference scores. b. differences between means. c. means. d. scores. 64. Twenty-five college students experience the effects of alcohol on reaction time. They perform very basic timed responses in a driving simulator both before and after consuming several alcoholic beverages. The researcher collects a reaction-time result for each of the 20 students before and after intoxication, for a total of 40 measures. What is the null hypothesis for this paired-samples study? a. There is no difference in reaction time before and after consuming alcohol. b. Reaction time slows with alcohol consumption. c. There is a difference in reaction time before and after consuming alcohol. d. Reaction time quickens with alcohol consumption. 65. What would be the decision for a two-tailed paired-samples t test where the t statistic is 3.47, with cutoffs of ±2.145? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 66. Which of these is NOT an assumption of the paired-samples t test? a. The dependent variable is a scale variable. b. Participants are randomly selected. c. There are fewer than 30 sample data. d. The population is normally distributed. 67. Assume the following for a paired-samples t test: N = 18, Mdifference = 12.17, s = 22.4. What is the effect size using Cohen's d? a. 0.26 b. 0.68 c. 0.54 d. 2.31

Page 12

Name:

Class:

Date:

Chap_10_5e 68. One of the first steps in calculating the dependent-samples t statistic is: a. counterbalancing the order of the calculations. b. finding the variance for each of the measured variables and then pooling them. c. creating a difference score for each participant in the sample. d. averaging the scores on each of the measured variables and finding the difference between those means. 69. What is the paired-samples t test statistic for the following difference scores: –1, –24, +20, –28, +4? a. –1.45 b. –0.65 c. –0.29 d. -2.59 70. A paired-samples t test is reported as t(25) = 1.927, p > 0.05, d = 0.17. What is the effect size for this statistical test? a. There is no effect size reported. b. This is a small effect. c. This is a medium effect. d. This is a large effect. 71. The final step in calculating a confidence interval for a paired-samples t test is to convert the: a. effect sizes into t scores. b. t statistics into raw mean differences. c. t statistics into z statistics. d. z statistics into t statistics. 72. What would be the decision for a two-tailed paired-samples t test where the t statistic is –2.67, with cutoffs of ±2.776? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 73. What would be the decision for the following paired-samples t test: t(28) = 1.81, p > 0.05? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis

Page 13

Name:

Class:

Date:

Chap_10_5e 74. A psychologist is interested in whether working memory is influenced by sleep loss. The psychologist administers a measure of working memory to a group of subjects at 8 A.M. on day 1 of the study and then again at 8 A.M. on day 2 of the study, after keeping the subjects awake the entire night. Does sleep loss affect working memory? What statistical analysis should be performed to answer this question? a. single-sample t test b. z test c. standard deviation analysis d. paired-samples t test 75. An education researcher is worried that performance on a statistics aptitude test will improve simply because of repeated exposure to the instrument, which will cloud his ability to assess the impact of two unique educational interventions he wants to study. To help remove the effect of repeated exposure, the researcher could use: a. repeated measures. b. counterbalancing. c. a within-groups design. d. a single-sample t test. 76. What is the mean difference of the following difference scores: –1, –24, +20, –28, +4? a. –5.80 b. –7.40 c. 3.80 d. 4.20 77. In a within-groups design with two groups, the appropriate hypothesis test is a(n): a. single-sample t test. b. two-way analysis of variance. c. paired-samples t test. d. independent-samples t test. 78. Assume the following for a paired-samples t test: N = 15, Mdifference = 615.67, s = 628.50. What is the t statistic? a. 14.69 b. 3.79 c. 3.66 d. 2.15

Page 14

Name:

Class:

Date:

Chap_10_5e 79. A mood researcher investigates whether chocolate affects emotions. He recruits college students to take a mood inventory, ingest 0.25 pound of chocolate, and then complete the mood inventory again. Given the confidence interval [1.37, 4.21], what decision about the hypothesis should the researcher make? a. Reject the null hypothesis and conclude that chocolate affected mood because 0, or no effect, does not fall within the confidence interval. b. Reject the null hypothesis and conclude that chocolate lowers mood because the confidence interval includes small numbers. c. Fail to reject the null hypothesis because the confidence interval is positive. d. Fail to reject the null hypothesis and conclude that the effect size is small for this study. 80. The critical cutoff(s) for a two-tailed, paired-samples t test with 21 participants at a p level of 0.05 is (are): a. –2.080 and 2.080. b. –2.086 or 2.086. c. –1.721 and 1.721. d. –1.725 or 1.725. 81. Twenty-five college students experience the effects of alcohol on reaction time. They perform very basic timed responses in a driving simulator both before and after consuming several alcoholic beverages. The researcher collects a reaction time result for each of the 25 students before and after intoxication, for a total of 50 measures. What are the degrees of freedom for this study? a. 24 b. 25 c. 48 d. 50 82. What would be the decision for a two-tailed paired-samples t test where the t statistic is 2.07, with cutoffs of ±2.145? a. reject the null hypothesis b. fail to reject the null hypothesis c. reject the research hypothesis d. fail to reject the research hypothesis 83. The research hypothesis for a paired-samples t test is: a. H0: µ1 = µ2. b. H0: µ1 ≠ µ2. c. H1: µ1 = µ2. d. H1: µ1 ≠ µ2.

Page 15

Name:

Class:

Date:

Chap_10_5e 84. In a paired-samples t test, a possible confounding variable that can occur as a result of the within-groups design is: a. counterbalancing. b. repeated measures. c. order effects. d. error effects. 85. According to the null hypothesis, the mean difference for the comparison distribution in a paired-samples t test is: a. always 0. b. sometimes 0. c. always 1. d. sometimes 1. Enter the appropriate word(s) to complete the statement. 86. _______ helps to minimize order effects due to the presentation of the levels of the independent variable.

87. When conducting a paired-samples t test, the APA encourages the use and reporting of _______ and effect size.

88. A paired-samples t test has 14 scores from a pretest and 14 scores from a posttest; in total, there are _______ degrees of freedom.

89. For the paired-samples t test, _______ is used to assess effect size.

Page 16

Name:

Class:

Date:

Chap_10_5e 90. An order effect can only occur in a(n) _______ design.

91. A mean difference of 4.1 (sM = 0.23) is found based on 21 participants. The confidence interval would be expressed as [ _______ ].

92. A researcher conducts a dependent-samples t test and reports a t statistic with df = 26. The study included _______ participants.

93. When calculating a paired-samples t test, you would cross out the original data once _______ have been calculated for the participants.

94. When conducting a paired-samples t test, the APA encourages the use and reporting of confidence intervals and _______.

95. Counterbalancing helps to minimize order effects due to the presentation of the levels of the _______ variable.

Page 17

Name:

Class:

Date:

Chap_10_5e 96. Counterbalancing helps to minimize _______ due to the presentation of the levels of the independent variable.

97. One form of the t test in which scores are analyzed both before and after a manipulation is known as a(n) _______.

98. _______ refer to how a participant's behavior changes when the dependent variable is presented for a second time.

99. The paired-samples t test is a(n) _______-groups design.

100. A researcher should _______ the null hypothesis when reporting a statistical test of t(44) = 3.16, p < 0.05.

101. A researcher should _______ the null hypothesis when reporting a statistical test of t(29) = 1.91, p > 0.05.

Page 18

Name:

Class:

Date:

Chap_10_5e 102. The confidence interval of [3.42, 5.76] is centered on _______, the sample mean difference.

103. If the null hypothesis is true for a paired-samples t test, the mean of the distribution of mean differences is _______.

104. What are order effects? Why are they a problem? What is one way to reduce order effects?

105. (Table: TV and Aggressiveness) In a fictional study, a pretest–posttest design was used to examine the influence of a television program on children's aggressiveness. The number of aggressive responses was measured during an observation period both before and after the television program. Perform the six steps of hypothesis testing using the data in the table to determine if the number of aggressive behaviors differs after the children view the television program. Table: TV and Aggressiveness Participant 1 2 3 4 5 6 7

Before 6 4 12 9 10 2 14

After 9 4 11 12 14 7 12

Page 19

Name:

Class:

Date:

Chap_10_5e 106. Dr. Mitchell, a psychologist and researcher, was interested in assessing whether stress affects health. She measured student stress levels immediately after exposure to two stressors. The first stressor was spontaneous bursts of loud noise. The second stressor involved asking students to give an extemporaneous speech in front of the other participants. Stress levels were measured by asking students to complete brief self-report measures of stress immediately after exposure to the two stress conditions. Dr. Mitchell hypothesized that students would report higher stress after the public speaking stressor compared to the noise stressors. Can Dr. Mitchell use a t test to test her hypothesis? Why or why not?

107. (Table: TV and Aggressiveness) In a fictional study, a pretest–posttest design was used to examine the influence of a television program on children's aggressiveness. The number of aggressive responses was measured during an observation period both before and after the television program; the data are provided in the table. (a) Determine if there is a difference in the number of aggressive behaviors in children after having viewed the television program, using a 95% confidence interval. (b) Compute Cohen's d as a measure of effect size and interpret its meaning. Table: TV and Aggressiveness Participant 1 2 3 4 5 6 7

Before 6 4 12 9 10 2 14

After 9 4 11 12 14 7 12

Page 20

Name:

Class:

Date:

Chap_10_5e 108. (Table: Infant Attention) A researcher is interested in whether infants' attention to their mother's voice increases in the first week of life. The researcher selects 15 full-term infants in normal health who experienced uncomplicated deliveries and tests the number of seconds the infants oriented in the direction of their mother's voice on day 1 and on day 7 after delivery. The fictional data are provided in the table. (a) Assess these data as a two-tailed research hypothesis using a 95% confidence interval. (b) Compute Cohen's d as a measure of effect size and interpret its meaning. Table: Infant Attention Day 1 6 4 3 8 6 6 6 7 7 5 6 7 5 5 5

Day 7 7 7 6 8 8 8 8 8 6 7 7 7 7 8 6

109. Using the following information, calculate an effect size using Cohen's d for a paired-samples t test. Sample mean difference = 13.5 Population mean difference = 0 Population standard deviation = 9.560 N = 265

Page 21

Name:

Class:

Date:

Chap_10_5e 110. Dr. Lofgren was interested in the effects of test anxiety on concentration abilities. Using an anxiety questionnaire, he measured student anxiety levels when the students arrived at his laboratory and then again immediately before taking an examination. Dr. Lofgren hypothesized that participants in his study would have higher anxiety scores immediately prior to the completion of the exam compared to when they first came in. As hypothesized, Dr. Lofgren found that participants' anxiety scores were significantly higher immediately prior to the examination compared to baseline scores. As a result of this information, what type of t test was Dr. Lofgren MOST likely to use to test his hypothesis? Is Dr. Lofgren's hypothesis test one-tailed or twotailed? Explain your answers.

111. (Table: Infant Attention) A researcher is interested in whether infants' attention to their mother's voice increases in the first week of life. The researcher selects 15 full-term infants in normal health who experienced uncomplicated deliveries and tests the number of seconds the infants oriented in the direction of their mother's voice on day 1 and on day 7 after delivery. The fictional data are provided in the table. Perform all six steps of hypothesis testing on these data using a directional hypothesis test. Table: Infant Attention Day 1 6 4 3 8 6 6 6 7 7 5 6 7 5 5 5

Day 7 7 7 6 8 8 8 8 8 6 7 7 7 7 8 6

Page 22

Name:

Class:

Date:

Chap_10_5e 112. (Table: Sales: Before and After ) Television has been known to have an effect on the popularity of things. For example, popular shows about lawyers have preceded increased applications to law school, and the latest boom in shows about criminal profiling has increased students' interest in forensic psychology. A student double-majoring in psychology and marketing was interested in whether the popularity of a song could be affected by its appearance on a popular television show about family-owned recording company. He tracks the sales of music before and after the music is performed on the show. Hypothetical data (in millions) are provided in the table. (a) Compute the paired-samples t test and make a decision about a two-tailed hypothesis with a p level of 0.05. (b) Compute a 95% confidence interval. (c) Compute Cohen's d as a measure of effect size and interpret its meaning. Table: Sales: Before and After Sales before TV appearance 1.3 1.1 2.1 1.6

Sales after the song was covered on TV 1.6 1.5 1.4 1.8

Page 23

Name:

Class:

Date:

Chap_10_5e Answer Key 1. True 2. True 3. False 4. False 5. False 6. False 7. False 8. True 9. True 10. False 11. True 12. True 13. False 14. False 15. True 16. True 17. a 18. c 19. a 20. d 21. b 22. c 23. d 24. a 25. a 26. a Copyright Macmillan Learning. Powered by Cognero.

Page 24

Name:

Class:

Date:

Chap_10_5e 27. c 28. c 29. d 30. d 31. b 32. a 33. b 34. c 35. a 36. d 37. c 38. d 39. c 40. a 41. c 42. a 43. c 44. a 45. b 46. b 47. c 48. b 49. a 50. c 51. a 52. d 53. b 54. d Copyright Macmillan Learning. Powered by Cognero.

Page 25

Name:

Class:

Date:

Chap_10_5e 55. c 56. c 57. a 58. b 59. a 60. b 61. b 62. b 63. a 64. a 65. a 66. c 67. c 68. c 69. b 70. b 71. b 72. b 73. b 74. d 75. b 76. a 77. c 78. b 79. a 80. b 81. a 82. b Copyright Macmillan Learning. Powered by Cognero.

Page 26

Name:

Class:

Date:

Chap_10_5e 83. d 84. c 85. a 86. Counterbalancing 87. confidence intervals 88. 13 89. Cohen's d 90. within-groups 91. 3.62, 4.58 92. 27 93. difference scores 94. effect size 95. independent 96. order effects 97. paired-samples t test, dependent-samples t test 98. Order effects 99. within 100. reject 101. fail to reject 102. 4.59 103. 0, zero 104. Order effects can occur in within-groups designs when a person's response in one condition is influenced by what he or she did in a previous condition. Order effects are a problem because they introduce a confound into the experiment: We are unable to tell if any difference we observe between conditions is due to the independent variable we manipulated or to the ordering of the conditions. One way to reduce order effects is to counterbalance the ordering of the experiment conditions across participants, such that different participants get the conditions in different orders.

Page 27

Name:

Class:

Date:

Chap_10_5e 105. Step 1: The populations to be compared are (1) all children who have not seen the television program and (2) all children who have seen the television program. The comparison distribution will be a distribution of mean difference scores, and the hypothesis test will be a paired- or dependent-samples t test. One of the assumptions for the t test is met: The dependent variable, number of aggressive behaviors, is measured on an interval scale. We do not know whether this dependent variable is normally distributed, and the sample size is less than 30. Thus, the sampling distribution may not be normal. Finally, it is unlikely that the children were randomly selected. Therefore, we must be cautious in generalizing from the results. Step 2: Null hypothesis: The average number of aggressive behaviors before watching the television program is the same as that after having watched the television program: H0: µ1 = µ2. Research hypothesis: The average number of aggressive behaviors before watching the television program differs from the average number of aggressive behaviors after watching the television program: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a t distribution with a mean of 0 and sM = 1.017. Step 4: The critical value using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 6 is –2.447 or 2.447. Step 5: Table: TV and Aggressiveness Difference Participant 1 2 3 4 5 6 7

Before 6 4 12 9 10 2 14 M = 8.143 SD = 4.337

After 9 4 11 12 14 7 12 M = 9.857 SD = 3.436

Difference 3 0 –1 3 4 5 –2 Mdifference= 1.714 s = 2.690

t = (1.714 – 0)/1.017 = 1.685 Step 6: The calculated t value does not exceed the critical value. Therefore, we fail to reject the null hypothesis. We have no evidence that the number of aggressive behaviors exhibited after watching a television program differs from the number exhibited prior to watching the television program: t(6) = 1.685, p > 0.05. 106. A paired-samples t test would be appropriate because two measures are taken on the same participants. Difference scores can be computed, comparing depression before and after the students were exposed to social media. These difference scores can then be used to compute a mean difference, standard error of the differences, and the t statistic. Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_10_5e 107. (a) Table: TV and Aggressiveness Difference Participant 1 2 3 4 5 6 7

Before 6 4 12 9 10 2 14 M = 8.143 SD = 4.337

After 9 4 11 12 14 7 12 M = 9.857 SD = 3.436

Difference 3 0 –1 3 4 5 –2 Mdifference = 1.714 s = 2.690

df = 7 – 1 = 6; critical t cutoffs are ±2.447 Mdifference = 1.714 s = 2.690 sM = 1.017 Mupper = 2.145(0.306) + 1.467 = 2.123 Mlower = –2.145(0.306) + 1.467 = 0.811 The 95% confidence interval is [0.811, 2.123], indicating that the null hypothesis of no change (or zero) should be rejected. Infants attended to their mother's voice by 1.467 seconds longer, on average, on the seventh day compared to the first day. (b) d = (1.467 – 0)/1.187 = 1.24, which is a large effect size.

Page 29

Name:

Class:

Date:

Chap_10_5e 108. (a) df = 15 – 1 = 14; critical t cutoffs are ±2.145 Mdifference = 1.467 s = 1.187 sM = 0.306 Mupper = 2.145(0.306) + 1.467 = 2.123 Mlower = –2.145(0.306) + 1.467 = 0.811 The 95% confidence interval is [0.811, 2.123], indicating that the null hypothesis of no change (or zero) should be rejected. Infants attended to their mother's voice by 1.467 seconds longer, on average, on the seventh day compared to the first day. (b) d = (1.467 – 0)/1.187 = 1.24, which is a large effect size. 109. To calculate effect size for a paired-samples t test for Cohen's d, we use the following formula: Cohen's d = (M – µ)/s Cohen's d = (13.5 – 0)/9.560 = 1.41 For a sample mean difference of 13.5, a population mean difference of 0, and a population standard deviation of 9.560, the effect size using Cohen's d is 1.41. This is a large effect. 110. Dr. Lofgren probably used a paired-samples t test. The paired-samples t test is used to compare mean differences that result from each participant being measured twice. Every participant is in both samples. Dr. Lofgren's hypothesis test is a one-tailed test. In a one-tailed test, the research hypothesis is directional, positing either a mean decrease or a mean increase in the dependent variable. Conversely, a two-tailed test does not indicate a specific type (direction) of mean difference or change in the dependent variable; it merely indicates that there will be a mean difference. 111. Step 1: The populations to be compared are (1) full-term infants on the first day of life and (2) full-term infants on the seventh day of life. The comparison distribution will be a distribution of mean difference scores, and the hypothesis test will be a paired-samples t test. The researcher meets one of the assumptions for the t test: The dependent variable, seconds attended to the mother's voice, is measured on an interval scale. The researcher does not know whether this dependent variable is normally distributed, and the sample size is less than 30. Thus the sampling distribution may not be normal. Finally, it is unlikely that the infants were randomly selected. Therefore, the researcher must be cautious in generalizing from the results. Step 2: Null hypothesis: The average number of seconds attending to mother's voice on day 7 is similar to or less than the average number of seconds attending to mother's voice on day 1: H0: µ1 ≥ µ2. Research hypothesis: The average number of seconds attending to mother's voice on day 7 is greater than the average number of seconds attending to mother's voice on day 1: H1: µ1 < µ2. Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_10_5e Step 3: The comparison distribution is a t distribution with a mean of 0 and sM = 0.306. Step 4: The critical value using a p level of 0.05, a one-tailed hypothesis test, and the t distribution with df = 14 is 1.762. Step 5: Table: Infant Attention Difference Day 1 6 4 3 8 6 6 6 7 7 5 6 7 5 5 5 M = 5.733 s = 1.280

Day 7 7 7 6 8 8 8 8 8 6 7 7 7 7 8 6 M = 7.20 s = 0.775

Difference 1 3 3 0 2 2 2 1 –1 2 1 0 2 3 1 Mdifference = 1.467 s = 1.187

Mdifference = 1.467 s = 1.187

sM = 0.306

t = (1.467 – 0)/0.306 = 4.794 Step 6: The calculated t value exceeds the critical value. Therefore, we reject the null hypothesis. Infants spend longer attending to their mother's voice on day 7 (M = 7.20, SD = 0.78) than on day 1 (M = 5.73, SD = 1.28): t(14) = 4.79, p < 0.05.

Page 31

Name:

Class:

Date:

Chap_10_5e 112. (a) Table: Sales Difference Sales before TV appearance 1.1 1.1 2.1 1.6 M = 1.525 s = 0.435

Sales after the song was covered on TV 1.6 1.5 1.4 1.8 M = 1.575 s = 0.171

Difference 0.3 0.4 –0.7 0.2 Mdifference = 0.05 s = 0.507

sM = 0.254 t = (0.05 – 0)/0.254 = 0.197 The critical cutoff with 3 df is +/– 3.182. Our t statistic does not exceed our critical cutoff, so we fail to reject the null hypothesis. We do not know if a song being covered on a TV show results in change in sales. (b) Mlower = –3.182(0.254) + 0.05 = –0.758 Mupper = 3.182(0.254) + 0.05 = 0.858 The 95% confidence interval is [–0.758, 0.858], which includes the null hypothesized value of zero, so we would fail to reject the null hypothesis. We cannot say whether being covered on a TV show affects the sales of music. (c) d = (0.05 – 0)/0.507 = 0.10 This is a very weak effect.

Page 32

Name:

Class:

Date:

Chap_11_5e Indicate whether the statement is true or false. 1. The independent-samples t test is considered a within-groups design. a. True b. False 2. Effect size cannot be calculated for the independent-samples t test. a. True b. False 3. When calculating a confidence interval for an independent-samples t test, you replace the sample mean difference with the population mean difference. a. True b. False 4. For an independent-samples t test, the test result is compared to a distribution of differences between means. a. True b. False 5. Pooled variance is a weighted average of the two estimates of variance from the two samples in the independent-samples t test. a. True b. False 6. The independent-samples t test is considered a between-groups design. a. True b. False 7. The confidence interval for an independent-samples t test is an interval estimate. a. True b. False 8. For an independent-samples t test, the test result is compared to a distribution of mean differences. a. True b. False 9. Bayesian statistics help us account for both prior beliefs and probabilities. a. True b. False 10. Bayesian statistics help us account for both prior beliefs and chance. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_11_5e 11. Some researchers believe hypothesis testing should completely replace Bayesian statistics. a. True b. False 12. When calculating a confidence interval for an independent-samples t test, you replace the population mean difference with the sample mean difference. a. True b. False 13. When conducting an independent-samples t test, it is not necessary to have an equal number of participants in both groups. a. True b. False 14. When conducting an independent-samples t test, it is necessary to have an equal number of participants in both groups. a. True b. False 15. When reporting the test results for an independent-samples t test, N is reported rather than degrees of freedom. a. True b. False 16. Some researchers believe hypothesis testing should be completely replaced with Bayesian statistics. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 17. To determine the critical values or cutoffs for an independent-samples t test, we use: a. degrees of freedom for group 1. b. degrees of freedom for group 2. c. N – 1. d. total degrees of freedom. 18. The appropriate hypothesis test to use for a between-groups design with two groups is a(n): a. z test. b. paired-samples t test. c. independent-samples t test. d. single-sample t test.

Page 2

Name:

Class:

Date:

Chap_11_5e 19. In addition to reporting the results of statistical hypothesis testing, it is recommended that researchers report: a. confidence intervals and random selection results. b. effect size and confidence intervals. c. effect size and median differences. d. effect size and mean difference scores. 20. Dr. Simon wanted to know if IQ scores differed between male and female participants in his study. He gave 26 participants an IQ test and then examined IQ scores for gender differences. He hypothesized that there would be a statistically significant gender difference in IQ scores. Contrary to Dr. Simon's hypothesis, there were no differences in IQ scores between men and women in his study. What is the dependent variable in this study? a. IQ scores b. sample size of 26 c. gender d. participants 21. Checking an independent-samples t test for the assumption of normality is part of step _____ of the six steps. a. 1 b. 2 c. 3 d. 4 22. A psychopathology researcher was interested in determining whether her interventions for depression were effective in minimizing depression symptoms among the participants in her study. To assess the effectiveness of her treatment program, she administered a depression inventory prior to treatment and after treatment. She hypothesized that depression scores would be lower at time 2 compared to time 1. She then compared the mean differences between the two groups and found that her treatment was effective. Which statistical test was the researcher MOST likely to have used to test her hypothesis? a. single-sample t test b. z test c. independent-samples t test d. paired-samples t test 23. Bayesian statistics help us take into account: a. both probabilities and future predictions. b. only prior beliefs. c. both prior beliefs and probabilities. d. both prior beliefs and future predictions.

Page 3

Name:

Class:

Date:

Chap_11_5e 24. A cognitive psychologist is interested in whether working memory is impacted by sleep deprivation. The psychologist administers a measure of working memory to two groups of subjects. The subjects in one group were kept awake for the entire night, whereas the subjects in the other group maintained their normal sleep schedules. Which statistical analysis should be performed to answer the question "Does sleep loss affect working memory?" a. single-sample t test b. z test c. independent-samples t test d. paired-samples t test 25. There is(are) _____ degree(s) of freedom calculations for an independent-samples t test. a. 1 b. 2 c. 3 d. 4 26. In a(n) _____, a within-groups design with two groups is used to compare the distribution of mean difference scores. a. single-sample t test b. z test c. independent-samples t test d. paired-samples t test 27. There are three steps involved in creating the comparison distribution for the independent-samples t test, and those steps are repeated many times. Which of these is the sequence described in the text? a. randomly select a pair of scores, subtract the second score from the first, and record the difference between scores b. randomly select a group of scores, calculate their mean, and plot that mean on a distribution c. randomly select scores and calculate their mean as the group 1 mean, randomly select another group of scores and calculate their mean as the group 2 mean, and subtract the second mean from the first d. randomly select a score, subtract the mean from the score, and plot the difference 28. In a(n) _____, each participant is assigned to only one group so as to compare mean differences in a between-groups design. a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test

Page 4

Name:

Class:

Date:

Chap_11_5e 29. Dr. Simon wanted to know if IQ scores differed between male and female participants in his study. He gave 26 participants an IQ test and then examined IQ scores for gender differences. He hypothesized that there would be a statistically significant gender difference in IQ scores. Contrary to Dr. Simon's hypothesis, there were no differences in IQ scores between men and women in his study. Based on the information provided, does this study meet the assumption of normality? Why or why not? a. Yes; it is normally distributed because the sample size is large. b. No; it is not normally distributed because the small sample size is small. c. We do not know whether it is normally distributed, and the sample size is not at least 30 participants. d. Yes; the population is likely to be normally distributed because the sample size is less than 30. 30. The results of an independent-samples t-test were t(18) = 4.42, p < 0.05. These results are: a. statistically significant. b. generalizable. c. meaningful. d. not statistically significant. 31. In an independent-samples t test, how is the null hypothesis written in symbols? a. H1: µ1 ≠ µ2 b. H0: µ1 ≠ µ2 c. H1: µ1 = µ2 d. H0: µ1 = µ2 32. Dr. Simon wanted to know if IQ scores differed between male and female participants in his study. He gave 26 participants an IQ test and then examined IQ scores for gender differences. He hypothesized that there would be a statistically significant gender difference in IQ scores. Contrary to Dr. Simon's hypothesis, there were no differences in IQ scores between men and women in his study. From the results of the study, Dr. Simon must _____ the null hypothesis and _____ the research hypothesis. a. fail to reject; reject b. reject; fail to reject c. retest; reject d. reject; retest 33. If two sample means come from the exact same distribution, then the distribution of differences between means against which they will be compared should have a mean of: a. 2. b. 1. c. 0. d. –1.

Page 5

Name:

Class:

Date:

Chap_11_5e 34. Which formula for calculating the lower bound of a confidence interval is correct? a. –t(sdifference) + (MX – MY)sample b. t(smean) + (MX – MY)population c. t(sdifference) + (MX – MY)sample d. –t(smean) + (MX – MY)population 35. The fourth step in calculating a confidence interval for an independent-samples t test is to: a. look up the t statistics for the lower and upper bounds. b. draw a normal curve. c. write the percentages for the upper and lower bounds. d. convert the t statistics to raw differences between means. 36. Mehl et al. (2007) published in the journal Science the results of an extensive study of 396 men and women, comparing the number of words uttered per day by each sex. Is the population likely to be normally distributed? Why or why not? a. Yes; the population is likely to be normally distributed because the sample size is greater than 200. b. No; the population is not likely to be normally distributed because it has a small sample size. c. Yes; the population is likely to be normally distributed because the sample is greater than 30. d. We do not know whether the population is normally distributed because we do not have information on the population. 37. To report the exact p value associated with the test statistic for an independent-samples t test, we need to: a. use software to obtain the value. b. record the value from the t table under the 0.05 p level. c. compute Cohen's d. d. record the value from the t table under the 0.01 p level. 38. Which formula for calculating the upper bound of a confidence interval is correct? a. –t(sdifference) + (MX – MY)sample b. t(smean) + (MX – MY)population c. t(sdifference) + (MX – MY)sample d. –t(smean) + (MX – MY)population 39. Which of these illustrates a statistically insignificant result for an independent-samples t test? a. t(N – 1) = 1.89, p < 0.05 b. t(15) = 1.89, p < 0.05 c. t(15) = 1.89, p > 0.05 d. t(N – 2) = 1.89, p > 0.05 Copyright Macmillan Learning. Powered by Cognero.

Page 6

Name:

Class:

Date:

Chap_11_5e 40. The results of an independent-samples t test were t(18) = 4.42, p < 0.05. In this example, the sample size is: a. 16. b. 18. c. 19. d. 20. 41. The formula for the total degrees of freedom for the independent-samples t test is: a. N. b. dfX + dfY. c. N – 1. d. (N – 1)(N – 1). 42. The results of an independent-samples t test were t(18) = 4.42, p < 0.05. In this example, there are _____ degrees of freedom. a. 0.05 b. 4.42 c. 16 d. 18 43. To calculate effect size for an independent-samples t test, we use: a. Cohen's d. b. sample size. c. z score. d. standard deviation scores. 44. Which of these illustrates the APA format for reporting statistically significant results for an independentsamples t test? a. t(N – 1) = 3.69, p < 0.05 b. t(15) = 3.69, p < 0.05 c. t(15) = 3.69, p > 0.05 d. t(N – 2) = 3.69, p < 0.05 45. Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a prerecorded psychology lecture. At the end of the 20-minute lecture, students answered a 17-question quiz about the material that had just been presented. Which statistical test should the researchers use to analyze their data? a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test Copyright Macmillan Learning. Powered by Cognero.

Page 7

Name:

Class:

Date:

Chap_11_5e 46. Mehl et al. (2007) published in the journal Science the results of an extensive study of 396 men and women comparing the number of words uttered per day by each sex. Which statistical test should Mehl and colleagues use to analyze the data? a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test 47. A p level of 0.01 corresponds to a confidence interval of _____%. a. 10 b. 90 c. 95 d. 99 48. Some researchers argue that Bayesian statistics should replace: a. parametric statistics. b. hypothesis testing. c. nonparametric statistics. d. confidence intervals and effect sizes. 49. A study found statistically significant results for a hypothesis tested with an independent-samples t test. The author of the study reported her effect size for the test as 0.86. According to Cohen's convention, this effect size is considered: a. small. b. medium. c. large. d. to have no overlap. 50. A study found statistically significant results for a hypothesis tested with an independent-samples t test. The author of the study reported her effect size for the test as 0.61. According to Cohen's convention, this effect size is considered: a. small. b. medium. c. large. d. to have no overlap.

Page 8

Name:

Class:

Date:

Chap_11_5e 51. Dr. Simon wanted to know if IQ scores differed between male and female participants in his study. He gave 26 participants an IQ test and then examined the IQ scores for gender differences. He hypothesized that there would be a statistically significant gender difference in IQ scores. Contrary to Dr. Simon's hypothesis, there were no differences in IQ scores between men and women in his study. Which test was Dr. Simon MOST likely to use to test his hypothesis? a. paired-samples t test b. z test c. independent-samples t test d. single-sample t test 52. The statement "On average, older adults have the same response time as younger adults" is an example of: a. effect size. b. statistical significance. c. the null hypothesis. d. the research hypothesis. 53. When calculating effect size for an independent-samples t test, we _____ sample size. a. add b. divide by c. disregard d. multiply by 54. _____ statistics help us take into account both prior beliefs and probabilities. a. Inferential b. Nonparametric c. Parametric d. Bayesian 55. The comparison distribution for an independent-samples t test is a distribution of: a. differences between means. b. means. c. mean of difference of scores. d. scores. 56. The statement "On average, older adults have a different response time than younger adults" is an example of: a. effect size. b. statistical significance. c. the null hypothesis. d. the research hypothesis.

Page 9

Name:

Class:

Date:

Chap_11_5e 57. The first step in calculating a confidence interval for an independent-samples t test is to: a. look up the t statistics for the lower and upper bounds. b. draw a normal curve. c. write the percentages for the upper and lower bounds. d. convert the t statistics to raw score differences between means. 58. s2pooled is the symbol for: a. variance. b. pooled variance. c. standard deviation. d. pooled mean. 59. When statisticians describe pooling the variances, they mean: a. adding the standard deviations of the two samples. b. adding the variances of the two samples. c. taking the average of the two variances, accounting for sample size. d. taking the average of the two variances, disregarding sample size. 60. Which example best illustrates the APA's standards for reporting descriptive statistics for an independentsamples t test? a. On average, men scored significantly lower (M = 4.33, SD = 0.22) compared to women (M = 6.59, SD = 0.19) on reaction time. b. On average, men scored significantly lower (µ = 4.33, s = 0.22) compared to women (µ = 6.59, s = 0.19) on reaction time. c. On average, men scored significantly lower compared to women (M = 4.33, SD = 0.22 versus M = 6.59, SD = 0.19) on reaction time. d. On average, men scored significantly lower (M = 4.33, s = 0.22) compared to women (M = 6.59, s = 0.19) on reaction time. 61. A study found statistically significant results for a hypothesis tested with an independent-samples t test. The author of the study reported her effect size for the test as 0.19. According to Cohen's convention, this effect size is considered: a. small. b. medium. c. large. d. to have no overlap.

Page 10

Name:

Class:

Date:

Chap_11_5e 62. Researchers were interested in whether mindfulness meditation training decreases the number of headaches a person experiences. They randomly assigned 28 participants to a control group or a mindfulness training group, and noted the change in number of headaches each group reported from the week before training to the week after training. The dependent variable in this study is: a. training type. b. number of headaches. c. control group. d. relaxation training. 63. _____ is a weighted average of the two estimates of _____. a. Pooled variance; variance b. Variance; pooled variance c. Pooled variance; standard deviation d. Standard deviation; pooled variance 64. Researchers were interested in whether mindfulness meditation training decreases the number of headaches a person experiences. They randomly assigned 28 participants to a control group (no mindfulness meditation training) or a relaxation training group, and noted the change in number of headaches each group reported from the week before training to the week after training. The independent variable in this study is: a. training type. b. number of headaches. c. control group. d. relaxation training. 65. Researchers were interested in whether mindfulness meditation training decreases the number of headaches a person experiences. They randomly assigned 28 participants to a control group or a mindfulness training group, and noted the change in number of headaches each group reported from the week before training to the week after training. Which statistical analysis should be performed to answer the researchers' question? a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test 66. The mean of the comparison distribution for the null hypothesis of an independent-samples t test is: a. always 0. b. sometimes 0. c. always 1. d. sometimes 1.

Page 11

Name:

Class:

Date:

Chap_11_5e 67. Why is it necessary to use the pooled variance when conducting an independent-samples t test? a. We are working with two samples, and an estimate of spread based on two samples is likely to be more accurate than an estimate of spread based on a single sample. b. It is necessary to estimate the standard deviation of the two samples so as to compare the two samples to one another. c. Estimating the spread of the sample using the standard deviation increases the generalizability of results. d. Using the pooled variance helps the researcher identify skewness. 68. To calculate a confidence interval for an independent-samples t test, we use the: a. difference between means. b. difference between medians. c. mean of the first group only. d. mean of the second group only. 69. In a(n) _____, each participant is assigned to only one group so as to compare mean differences. a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test 70. In a(n) _____, one sample is used to compare a distribution of means during hypothesis testing. a. single-sample t test b. z test c. independent-samples t test d. dependent-samples t test 71. In an independent-samples t test, how is the research hypothesis written in symbols? a. H1: µ1 ≠ µ2 b. H0: µ1 ≠ µ2 c. H1: µ1 = µ2 d. H0: µ1 = µ2 72. In an independent-samples t test, the _____ is used rather than the _____. a. standard deviation; pooled variance b. pooled variance; standard deviation c. standard deviation; variance d. variance; pooled variance

Page 12

Name:

Class:

Date:

Chap_11_5e 73. A study found statistically significant results for a hypothesis tested with an independent-samples t test. The author of the study reported her effect size for the test as 0.43. What is true of the two sample means? a. The two sample means overlap by 43%. b. The two sample means likely come from the same distribution. c. The two sample means do not indicate meaningful differences between groups. d. The two sample means are 0.43 standard deviation apart. 74. The formula

is used to calculate the:

a. pooled variance estimate. b. standard deviation for an independent-samples t test. c. squared standard error. d. estimated standard deviation of the distribution of mean differences. 75. A p level of 0.05 corresponds to a confidence interval of _____%. a. 68 b. 90 c. 95 d. 99.9 76. Dr. Simon wanted to know if IQ scores differed between male and female participants in his study. He gave 26 participants an IQ test and then examined IQ scores for gender differences. He hypothesized that there would be a statistically significant gender difference in IQ scores. Contrary to Dr. Simon's hypothesis, there were no differences in IQ scores between men and women in his study. What is the independent variable in this study? a. IQ scores b. sample size of 26 c. gender d. participants 77. A study found statistically significant results for a hypothesis tested with an independent-samples t test. The author of the study reported her effect size for the test as 1.13. What is true of the two sample means? a. The two sample means overlap by 85%. b. The two sample means likely come from the same distribution. c. The two sample means do not indicate meaningful differences between groups. d. The two sample means are 1.13 standard deviations apart.

Page 13

Name:

Class:

Date:

Chap_11_5e 78. Squared standard deviation is equivalent to the: a. standard error. b. variance. c. pooled variance. d. z test. Enter the appropriate word(s) to complete the statement. 79. A researcher reports the results of an independent samples t test as t(26) = 3.18. There were _______ participants in the researcher's sample.

80. We use the pooled variance to compute the standard error in an independent-samples t test, and when calculating Cohen's d we use the _______ in the denominator.

81. The results for an independent-samples t test were t(26) = 3.18, p < 0.05. Based on the results, the researcher should _______ the null hypothesis.

82. An independent-samples t test is a hypothesis test for comparing two means from a(n) _______-groups design.

83. When estimating variance for an independent-samples t test, variance from the larger sample counts for _______ variance than does the variance from the smaller sample.

Page 14

Name:

Class:

Date:

Chap_11_5e 84. A(n) _______ is a hypothesis test for comparing two means from a between-groups design.

85. For an independent-samples t test, the total _______ equals dfX + dfY.

86. The results for an independent-samples t test were t(29) = 1.78, p > 0.05. Based on the results, the researcher should _______ the null hypothesis.

87. In an independent-samples t test, each participant is assigned to _______ group(s).

88. _______ statistics help us account for both prior beliefs and probabilities.

89. _______is a measure of effect size for an independent-samples t test.

90. Pooled variance involves taking the _______ of the two variance samples.

Page 15

Name:

Class:

Date:

Chap_11_5e 91. If there are at least _______ participants in a study, it is likely that the normality assumption is met.

92. When calculating Cohen's d for an independent-samples t test, we use the _______ in the denominator.

93. Cohen's d is a measure of _______.

94. The confidence interval for the independent-samples t test is centered on the _______.

95. Calculating the _______ involves taking the weighted average of the two variance samples.

Page 16

Name:

Class:

Date:

Chap_11_5e 96. (Table: Word Usage and Gender) Women 17,212 15,325 14,032 18,645 15,790

Men 16,322 14,636 17,045 18,873 13,071

Mehl et al. (2007) reported in the journal Science the results of an extensive study of 396 men and women, comparing the number of words uttered per day by each sex. Volunteer participants wore inconspicuous recording devices that recorded the subjects' daily word usage. Is there any validity to the notion that women talk more than men? The following fictional data produce results similar to those obtained by Mehl et al. (2007). Perform all six steps of hypothesis testing on the data to answer this question.

97. A researcher presented the findings of a study at a local conference. In the study, the researcher investigated the impact of heat stress on the performance of test takers. She hypothesized that test takers exposed to high levels of heat in a temperature-controlled room would make significantly more errors compared to test takers in a standard-temperature temperature-controlled room. Prior to hypothesis testing, the researcher conducted a power analysis and found that she needed 46 participants to have sufficient power. At the conference, the researcher reported the following results: t(44) = 2.83, p < 0.05, Cohen's d = 0.9. The researcher was quite confident in her findings and concluded that temperature affects test-taking performance and that the effect should be factored into scoring exams. Another researcher at the conference criticized the researcher's results, citing another study that found evidence to the contrary. How could the original researcher defend her results?

Page 17

Name:

Class:

Date:

Chap_11_5e 98. (Table: Faculty Salaries) Assistant Professor 76 82 111 80 60

Full Professor 156 95 137 169 105

Does the average salary of a new assistant professor differ from that of a tenured full professor? Five new assistant professors and five tenured full professors were selected at random from a large public university. Their 2018–2019 salaries (in thousands of dollars) appear in the table. (a) Calculate a 95% confidence interval and assess the research hypotheses. (b) Compute the effect size for this study and interpret its meaning.

99. (Table: Faculty Salaries) Assistant Professor 76 82 111 80 60

Full Professor 156 95 137 169 105

Does the average salary of a new assistant professor differ from that of a tenured full professor? Five new assistant professors and five tenured full professors were selected at random from a large public university. Their 2018–2019 salaries (in thousands of dollars) appear in the table. Perform all six steps of the appropriate hypothesis test on this set of data.

Page 18

Name:

Class:

Date:

Chap_11_5e 100. To find out whether talking to plants affects their health, a researcher conducts a study in which he talks to one group of five plants and does not talk to a second group of five plants. After 2 weeks, he records the number of blooms produced by each plant (the dependent measure of their health). The researcher finds that, on average, the five plants he talked to produced 4.3 blooms (SD = 2.2), while the five plants he did not talk to produced 3.2 blooms (SD = 2.0). (a) Perform an independent-samples t test to determine whether talking to the plants significantly affected their health. (b) Calculate the size of the effect of talking to plants and interpret the effect size using Cohen's conventions. (c) What do the calculations in (a) and (b) suggest about the researcher's power to detect an effect?

101. A researcher was interested in whether there are differences in life satisfaction between older adults and younger adults. Ten older adults (older than age 70) and 10 younger adults (between 20 and 30 years of age) were given a life satisfaction test. Scores on the measure ranged from 0 to 60, with high scores indicating high life satisfaction. The researcher found a statistically significant difference between the average life satisfaction of the two groups: that for the older adults was 44.3 (SD = 8.68), while that for the younger adults was 28.1 (SD = 8.66), t(18) = 4.18, p < 0.05. Calculate the size of this effect, and then use Cohen's conventions to interpret its size.

102. Some researchers argue that Bayesian statistics should replace hypothesis testing. What is the rationale for this suggestion?

103. Dietz and Henrich (2014) were interested in the impact of texting on student learning. A group of 99 college students were randomly assigned to text (N = 50) or not text (N = 49) during a prerecorded psychology lecture. At the end of the 20-minute lecture, students answered a 17-question quiz about the material that had just been presented. On average, the researchers found that students who texted during the lecture answered fewer quiz questions correctly as compared to students who had not texted during the lecture. (a) Identify the populations. (b) What is the appropriate hypothesis test for analyzing the data, and what is the comparison distribution? (c) Evaluate whether the assumptions for the hypothesis test are met in this study.

Page 19

Name:

Class:

Date:

Chap_11_5e 104. (Table: Diets) Soy Diet 2.4 1.7 2.5 1.9 2.0 3.1

Traditional Diet 1.2 0.9 1.8 1.2 0.5 1.7

An article in the journal Applied Nutritional Investigation reported the results of a comparison of two different weight-loss programs (Liao, 2007). In the study, obese participants were randomly assigned to one of two groups: (1) the soy group, a low-calorie group that ate only soy-based proteins; or (2) the traditional group, a low-calorie group that received two-thirds of their protein from animal products and one-third from plant products. One of the dependent measures collected was the amount of body fat loss as a percentage of initial body weight. Fictional data on the percent body fat loss appears in the table. These data produce results similar to those of the 2007 study (note that a smaller n than that used in the study is presented for ease of calculation). Perform all six steps of the appropriate hypothesis test on this set of data.

105. A researcher is interested in whether there are significant differences between men and women and religious preferences. In planning his hypothesis tests, the researcher identified gender as the independent variable and religious preferences (e.g., Catholic, Protestant, Jewish) as his dependent variable. Can the researcher use an independent-samples t test to test his hypothesis? Why or why not?

Page 20

Name:

Class:

Date:

Chap_11_5e 106. Imagine that researchers are interested in whether relaxation training decreases the number of headaches a person experiences. Then imagine that they randomly assign 36 patients who frequently experience headaches to a control group or a mindfulness meditation training group, and note the change in number of headaches each group reports from the week before training to the week after training. (a) Identify the populations. (b) What is the appropriate hypothesis test to use for analyzing these data, and what is the comparison distribution? (c) Evaluate whether the assumptions for the hypothesis test are met in this hypothetical study.

Page 21

Name:

Class:

Date:

Chap_11_5e Answer Key 1. False 2. False 3. False 4. True 5. True 6. True 7. True 8. False 9. True 10. False 11. False 12. True 13. True 14. False 15. False 16. True 17. d 18. c 19. b 20. a 21. a 22. d 23. c 24. c 25. c 26. d Copyright Macmillan Learning. Powered by Cognero.

Page 22

Name:

Class:

Date:

Chap_11_5e 27. c 28. c 29. c 30. a 31. d 32. a 33. c 34. a 35. d 36. d 37. a 38. c 39. c 40. d 41. b 42. d 43. a 44. b 45. c 46. c 47. d 48. b 49. c 50. b 51. c 52. c 53. c 54. d Copyright Macmillan Learning. Powered by Cognero.

Page 23

Name:

Class:

Date:

Chap_11_5e 55. a 56. d 57. b 58. b 59. c 60. a 61. a 62. b 63. a 64. a 65. c 66. a 67. a 68. a 69. c 70. a 71. a 72. b 73. d 74. d 75. c 76. c 77. d 78. b 79. 28 80. pooled standard deviation 81. reject 82. between Copyright Macmillan Learning. Powered by Cognero.

Page 24

Name:

Class:

Date:

Chap_11_5e 83. more 84. independent-samples t test 85. degrees of freedom, df 86. fail to reject 87. one, 1 88. Bayesian 89. Cohen's d 90. weighted average 91. 30, thirty 92. pooled standard deviation 93. effect size 94. difference between means 95. pooled variance 96. Step 1: The populations to be compared are (1) all men and (2) all women. The comparison distribution will be a distribution of differences between means, and the hypothesis test will be an independent-samples t test. One of the assumptions for the t test is met: The dependent variable, number of words uttered in a day, is measured on an interval scale. We do not know whether the number of words uttered is normally distributed, and the sample size is small so the sampling distribution may not be normally distributed. Finally, it does not appear that the participants were randomly selected. Therefore, we will have to be cautious when generalizing our results. Step 2: Null hypothesis: The average number of words uttered by women is less than or equal to the average number of words uttered by men: H0: µ1 ≤ µ2. Research hypothesis: The average number of words uttered by women is greater than the average number of words uttered by men: H1: µ1 > µ2. Step 3: The comparison distribution is a t distribution, with a mean of 0 and sdifference =

Here is the work: X 17,212 15,325 14,032 18,645 15,790

X–M 1011 –876 –2169 2444 –411

(X–M)2 1,022,525 767,026 4,703,693 5,974,114 168,757

Y 16,327 14,636 17,042 18,873 13,081

Y–M 335 –1356 1050 2881 –2911

(Y–M)2 112,359 1,838,194 1,102,920 8,301,313 8,472,757 Page 25

Name:

Class:

Date:

Chap_11_5e M = 16,200.80 SD = 1777.37

M = 15,991.80 SD = 2226.41

dfX = 5 – 1 = 4 dfY = 5 – 1 = 4 dftotal = 4 + 4 = 8

Step 4: The critical value using a p level of 0.05, a one-tailed hypothesis test, and the t distribution with df = 8 is 1.86. Step 5:

Page 26

Name:

Class:

Date:

Chap_11_5e Step 6: The calculated t value does not exceed the critical value. Therefore, we fail to reject the null hypothesis. We do not have any evidence that the number of words uttered by women is greater than the number of words uttered by men: t(8) = 0.164, p > 0.05. 97. Several aspects of the researcher's statistics are particularly strong. First, the researcher conducted a power analysis prior to hypothesis testing, which should have helped to protect against Type II error. Second, the researcher's sample size was greater than 30, suggesting that the data were likely to have been normally distributed. Third, the researcher found statistically significant results on the basis of an alpha of 0.05. Finally, the researcher's effect size was large, indicating meaningful differences between the groups. The researcher should highlight these factors in defending her results. 98. (a) The confidence interval is calculated as:

= – 2.306(16.48) – 50.6 = –88.603

= 2.306(16.48) – 50.6 = –12.597 The confidence interval of [–88.603, –12.597] does not include zero, indicating that there is a difference between groups, so we can reject the null hypothesis. (b) The effect size is calculated as:

This is a large effect size, which makes sense with the significant hypothesis test and such a small sample size. 99. Step 1: The populations to be compared are (1) all new assistant professors and (2) all tenured full professors. The comparison distribution will be a distribution of differences between means, and the hypothesis test will be an independent-samples t test. One of the assumptions for the t test is met: The dependent variable, salary, is Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_11_5e measured on an interval scale. However, we do not know whether the population is normally distributed, and there are not at least 30 samples. It is unlikely, however, that salary is normally distributed in the populations. Furthermore, although professors were randomly selected from the large public university, they were not randomly selected from all professors. Therefore, we will have to be cautious when generalizing the results. Step 2: Null hypothesis: The average salary of new assistant professors is the same as that of tenured full professors: H0: µ1 = µ2. Research hypothesis: The average salary of new assistant professors differs from that of tenured full professors: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a t distribution with a mean of 0 and sdifference = 16.48. Here is the work: X 76 82 111 80 60 M = 81.80 SD = 18.472

X–M –5.80 0.20 29.20 –1.80 –21.80

(X–M)2 33.64 0.04 852.64 3.24 475.24

Y 156 95 137 169 105 M = 132.40 SD = 31.887

Y–M 23.60 –37.40 4.60 36.60 –27.40

(Y–M)2 556.96 1398.76 21.16 1339.56 750.76

dfX = 5 – 1 = 4 dfY = 5 – 1 = 4 dftotal = 4 + 4 = 8

Page 28

Name:

Class:

Date:

Chap_11_5e

Step 4: The critical values using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 8 are –2.306 and 2.306. Step 5:

Step 6: The calculated t value exceeds the critical value. Therefore, we reject the null hypothesis. Our data provide evidence that the salary of assistant professors is, on average, less than that of tenured full professors: t(8) = –3.070, p < 0.05. 100. (a) Talking to the plants did not significantly affect their health: t(8) = 0.827, p > 0.05. (b) The size of the effect was Cohen's d = 0.52, which according to Cohen's conventions is a medium effect size. (c) Given that there was a medium effect size, perhaps the researcher did not have enough power to detect an effect. Increasing the sample size would improve his ability to detect an effect. 101. This is a large effect, according to Cohen's conventions. 102. Students' answers may differ, but will likely focus on the differences in probability distributions used by these two types of testing. Probabilities are based on prior beliefs or previous research in Bayesian statistics, similar to conditional probabilities. However, traditional hypothesis testing uses probability distributions based on chance using p values.

Page 29

Name:

Class:

Date:

Chap_11_5e 103. (a) One population is all college students who text during lectures, and the second population is all college students who do not text during lectures. (b) The appropriate hypothesis test for these data is an independent-samples t test, and the comparison distribution is a distribution of differences between means. (c) The dependent variable is the number of quiz questions answered correctly, which is a ratio variable; thus, this meets the requirement that the dependent variable be a scale variable. We cannot know whether the data are normally distributed, but because the sample size is greater than 30, we can be confident that the sampling distribution is likely to be normally distributed. Finally, it is unlikely that the participants for this study were randomly selected. 104. Step 1: The populations to be compared are (1) all obese people on a low-calorie soy diet and (2) all obese people on a low-calorie traditional diet. The comparison distribution will be a distribution of differences between means, and the hypothesis test will be an independent-samples t test. One of the assumptions for the t test is met: The dependent variable, percent body fat loss, is measured on an interval scale. We do not, however, know whether we met the assumption that our population is normally distributed, and our sample size is less than 30. Furthermore, it does not seem as though the participants were randomly selected. Step 2: Null hypothesis: The average percent body fat loss is the same for those on a low-calorie soy diet and those on a low-calorie traditional diet: H0: µ1 = µ2. Research hypothesis: The average percent body fat loss differs for the low-calorie soy diet and the low-calorie traditional diet: H1: µ1 ≠ µ2. Step 3: The comparison distribution is a t distribution with a mean of 0 and sdifference = 0.303. Here is the work: X 2.4 1.7 2.5 1.9 2.1 3.1 M = 2.283 SD = 0.500

X–M 0.117 –0.583 0.217 –0.383 –0.183 0.817

(X–M)2 0.014 0.340 0.047 0.147 0.034 0.667

Y 1.2 0.9 1.8 1.2 0.5 1.7 M = 1.217 SD = 0.488

Y–M –0.017 –0.317 0.583 –0.017 –0.717 0.483

(Y–M)2 0.000 0.100 0.340 0.000 0.514 0.234

Page 30

Name:

Class:

Date:

Chap_11_5e dfX = 6 – 1 = 5 dfY = 6 – 1 = 5 dftotal = 5 + 5 = 10

Step 4: The critical values using a p level of 0.05, a two-tailed hypothesis test, and the t distribution with df = 10 are –2.228 and 2.228. Step 5:

Step 6: The calculated t value exceeds the critical value. Therefore, we reject the null hypothesis. The percent body fat lost on a low-calorie soy diet (M = 2.283, SD = 0.500) is greater than that lost on a low-calorie traditional diet (M = 1.217, SD = 0.488), t(10) = 3.727, p < 0.05. 105. No; this hypothesis would not be appropriate for an independent-samples t test. An assumption that must be met for the independent-samples t test is that the dependent variable is a scale variable. The dependent variable religious preference is not a scale variable; it is a nominal variable.

Page 31

Name:

Class:

Date:

Chap_11_5e 106. (a) One population is all people who frequently experience headaches and are not undergoing mindfulness meditation training; the second population is all people who frequently experience headaches and are undergoing mindfulness meditation training. (b) The appropriate hypothesis test is an independent-samples t test, and the comparison distribution is a distribution of differences of means. (c) The assumption that the dependent variable is measured on an interval or ratio scale is met: Number of headaches is ratio data. We do not know whether the population is normally distributed, but there are at least 30 participants in the sample, so the sampling distribution is likely to be normal. Finally, it is unlikely that these participants were randomly selected from the population.

Page 32

Name:

Class:

Date:

Chap_12_5e Indicate whether the statement is true or false. 1. When the null hypothesis is true, the F ratio for analysis of variance is expected, on average, to be 0. a. True b. False 2. In the Tukey test, HSD stands for highly significant differences. a. True b. False 3. R2 is the proportion of variance in the dependent variable accounted for by the independent variable. a. True b. False 4. When the null hypothesis is true, the F ratio for analysis of variance is expected, on average, to be 1. a. True b. False 5. The F ratio measures two sources of variability: within-groups variance and between-groups variance. a. True b. False 6. The values observed on the F table when there are only two groups are the square of those on the t table. a. True b. False 7. The Bonferroni test is a less conservative post hoc test than is the Tukey HSD. a. True b. False 8. Populations that have the same variance are referred to as heteroscedastic. a. True b. False 9. Populations that have the same variance are referred to as homoscedastic. a. True b. False 10. When the F statistic is less than the critical cutoff value, additional follow-up analyses are needed. a. True b. False

Page 1

Name:

Class:

Date:

Chap_12_5e 11. As the variability of between-groups means increases, the F statistic decreases. a. True b. False 12. R2 is the proportion of variance in the independent variable accounted for by the dependent variable. a. True b. False 13. To test the assumption of homoscedasticity, you would check whether the largest variance is not more than twice the smallest variance. a. True b. False 14. The effect-size measure R2 is a less biased measure of effect size than is ω2. a. True b. False 15. R2 is a post hoc test that tells us which means are different. a. True b. False 16. Populations that have different variances are referred to as heteroscedastic. a. True b. False 17. As the number of t tests increases, the risk of a Type I error decreases. a. True b. False 18. An ANOVA is typically used when the dependent variable is nominal with more than two groups. a. True b. False 19. If between-groups variability decreases, within-groups variability necessarily decreases. a. True b. False 20. Post hoc tests are frequently carried out before the null hypothesis has been rejected in an ANOVA. a. True b. False

Page 2

Name:

Class:

Date:

Chap_12_5e 21. The effect-size measure ω2 is a less biased measure of effect size than is R2. a. True b. False 22. When the F statistic is more than the critical cutoff value, additional follow-up analyses are needed. a. True b. False 23. Populations that have different variances are referred to as homoscedastic. a. True b. False 24. Post hoc tests are frequently carried out after the null hypothesis has been rejected in an ANOVA. a. True b. False 25. The values observed on the F table when there are only two groups are the same as those on the t table. a. True b. False 26. An ANOVA is typically used when the independent variable is nominal with more than two groups. a. True b. False 27. The F statistic is more versatile than the t test. a. True b. False 28. The effect size measure ω2 is the proportion of variance in the dependent variable accounted for by the independent variable. a. True b. False 29. The Bonferroni test is a more conservative post hoc test than is the Tukey HSD. a. True b. False 30. As the variability of between-groups means increases, the F statistic increases. a. True b. False

Page 3

Name:

Class:

Date:

Chap_12_5e Indicate the answer choice that best completes the statement or answers the question. 31. As the amount of overlap decreases among sample distributions, the: a. less confounding variability there is in the groups. b. less confident we are that the samples represent true differences in the population. c. more confident we are that the samples represent true differences in the population. d. less likely it is that we will reject the null hypothesis. 32. There is a different F distribution for every: a. sample size. b. level of the independent variable. c. combination of sample size and number of samples. d. within-groups degrees of freedom. 33. MSwithin is obtained by dividing: a. SSbetween by dfbetween. b. SSbetween by SSwithin. c. dfbetween by SSwithin. d. SSwithin by dfwithin. 34. According to Cohen's conventions, an R2 of 0.05 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous 35. The F ratio is a ratio of: a. two (or more) sample means. b. two variances. c. sample means divided by sample variances. d. two sum-of-squares estimates. 36. If between-groups variance is much larger than within-groups variance, we infer that the sample means are _____ one another, and we _____ the null hypothesis. a. different from; reject b. similar to; fail to reject c. different from; fail to reject d. similar to; reject

Page 4

Name:

Class:

Date:

Chap_12_5e 37. Given a scale dependent variable and a nominal independent variable with three or more levels, we could use a(n) _____ to analyze the data. a. chi-square test b. z test c. t test d. ANOVA 38. If there are only two samples, then the F distribution is the: a. sum of the two samples. b. square of the t distribution. c. square of the z distribution. d. null hypothesis distribution for N = 2. 39. To evaluate Tukey's HSD, you would consult the _____ table. a. z b. F c. q d. R 40. If the p level is 0.05 and the null hypothesis is true, what is the probability of making a Type I error (i.e., that we will fail to reject the null hypothesis)? a. 0.05 b. 0.90 c. 0.95 d. 0.00 41. A post hoc test is warranted when the researcher: a. rejects the null hypothesis and there are more than two groups. b. fails to reject the null hypothesis in an ANOVA. c. rejects the null hypothesis when performing an independent-groups t test. d. has an a priori prediction about which group means will differ.

Page 5

Name:

Class:

Date:

Chap_12_5e 42. This table reflects the fictional results of a study examining the impact of reinforcement on four groups of participants with different psychological disorders. Table: Between-Groups Source Table Source

Between

36.95

12.32

Within

63.20

3.95

Total

100.15

The effect-size value, R2, for these data is: a. 0.37. b. 3.12. c. 0.12. d. 0.24. 43. According to conventions, an ω2 of 0.07 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous 44. This table reflects the fictional results of a study examining the impact of reinforcement on four groups of participants with different psychological disorders. Table: Between-Groups Source Table Source

Between

36.95

12.32

Within

63.20

3.95

Total

100.15

The effect-size value, ω2, for these data is: a. 0.37. b. 3.12. c. 0.12. d. 0.24.

Page 6

Name:

Class:

Date:

Chap_12_5e 45. Using the Bonferroni post hoc test, to make all possible comparisons between the means of three groups while keeping the overall p level at 0.05, what should the p level be for each comparison? a. 0.050 b. 0.001 c. 0.017 d. 0.008 46. This table reflects the fictional results of a study examining the impact of reinforcement on four groups of participants with different psychological disorders. Table: Between-Groups Source Table Source

Between

36.95

12.32

Within

63.20

3.95

Total

100.15

The F value for these data is: a. 4.11. b. 0.65. c. 3.12. d. 12.32. 47. Which of the following is NOT an assumption of an ANOVA? a. random selection b. independent variable has only three levels c. normally distributed population d. samples come from populations with the same variances 48. In which situation would it NOT be appropriate for a researcher to use a one-way ANOVA in analyzing the data? a. A researcher is interested in a longitudinal study that follows a group of kindergarten children through high school and assesses their body mass index at four time points. b. A researcher classifies a group of college students on the basis of their political leanings as liberal, conservative, libertarian, or green. c. A researcher is interested in the effects of distraction on driving performance and randomly assigns participants to one of three distraction groups. d. A researcher is interested in whether a sensitivity training class changes attitudes toward minority populations and assesses these attitudes before, immediately after, and 6 months after the sensitivity training class. Copyright Macmillan Learning. Powered by Cognero.

Page 7

Name:

Class:

Date:

Chap_12_5e 49. An F statistic calculated on 2 and 38 degrees of freedom equals 2.90. Which decision should be made about a hypothesis tested at the p = 0.05 level? a. Fail to reject the null hypothesis and conclude there are differences among the groups. b. Fail to reject the null hypothesis and conclude there are no significant differences among the groups. c. Reject the null hypothesis and conclude there are differences among the groups. d. Reject the null hypothesis and conclude there are no significant differences among the groups. 50. The proportion of variance in the dependent variable that is accounted for by the independent variable is _____. a. t b. Cohen's d c. R2 d. z 51. If the cutoff q statistic is 4.69, with a p level of 0.05, which of these would be significant? a. an HSD of 4.8 b. a group mean difference of 4.7 c. an F of 5.2 d. an HSD of 4.6 52. When a researcher violates the assumption that samples are randomly selected, this affects the researcher's ability to: a. generalize the results of a study beyond the sample. b. differentiate between-groups variance and within-groups variance. c. meet the homogeneity of variance assumption. d. reject the null hypothesis. 53. Which of the following is the formula for calculating ω2? a. b. c. d.

Page 8

Name:

Class:

Date:

Chap_12_5e 54. The F statistic increases when: a. within-groups variance decreases. b. between-groups variance decreases. c. between-groups variance increases. d. within-groups variance decreases and between-groups variance increases. 55. Within-groups variance reflects the: a. amount of difference between means that we would expect to occur as a result of experimental manipulation. b. average of the sample variances. c. mean averages. d. total amount of difference we would expect to occur, given both chance and experimental manipulation. 56. In a z test and a t test, when there is no difference between groups, the test statistic is equal to _____; in ANOVA, when there are no differences among groups, the test statistic is equal to _____. a. 0; 1 b. 1; 0 c. 1; 2 d. 2; 1 57. R2, the effect size for ANOVA, is calculated as a ratio of: a. total sum of squares to within-groups sum of squares. b. within-groups mean squares to between-groups mean squares. c. total mean squares to total sum of squares. d. between-groups sums of squares to total sum of squares. 58. Within-groups degrees of freedom is calculated by: a. subtracting 1 from the total number of groups in the study. b. multiplying the number of subjects by the number of conditions in the study and then subtracting 1. c. subtracting 1 from the total number of subjects in the study. d. for each condition, subtracting 1 from the number of subjects in that group and then adding together the totals for all the groups. 59. Changes in between-groups variance result in changes in the _____ of the F ratio. a. denominator b. numerator c. sign d. significance

Page 9

Name:

Class:

Date:

Chap_12_5e 60. The assumptions of ANOVA are that samples are selected: a. randomly and that the samples come from populations with unequal variances. b. randomly, the population distribution is normal, and the samples come from populations with equal variances. c. randomly and the population distribution is normal. d. from populations that are heteroscedastic and normally distributed. 61. Interested in the effects of different kinds of instruction on video game performance, Venera asks 36 college freshmen to each play one hour of Ratchet and Clank. Participants are randomly assigned to one of three instruction groups: (1) complete the tasks as quickly as possible, (2) conserve as much health as possible (i.e., play more carefully), or (3) find gold bolts (worth lots of money in equipment and ammunition). If Venera averages the scores for each instruction group and then compares them, any differences in the means of the instruction groups reflect: a. individual differences in hand–eye coordination. b. inherent differences in the ability of the college freshmen to play video games. c. within-groups variance. d. between-groups variance. 62. Homoscedasticity can be tested for by: a. making sure that the within-groups variance is no more than two times the between-groups variance. b. making sure that the largest sample variance is no more than two times the smallest variance. c. using ANOVA. d. making sure that the largest sample variance is no more than five times the smallest sample variance. 63. When the F statistic is significant: a. the values of the means that are closest together are the only ones that are significant. b. the values of the means that are furthest apart are the only ones that are significant. c. we do not know which specific means differ significantly. d. we fail to reject the null hypothesis. 64. Micayla is interested in the effects of alcohol consumption on style of play while playing the video game Fortnite. She asks a group of 15 college students to play Fortnite under one of the following conditions: no alcohol, 1 ounce of alcohol, 2 ounces of alcohol, or 3 ounces of alcohol. Micayla measures on a scale of 1 to 10 how prosocial the students' style of play is. After collecting the data, Micayla performs an ANOVA and finds that she can reject the null hypothesis. On this basis, what does Micayla know? a. Greater levels of alcohol consumption are associated with a more prosocial style of play. b. There is a difference among the groups somewhere, but she does not know where. c. Moderate alcohol consumption is associated with prosocial play, but greater alcohol consumption is associated with an antisocial style of play. d. Drinking any amount of alcohol affects the style of play. Copyright Macmillan Learning. Powered by Cognero.

Page 10

Name:

Class:

Date:

Chap_12_5e 65. According to Cohen's conventions, an R2 of 0.15 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous 66. The critical cutoffs for a two-tailed Tukey HSD test comparing three group means with 14 within-groups degrees of freedom are: a. –3.03 and 3.08. b. –3.70 and 3.70. c. –4.11 and 4.11. d. –4.89 and 4.89. 67. An F statistic calculated on 3 and 36 degrees of freedom equals 2.91. Which decision should be made about a hypothesis tested at the p = 0.05 level? a. Fail to reject the null hypothesis and conclude there are differences among the groups. b. Fail to reject the null hypothesis and conclude there are no significant differences among the groups. c. Reject the null hypothesis and conclude there are differences among the groups. d. Reject the null hypothesis and conclude there are no significant differences among the groups. 68. If there are only two samples and sample size is infinity, then the F distribution is the: a. sum of the two samples. b. t distribution. c. null hypothesis distribution for N = 2. d. square of the z distribution. 69. When the p level is.0.05 and the null hypothesis is true, the probability of making a Type I error when performing a single hypothesis test is _____, but if three hypothesis tests are performed, then the probability of NOT making a Type I error is _____, which means that the probability of making a Type I error at least once in these three hypothesis tests is _____. a. 0.05; 0.86; 0.14 b. 0.95; 0.85; 0.15 c. 0.05; 0.95; 0.15 d. 0.00; 0.95; 0.85 70. Changes in within-groups variance result in changes in the _____ of the F ratio. a. denominator b. numerator c. sign d. significance Copyright Macmillan Learning. Powered by Cognero.

Page 11

Name:

Class:

Date:

Chap_12_5e 71. With two groups, the square root of F is equal to: a. t. b. 0.0. c. 1.0. d. 2t. 72. The Bonferroni, Scheffé, and Tukey tests are all examples of: a. hypothesis tests. b. post hoc tests. c. effect-size statistics. d. confidence intervals. 73. In which situation would a researcher need to use a between-groups ANOVA to analyze the data? a. A researcher is interested in a longitudinal study that follows a group of kindergarten children through high school and assesses their body mass index at four time points. b. A researcher classifies a group of college students on the basis of their political leanings as liberal, conservative, libertarian, or green. c. A researcher is interested in the effects of distraction on driving performance and randomly assigns participants to one of three distraction groups. d. A researcher is interested in whether a sensitivity training class changes attitudes toward minority populations and assesses these attitudes before and after the sensitivity training class. 74. Between-groups degrees of freedom is calculated by: a. subtracting 1 from the total number of groups in the study. b. multiplying the number of subjects by the number of conditions in the study and then subtracting 1. c. subtracting 1 from the total number of subjects in the study. d. subtracting 1 from the number of subjects within each group and then adding those numbers together. 75. Cohen's d cannot be used as a measure of effect size for ANOVA because: a. there is no effect size measure for ANOVA. b. its values no longer have meaning when more than two groups are compared. c. it can compare only two means. d. it can only compute effect size for the post hoc tests.

Page 12

Name:

Class:

Date:

Chap_12_5e 76. For the data in the source table, calculate the effect size R2. Source Between-groups Within-groups Total

SS 10.1 17.6 27.7

df 2 12 14

MS 5.05 1.47

a. 3.44 b. 0.36 c. 0.25 d. 0.57 77. The z, t, and F distributions have something in common—the denominator of the test statistic: a. contains a measure of difference among group means. b. contains a measure of difference within the various groups. c. is a squared number. d. represents what would be expected if the null hypothesis were true. 78. A one-way ANOVA includes: a. one nominal dependent variable with more than two levels and a scale independent variable. b. one nominal independent variable with more than two levels and a scale dependent variable. c. one nominal dependent variable with more than three levels and a scale independent variable. d. one nominal independent variable with more than three levels and a scale dependent variable. 79. The z, t, and F distributions have something in common—the numerator of the test statistic: a. contains a measure of difference among group means. b. contains a measure of difference within the various groups. c. is a squared number. d. represents what would be expected to happen by chance. 80. When computing Tukey HSD with samples of different sizes, what additional step must be performed? a. Calculate a weighted sample size called a harmonic mean. b. Calculate HSD for each pair of means. c. Compare the means of all groups using a Bonferroni test first. d. Compute R2.

Page 13

Name:

Class:

Date:

Chap_12_5e 81. The symbol for variance in ANOVA is: a. s2, which stands for standard deviation squared. b. MS, which stands for mean squared. c. SD2, which stands for standard deviation squared. d. MS2, which stands for means squared. 82. For the data in the source table, calculate the effect size ω2. Source Between-groups Within-groups Total

SS 10.1 17.6 27.7

df 2 12 14

MS 5.05 1.47

a. 3.44 b. 0.36 c. 0.25 d. 0.57 83. The grand mean is the: a. mean of every score in the study, regardless of the sample. b. average of each of the group means. c. mean of all the scores in the study squared. d. average difference of each of the group means. 84. In which situation would a researcher need to use a within-groups ANOVA to analyze the data? a. A researcher is interested in a longitudinal study that follows a group of kindergarten children through high school and assesses their body mass index at four time points. b. A researcher classifies a group of college students on the basis of their political leanings as liberal, conservative, libertarian, or green. c. A researcher is interested in the effects of distraction on driving performance and randomly assigns participants to one of three distraction groups. d. A researcher is interested in whether a sensitivity training class changes attitudes toward minority populations and assesses these attitudes before and after the sensitivity training class. 85. A measure of the differences among group means is: a. the z score. b. between-groups variance. c. within-groups variance. d. the proportionate reduction in error. Copyright Macmillan Learning. Powered by Cognero.

Page 14

Name:

Class:

Date:

Chap_12_5e 86. F is obtained by dividing _____ by _____. a. SSbetween; dfbetween b. MSbetween; MSwithin c. dfbetween; SSbetween d. SSbetween; SSwithin 87. Using the Bonferroni post hoc test, to make all possible comparisons between the means of four groups while keeping the overall p level at 0.05, what should the p level be for each comparison? a. 0.050 b. 0.001 c. 0.013 d. 0.008 88. According to Cohen's conventions, an R2 of 0.01 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous 89. Using the Bonferroni post hoc test, the overall probability of making a Type I error can be kept at a p level of 0.05 by: a. dividing 0.05 by the number of comparisons you plan to make and using the result as the p level for each comparison. b. dividing 0.05 by 2 and using the result as the p level for each comparison. c. determining the amount of difference you should have between each of your means based on a p level of 0.01. d. finding the honestly significant difference. 90. When the independent variable has more than two levels, a researcher cannot just run multiple t tests, because as more statistical tests are run: a. it becomes increasingly more difficult to reject the null hypothesis. b. the probability of making a Type I error in one of the tests increases. c. degrees of freedom are lost. d. the probability of making a Type III error in one of the tests increases.

Page 15

Name:

Class:

Date:

Chap_12_5e 91. According to conventions, an ω2 of 0.02 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous 92. If the p level is 0.05 and the null hypothesis is true, what is the probability of NOT making a Type I error (i.e., we will correctly fail to reject the null hypothesis)? a. 0.05 b. 0.90 c. 0.95 d. 0.00 93. Because of concerns about accurate measures of effect size, what measure has been proposed as an unbiased alternative to R2? a. η2 b. Cohen's d c. ω2 d. λ 94. The F ratio is calculated by: a. dividing a measure of within-groups variability by a measure of between-groups variability. b. averaging the results from the t tests for all possible two-group comparisons. c. dividing a measure of between-groups variability by a measure of within-groups variability. d. squaring all the scores. 95. MSbetween is obtained by dividing: a. SSbetween by dfbetween. b. SSbetween by SSwithin. c. dfbetween by SSbetween. d. SSwithin by SSwithin. 96. According to conventions, an ω2 of 0.17 is considered to be a(n) _____ effect size. a. small b. medium c. large d. erroneous

Page 16

Name:

Class:

Date:

Chap_12_5e 97. A measure of variability that is essentially an average of the sample variances is: a. the z score. b. between-groups variance. c. within-groups variance. d. the proportionate reduction in error. 98. Interested in the effects of different kinds of instruction on video game performance, Venera asks 36 college freshmen to each play one hour of Ratchet and Clank. Participants are randomly assigned to one of three instruction groups: (1) complete the tasks as quickly as possible, (2) conserve as much health as possible (i.e., play more carefully), or (3) find gold bolts (worth lots of money in equipment and ammunition). Obviously, even in a single instruction group, not all players will obtain the same final score. These differences in an instruction group reflect: a. between-groups variance. b. within-groups variance. c. effects of instruction. d. effects of confounding variables. 99. In Tukey's HSD post hoc test, the HSD stands for: a. honorary standardized description. b. honorary standardized difference. c. honestly significant difference. d. honestly standardized difference. 100. If an F statistic is negative, which of these statements is true? a. The within-groups variance exceeds the between-groups variance. b. The difference among the group means is greater than what would have occurred by chance. c. There has been a calculation error. d. The difference among the group means is less than what would have occurred by chance. Enter the appropriate word(s) to complete the statement. 101. The values on the F table under two samples are the same as those on the t table except that they are _______.

Page 17

Name:

Class:

Date:

Chap_12_5e 102. Using a p level of 0.05, the critical value for an ANOVA with 4 and 24 degrees of freedom is _______.

103. Both R2 and ω2 for an ANOVA provide the proportion of _______ in the dependent variable that is accounted for by the independent variable.

104. Populations that have different variances are _______.

105. Using a p level of 0.01, the critical value for an ANOVA with 4 and 24 degrees of freedom is _______.

106. If you were to make all possible two-group comparisons using t tests with five conditions of the independent variable, you would run _______ t tests.

107. If you were to make all possible two-group comparisons using t tests with six conditions of the independent variable, you would run _______ t tests.

Page 18

Name:

Class:

Date:

Chap_12_5e 108. The _______ is the mean of every score in a study, regardless of which sample the score came from.

109. If dfwithin = 20 and there are four groups in the study with an equal number of participants in each group, there are _______ participants in each group.

110. If there are more than two levels of the independent variable and different participants were exposed to each level of the independent variable, then the researcher should analyze the data with a(n) _______ ANOVA.

111. A researcher compares different groups of psychiatric patients on some dependent measure. The variability among psychiatric patients as a whole is known as _______.

112. If dfbetween = 2, then there are _______ groups in the study.

113. If dfwithin = 24 and there are three groups in the study with an equal number of participants in each group, there are _______ participants in each group.

Page 19

Name:

Class:

Date:

Chap_12_5e 114. The Tukey HSD test is sometimes called the _______ because of the statistic on which it is based.

115. If there are more than two levels of the independent variable and the same participants were exposed to every level of the independent variable, then the researcher should analyze the data with a(n) _______ ANOVA.

116. To make comparisons between three or more groups for differences on a dependent variable, you would use a(n) _______.

117. The hypothesis test used when there is one independent variable with more than two nominal levels and a scale-dependent variable is the _______ ANOVA.

118. Homoscedastic populations have _______variance(s).

119. A researcher compares different groups of psychiatric patients on some dependent measure. The variability among depressed patients as a group is known as _______.

Page 20

Name:

Class:

Date:

Chap_12_5e 120. Heteroscedastic populations have _______ variance(s).

121. If dfwithin = 18 and there are three groups in the study with an equal number of participants in each group, the total number of participants in the study (across all conditions) is N = _______.

122. If you were to make all possible two-group comparisons using t tests with four conditions of the independent variable, you would run _______ t tests.

123. Populations that have the same variance are _______.

124. If dfwithin = 24 and there are four groups in the study with an equal number of participants in each group, the total number of participants in the study (across all conditions) is N = _______.

125. If dfbetween = 4, then there are _______ groups in the study.

Page 21

Name:

Class:

Date:

Chap_12_5e 126. _______ tests allow the statistician to make multiple comparisons among several means.

127. In ANOVA, both R2 and ω2 are measures of _______.

128. (Table: Faculty Salaries and Academic Department) Do university faculty members' salaries depend on the department in which they teach? The table data depict the 2018–2019 academic year salaries (in thousands of dollars) randomly selected from all faculty in their respective departments at a large public university. Perform the six steps of hypothesis testing on this set of data, presenting your final calculations in a source table. Table: Faculty Salaries and Academic Departments Psychology

Chemistry

English

125.1 103.6 134.8 111.4 162.2

173.8 110.5 121.5 105 117.3

77.4 111.8 100.7 78.6 145.4

Computer Science 204.1 233.8 176.8 153.1 196.4

Page 22

Name:

Class:

Date:

Chap_12_5e 129. (Table: Subsequent Memory) In an investigation of encoding conditions on subsequent memory for a list of words, Luo, Hendricks, and Craik (2007) assigned participants to one of three encoding conditions: Some participants saw a list of nouns that they were to remember (word condition), others saw the words accompanied by pictures of the objects (picture condition), and (3) still others saw the words and heard sound effects matching the objects (sound-effects condition). The researchers measured the proportion of words participants remembered correctly in a later recognition test. The fictional data in the table produce results similar to those of the original study. Use these data to perform the six steps of hypothesis testing. Table: Subsequent Memory Word Alone

Word + Picture

Word + Sound Effect

0.44 0.63 0.51

0.77 0.63 0.74

0.89 0.74 0.78

0.61

0.87

130. (Table: Music and Drinking) A social psychologist is interested in whether the type of music a college student is listening to has any effect on the number of beers the student drinks. Fifteen college students were randomly assigned to one of four listening conditions, and each student's beer consumption was covertly recorded during trips to four different taverns that played the type of music to which the student was assigned. The data are shown in the table. List each of the assumptions for performing an ANOVA, and for each assumption evaluate whether the described study meets the assumption. Table: Music and Drinking Rock 'n' Roll 5 3 5 3 4

Hip Hop 7 4 6 4 5

Country 5 4 7 6 9

Jazz 2 4 1 2 3

Page 23

Name:

Class:

Date:

Chap_12_5e 131. (Table: Faculty Salaries) The table depicts data for the 2018–2019 academic year salaries (in thousands of dollars) for university faculty randomly selected from all faculty in their respective departments at a large public university. List each of the assumptions for performing an ANOVA. For each assumption, evaluate whether the described set of data meets the assumption. Table: Faculty Salaries Psychology 125.1 103.6 134.8 111.4 162.2

Chemistry 173.8 110.5 121.5 105 117.3

English 77.4 111.8 100.7 78.6 145.4

Computer Science 204.1 233.8 176.8 153.1 196.4

132. Explain the logic of the analysis of variance, paying special attention to how the variance is partitioned (i.e., how variance components are divided up). Be sure to mention how a procedure that analyzes variance can be used to test a hypothesis about population means.

133. The expected value for an F ratio is equal to 1.00 when there is no treatment effect. Why?

134. Why do researchers have to make a decision about whether to perform a two-tailed or one-tailed hypothesis test for t tests but not for ANOVA?

Page 24

Name:

Class:

Date:

Chap_12_5e 135. (Table: Herbal Remedies) A researcher is interested in whether herbal remedies are effective in relieving allergies, and if so, which ones are most effective. The researcher takes a group of 20 patients with allergies and randomly assigns each one to receive herbal tea, a homeopathic administration of allergens, a traditional antihistamine, or a placebo pill. The dependent measure is the number of allergy complaints by patients during weeks 2 and 3 of the treatments. Perform the six steps of hypothesis testing on the set of fictional data in the table. Your final calculations should be presented in a source table. Table: Herbal Remedies Herbal Tea 2 4 0 2 3

Homeopathy 3 2 2 1 3

Antihistamine 2 1 0 3 4

Placebo 5 3 6 4 7

136. (Table: Smartphones) In an investigation of the impact on the mere presence of one's smartphone on subsequent memory and fluid intelligence, Ward, Duke, Gneezy, and Bos (2017) assigned participants to one of three conditions: Participants were asked to place their phones in view on their desks (desk condition), out of sight in their pocket or bag (pocket/bag condition), or in a separate rom (other-room condition). For the assessment of fluid intelligence, the researchers measured the number of matrices solved from a 10-item subset of Raven's Progressive Matrices. The fictional data in the table produce results similar to those of the original study. Use these data to perform the six steps of hypothesis testing. Table: Smartphones Phone on Desk 6 4 5 6 5

Phone in Pocket/Bag 8 7 6 9 8

Phone in Other Room 9 10 8 8 10

Page 25

Name:

Class:

Date:

Chap_12_5e 137. (Table: Beer and Attractiveness & Table: Beer and Attractiveness, Group Means) Imagine that you have performed a "beer goggling" experiment: You randomly assigned 21 participants to drink either 0, 1, or 3 beers, and then had all participants rate the attractiveness of the same fellow college student on a scale from 0 to 50, with higher scores indicating greater perceived attractiveness. The ANOVA source table and the group means are provided in the tables. Use this information to answer the questions. Table: Beer and Attractiveness Source Between Within Total

SS 520.096 500.854 1020.950

df 2 18 20

MS 260.048 27.825

F 9.346

Table: Beer and Attractiveness, Group Means Group 0 beer 1 beer 3 beers

M 21.271 28.571 35.824

(a) Compute both measures of effect size for this ANOVA and interpret their meaning. (b) Is it appropriate to run a post hoc analysis for this experiment? Why or why not? (c) If it is appropriate to run a post hoc analysis for this experiment, perform the analysis using Tukey's HSD with a p level of 0.05. Clearly indicate which groups differ significantly from one another. (d) If this post hoc analysis were performed as Bonferroni tests, what kind of statistical test would we compute for each comparison and what p level would each comparison be compared against?

Page 26

Name:

Class:

Date:

Chap_12_5e Answer Key 1. False 2. False 3. True 4. True 5. True 6. True 7. False 8. False 9. True 10. False 11. False 12. False 13. True 14. False 15. False 16. True 17. False 18. False 19. False 20. False 21. True 22. True 23. False 24. True 25. False 26. True Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_12_5e 27. True 28. True 29. True 30. True 31. c 32. c 33. d 34. b 35. b 36. a 37. d 38. b 39. c 40. a 41. a 42. a 43. b 44. d 45. c 46. c 47. b 48. b 49. b 50. c 51. a 52. a 53. d 54. d Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_12_5e 55. b 56. a 57. d 58. d 59. b 60. b 61. d 62. b 63. c 64. b 65. c 66. b 67. c 68. d 69. a 70. a 71. a 72. b 73. c 74. a 75. c 76. b 77. b 78. b 79. a 80. a 81. b 82. c Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_12_5e 83. a 84. a 85. b 86. b 87. d 88. a 89. a 90. b 91. a 92. c 93. c 94. c 95. a 96. c 97. c 98. b 99. c 100. c 101. squared 102. 2.78 103. variance 104. heteroscedastic 105. 4.22 106. 10, ten 107. 15, fifteen 108. grand mean 109. six, 6 110. between-groups Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_12_5e 111. between-groups variability 112. three, 3 113. nine, 9 114. q test 115. within-groups 116. ANOVA 117. one-way 118. the same, similar 119. within-groups variability 120. different 121. 21 122. six, 6 123. homoscedastic 124. 28 125. five, 5 126. Post hoc 127. effect size

Page 31

Name:

Class:

Date:

Chap_12_5e 128. Step 1: The four populations are (1) all psychology faculty, (2) all chemistry faculty, (3) all English faculty, and (4) all computer science faculty. The comparison distribution will be an F distribution and the hypothesis test will be a one-way between-groups ANOVA. The participants were randomly selected from only one university, so we should be cautious when generalizing to all university faculty. It is likely that the population distributions of salary are not normal. Finally, the homogeneity of variance assumption is not violated because the largest variance, s2 = 913.43, is not more than two times the smallest variance s2 = 523.38. Step 2: Null hypothesis: Salary does not differ depending on faculty members' department Research hypothesis: The faculty members' salaries differ depending on their department. Step 3: The comparison distribution is an F distribution with df = (3, 16). Step 4: The critical value is 3.24. Step 5: (Table: Faculty Salaries and Academic Departments) The answer should look like the table. Table: Faculty Salaries and Academic Departments Source Between Within Total

SS 22,551.43 11,939.54 34,490.97

df 3 16 19

MS 7517.17 746.22

F 10.07

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. There is evidence that faculty members' salaries differ depending on the department in which they teach, F (3, 16) = 10.07, p < 0.05.

Page 32

Name:

Class:

Date:

Chap_12_5e 129. Step 1: The three populations are (1) all people memorizing words alone, (2) all people memorizing words with pictures, and (3) all people memorizing words with sound effects. The comparison distribution will be an F distribution, and the hypothesis test will be a one-way between-groups ANOVA. The participants were not randomly selected, so we must generalize with caution. The sample distributions do not suggest skew in the population. But because N is small, it is unclear whether the distribution of the population is normal. Finally, the homogeneity of variance assumption is supported because the largest variance (0.008 for word alone) is less than two times greater than the smallest variance (0.005 for word + sound effect). We will proceed with some caution. Step 2: Null hypothesis: Memory performance does not depend on the type of encoding condition: H0: µ1 = µ2= µ3. Research hypothesis: Memory performance differs as a function of the encoding condition. Step 3: The comparison distribution is an F distribution with df = (2, 9). Step 4: The critical value is 4.26. Step 5: (Table: Subsequent Memory, Answer) The answer should look like the table. Table: Subsequent Memory, Answer Source Between Within Total

SS 0.149 0.058 0.207

df 2 9 11

MS 0.075 0.006

F 12.50

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. Memory performance does differ as a function of the encoding condition, F(2, 9) = 12.50, p < 0.05.

Page 33

Name:

Class:

Date:

Chap_12_5e 130. The first assumption is that the participants are randomly selected from the population. It is unlikely that these participants were randomly selected from the population of all college students. The second assumption is that the dependent variable is normally distributed in the population. The data themselves do not appear to be badly skewed. But the sample size is small, so the sampling distribution may not be normal. The final assumption is that the group variances are homoscedastic. The variance for rock 'n' roll is s2 = 1.00; for hip hop, it is s2 = 1.70; for country, it is s2 = 3.70; and for jazz, it is s2 = 1.30. Because the sample sizes are equal and the largest variance is more than two times the smallest variance, the homogeneity of variance assumption may be violated and we should proceed with caution. A sample computation of s2 is included here for the rock 'n' roll data: X 5 3 5 3 4

(X – M) 1 –1 1 –1 0

(X – M)2 1 1 1 1 0

Σ(X – M)2 = 4.0 s2 = 4.0/(N – 1) = 1.00

Page 34

Name:

Class:

Date:

Chap_12_5e 131. The first assumption is that the participants are randomly selected from the population. These faculty were randomly selected from all faculty at a large public university. As long as this university is the target population, the first assumption is met. However, if we wish to draw conclusions about all university faculty, we do not have random sampling and, therefore, should be cautious in our generalizations. The second assumption is that the dependent variable is normally distributed in the population. It is unlikely that a dependent variable such as salary is normally distributed, and the data themselves suggest some skew. Thus, the second assumption may be violated. The final assumption is that the group variances are homoscedastic. The variance for psychology faculty is s2 = 523.382, that for chemistry faculty is s2 = 765.327, that for English faculty is s2 = 782.742, and that for computer science faculty is s2 = 913.433. Given equal sample sizes and the fact that the largest variance is less than two times the smallest variance, the homogeneity of variance assumption is met. A sample computation of s2 is included here for psychology faculty: X (X – M) 125.1 –2.32 103.6 –23.82 134.8 7.38 111.4 –16.02 162.2 34.78 Σ(X – M)2 = 2093.528

(X – M)2 5.3824 567.3924 54.4644 256.6404 1209.6484

s2 = 2093.528/(N – 1) = 523.382 132. In ANOVA, a measure of between-groups variability is divided from a measure of within-groups variability. The within-groups variability is calculated by looking at the variability within but not across samples. Therefore, it is not contaminated with the treatment effect, and it is an estimate of the amount of variability expected by chance. To calculate the between-groups variability, the grand mean (mean of all our data) is subtracted from each group mean. This result gives a measure of the differences among the groups, as this figure will be larger as the difference between the means increases. This between-groups variability contains variability due to both the treatment effect and the variability expected to occur by chance. By dividing between-groups variability (treatment effect + chance) by within-groups variability (chance), we obtain a picture of whether the variability due to the treatment is greater than would be expected by chance. 133. The F ratio is calculated by dividing between-groups variance by within-groups variance. Within-groups variance is the amount of variability we would expect to occur by chance. Thus, if chance is the only contributor to the between-groups variance (i.e., there is no treatment effect), then we would expect the between-groups variance to equal the within-groups variance, so the F ratio would be equal to 1.00. 134. With a t test there are only two means, and the t test tells us whether they are significantly different. In essence, we have the option to look for any significant difference with the t test or for only differences in a certain direction. Such decisions do not need to be made in ANOVA, because ANOVA is automatically a two-tailed test. All ANOVA can tell us is that there is a difference among group means somewhere, but we cannot use ANOVA to figure out exactly where the difference is.

Page 35

Name:

Class:

Date:

Chap_12_5e 135. Step 1: The four populations are (1) allergy patients who drink herbal tea, (2) allergy patients who receive homeopathic medicine, (3) allergy patients who take an antihistamine, and (4) allergy patients who take a placebo pill. The comparison distribution will be an F distribution, and the hypothesis test will be a one-way betweengroups ANOVA. The participants were not randomly selected so we must generalize with caution. It is unclear whether the underlying population distributions are normal. Finally, the homogeneity of variance assumption is violated because the largest variance s2 = 2.50 is more than two times the smallest variance s2 = 0.70. Step 2: Null hypothesis: The number of allergy complaints does not differ depending on treatment: H0: µ1 = µ2= µ3 = µ4. Research hypothesis: The number of allergy complaints differs depending on the treatment condition. Step 3: The comparison distribution is an F distribution with df = (3, 16). Step 4: The critical value is 3.24. Step 5: (Table: Herbal Remedies) The answer should look like the table. Table: Herbal Remedies Source Between Within Total

SS 30.95 31.60 62.55

df 3 16 19

MS 10.32 1.98

F 5.22

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. The number of allergy complaints did depend on the type of treatment received, F(3, 16) = 5.22, p < 0.05.

Page 36

Name:

Class:

Date:

Chap_12_5e 136. Step 1: The three populations are (1) all people solving problems with their phones on their desks, (2) all people solving problems with their phones in their pockets or bags, and (3) all people solving problems with their phones in another room. The comparison distribution will be an F distribution, and the hypothesis test will be a one-way between-groups ANOVA. The participants were not randomly selected, so we must generalize with caution. The sample distributions do not suggest skew in the population. But because N is small, it is unclear whether the distribution of the population is normal. Finally, the homogeneity of variance assumption is reasonably supported because the largest variance (1.30 for pocket/bag) is less than two times greater than the smallest variance (0.70 for desk). We will proceed with some caution. Step 2: Null hypothesis: Problem-solving performance does not depend on the smartphone location condition: H0: µ1 = µ2= µ3. Research hypothesis: Problem-solving performance differs as a function of the smartphone location condition. Step 3: The comparison distribution is an F distribution with df = (2, 12). Step 4: The critical value is 3.89. Step 5: (Table: Smartphones, Answer) The answer should look like the table. Table: Smartphones, Answer Source Between Within Total

SS 36.933 12.000 48.933

df 2 12 14

MS 18.467 1.000

F 18.467

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. Problem-solving performance does differ as a function of the smartphone location condition, F(2, 12) = 18.467, p < 0.05. 137. (a) R2 = SSbetween/SStotal = 520.096/1020.950 = 0.509; ω2 = [SSbetween – ((dfbetween)(MSwithin)/(SStotal + MSwithin)] = [520.096 – (2)( 27.825)/(1020.950 + 27.825) = 464.446/1048.775 = 0.443; both measures indicate a large effect size. (b) It is appropriate to conduct post hoc tests because the ANOVA is significant. Also, because there are three groups, we need to find out where the significant differences are. (c) The critical value of the q statistic is 3.61, and sM = 1.994. 0 versus1 beer: 3.66 0 versus 3 beers: 7.30 1 versus 3 beers: 3.64 All levels of beer exceed the critical value of the q statistic. Therefore, all the groups differ significantly. Attractiveness ratings increase as the number of beers consumed increases. (d) When performing Bonferroni tests, we compute independent-samples t tests for each comparison. There are three groups in this study, so we would need three post hoc tests, as we did in (c). To compute the p level for each comparison test, we would divide 0.05 by the number of comparisons, 3, to get a p level of 0.017 for each individual post hoc comparison. Copyright Macmillan Learning. Powered by Cognero.

Page 37

Name:

Class:

Date:

Chap_13_5e Indicate whether the statement is true or false. 1. A between-groups hypothesis test is preferred over a within-groups hypothesis test. a. True b. False 2. Effect sizes help us determine whether the differences observed are large enough to matter in a within-groups ANOVA. a. True b. False 3. The effect size measure for the one-way within-groups ANOVA is R2. a. True b. False 4. Because the same participants are used in each condition, post hoc tests are not necessary when conducting a within-groups ANOVA. a. True b. False 5. The Tukey HSD test helps us determine which means are statistically significantly different from each other in a within-groups ANOVA. a. True b. False 6. Effect sizes help us determine which means are statistically significantly different from each other in a withingroups ANOVA. a. True b. False 7. When conducting the one-way within-groups ANOVA, two F statistics are calculated, but only the between-groups F statistic is of interest. a. True b. False 8. When conducting the one-way within-groups ANOVA, two F statistics are calculated, but only one is of primary interest. a. True b. False

Page 1

Name:

Class:

Date:

Chap_13_5e 9. When conducting the one-way within-groups ANOVA, two F statistics are calculated, but only the subjects F statistic is of interest. a. True b. False 10. When calculating effect size for one-way within-groups ANOVA, variability between means is considered relative to total variability after removing the variability that can be accounted for by the participants. a. True b. False 11. Order effects likely have been addressed if counterbalancing has occurred in the research. a. True b. False 12. When calculating effect size for one-way within-groups ANOVA, variability between means is considered relative to total variability. a. True b. False 13. A WEIRD sample is one that contains outliers among its data points. a. True b. False 14. The effect-size measure for the one-way within-groups ANOVA is Cohen's d. a. True b. False 15. Each participant experiences all levels of the independent variable in a within-groups design. a. True b. False 16. The primary benefit of the within-groups design is that between-groups variability is reduced. a. True b. False 17. The one-way within-groups ANOVA has all of the same assumptions as the one-way between-groups ANOVA. a. True b. False 18. The one-way within-groups ANOVA is the multiple-group equivalent of an independent-samples t test. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 2

Name:

Class:

Date:

Chap_13_5e 19. Each participant experiences all levels of the dependent variable in a within-groups design. a. True b. False 20. The one-way within-groups ANOVA is the multiple-group equivalent of a paired-samples t test. a. True b. False 21. WEIRD samples mean that a researcher's results may not generalize to all human beings. a. True b. False 22. The primary benefit of the within-groups design is that within-groups variability is reduced. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 23. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. A significant F test indicates that there are differences between the groups. What post hoc tests are you most interested in performing given these group means: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. comparisons of preference for pop music with the other genres because of their popularity b. computing all between-group comparisons c. comparisons of preference for country music with the other genres, because the mean is noticeably smaller d. any comparison of preference for country and rock music because of their low popularity 24. _____ degrees of freedom must be calculated in a within-groups ANOVA. a. Five b. Two c. Three d. Four

Page 3

Name:

Class:

Date:

Chap_13_5e 25. A one-way within-groups ANOVA on 2 and 20 degrees of freedom results in an F statistic of 3.32. What decision should be made about the hypothesis based on this statistic? a. Fail to reject the null hypothesis because the F statistic does not exceed the critical cutoff. b. Fail to reject the null hypothesis and conclude there is no effect of the independent variable. c. Reject the null hypothesis and conclude there are differences among the levels of the independent variable. d. Reject the null hypothesis and conclude the independent variable has no effect. 26. According to the conventions established by Jacob Cohen, an effect size of 0.014 for a one-way withingroups ANOVA is considered to be: a. small. b. medium. c. medium to large. d. large. 27. A benefit of within-groups designs is that they reduce error due to differences between the groups by using the same participants in each group. If differences are then observed between the groups, those differences are attributed to: a. extraneous variables. b. differences between participants. c. failure to control other important variables. d. the manipulation of the independent variable. 28. Which formula for calculating the subjects sum of squares for a within-groups ANOVA is correct? a. Σ(Mparticipant – GM)2 b. Σ(Mparticipant – X)2 c. Σ(M)(n – 1) d. Σ (Mparticipant – GM) 29. Which F statistic represents a new calculation as part of the within-groups ANOVA? a. Fbetween b. Fwithin c. Fsubjects d. Ftotal

Page 4

Name:

Class:

Date:

Chap_13_5e 30. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 11.1 7.8 9.1 28

df 2 4 8 14

MS 5.55 1.95 1.14

F 4.879 1.714

a. 0.33 b. 0.40 c. 0.55 d. 0.46 31. The vast majority of psychological studies are conducted using participants from countries in Europe and North America. This limits the generalizability of the research findings, and the samples are referred to as _____ samples. a. YAVIS b. WEIRD c. COG d. BADGER 32. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. What is the Tukey HSD value when comparing the data for hip hop and country, given the following means for preference: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. 6.905 b. 5.905 c. 1.450 d. –6.905

Page 5

Name:

Class:

Date:

Chap_13_5e 33. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 13.16 7.67 21.42 42.25

df 3 5 15 23

MS 4.387 1.534 1.428

F 3.07 1.07

a. small b. medium c. large d. too small to be of any interest 34. For the data in the source table, what is the F statistic of interest to researchers? Source Between-groups Subjects Within-groups Total

SS 29.52 9.45 24.87 63.84

df 2 6 12 20

MS 14.760 1.575 2.073

a. 0.14 b. 0.76 c. 7.12 d. 9.37 35. WEIRD samples are samples in which: a. there is little within-subject variability. b. there are a few extreme scores that impact the test statistics. c. the participants come from wealthy, educated, Western democratic countries. d. only some participants respond to the experimental manipulation. 36. Why is the subjects sum of squares calculated separately from the within-groups sum of squares? a. to remove Type I error b. to remove the variability between groups c. to remove the variability within groups d. to remove the variability due to participant differences

Page 6

Name:

Class:

Date:

Chap_13_5e 37. For the data in the source table, calculate the F statistic of interest to researchers. Source Between-groups Subjects Within-groups Total

SS 13.16 7.67 21.42 42.25

df 3 5 15 23

MS 4.387 1.534 1.428

a. 0.33 b. 1.07 c. 2.86 d. 3.07 38. A one-way within-groups ANOVA is also known as a _____ ANOVA. a. between-groups b. single samples c. repeated-measures d. multiple groups 39. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 29.52 9.45 24.87 63.84

df 2 6 12 20

MS 14.760 1.575 2.073

a. small b. medium c. large d. too small to be of any interest

Page 7

Name:

Class:

Date:

Chap_13_5e 40. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 29.52 9.45 24.87 63.84

df 2 6 12 20

MS 14.760 1.575 2.073

a. 0.275 b. 0.543 c. 0.462 d. 0.390 41. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. What are the critical cutoffs for the two-tailed Tukey HSD tests for these data, assuming a p level of 0.05, given the following group means: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. –3.53 and 3.53 b. –3.90 and 3.90 c. –4.41 and 4.41 d. –3.85 and 3.85 42. For the data in the source table, what decision can be made about the hypotheses being tested? Source Between-groups Subjects Within-groups Total

SS 13.16 7.67 21.42 42.25

df 3 5 15 23

MS 4.387 1.534 1.428

Page 8

Name:

Class:

Date:

Chap_13_5e 43. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. What is the Tukey HSD value when comparing the data for country and rock given the following means for preference: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. 1.240 b. 6.905 c. –6.905 d. –5.905 44. For which of the following reasons are within-groups designs preferred over between-groups designs? a. Variability due to participants' differences is held constant across levels of the independent variable in the within-groups design, resulting in less within-groups variability. b. Loss of participants has a more significant negative impact on between-groups designs as compared to within-groups designs. c. Control of extraneous variables is easier when using between-groups designs than when using withingroups designs. d. Within-groups designs take less time to carry out compared to between-groups designs. 45. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. What is the Tukey HSD value when comparing the data for pop and country given the following means for preference: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. 7.619 b. 6.905 c. 5.905 d. 1.600 46. In addition to hypothesis testing, effect size calculations are recommended as a way to assess: a. the power of the statistical analysis. b. the number of subjects needed to correctly reject the null hypothesis. c. where the significant differences exist among groups. d. whether the significant differences found are large enough to matter. 47. A one-way within-groups ANOVA on 2 and 32 degrees of freedom results in an F statistic of 3.44. What decision should be made about the hypothesis based on this statistic? a. Fail to reject the null hypothesis because the F statistic does not exceed the critical cutoff. b. Fail to reject the null hypothesis and conclude there is no effect of the independent variable. c. Reject the null hypothesis and conclude there are differences among the levels of the independent variable. d. Reject the null hypothesis and conclude the independent variable has no effect. Copyright Macmillan Learning. Powered by Cognero.

Page 9

Name:

Class:

Date:

Chap_13_5e 48. According to the conventions established by Jacob Cohen, an effect size of 0.17 for a one-way withingroups ANOVA is considered to be: a. small. b. medium. c. medium to large. d. large. 49. Compared to a one-way between-groups ANOVA, the effect size calculation for the one-way ANOVA differs in: a. Cohen's guidelines for interpreting the effect size values. b. that variability due to subjects is removed from total variability in the denominator. c. its ease of computation. d. the content of the numerator. 50. Buying jeans can be a challenge for women. A local shop owner wants to be selective in which brands of jeans she carries in her store. She gets six of her closest friends to try five different pairs of jeans, with the requirement that they wear each pair for two whole days, without washing, and they rate the jeans at the end of the second day. What research design should she use to analyze her data? a. one-way between-groups ANOVA b. independent-samples t tests c. one-way within-groups ANOVA d. paired-samples t tests 51. Dr. Ravanelli was interested in investigating the effects of stress on memory. He exposed participants to public-speaking stress and noise stress, and gave them a memory measure immediately after exposure to the two stressors. Dr. Ravanelli also gave participants the memory measure at baseline, when they first entered the research laboratory. Dr. Ravanelli hypothesized that participants' memories would be most affected by the two stressors compared to the baseline condition. What statistical test should Dr. Ravanelli use to test his hypothesis? a. between-groups ANOVA b. independent samples t test c. within-groups ANOVA d. dependent samples t test 52. Order effects in a within-groups ANOVA can be resolved by using: a. random sampling. b. counterbalancing. c. a larger sample size. d. mean differences.

Page 10

Name:

Class:

Date:

Chap_13_5e 53. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 11.1 7.8 9.1 28

df 2 4 8 14

MS 5.55 1.95 1.14

F 4.879 1.714

a. small b. medium c. large d. too small to be of any interest 54. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. What is the Tukey HSD value when comparing the data for pop and hip hop given the following means for preference: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. 0.150 b. 0.714 c. 1.000 d. 1.714 55. _____ sums of squares must be calculated in a within-groups ANOVA. a. Five b. Two c. Three d. Four 56. What types of designs are normally preferred to reduce variability by reducing individual differences? a. between-groups designs b. post hoc designs c. within-groups designs d. correlational designs

Page 11

Name:

Class:

Date:

Chap_13_5e 57. For the data in the source table, what decision can be made about the hypotheses being tested? Source Between-groups Subjects Within-groups Total

SS 29.52 9.45 24.87 63.84

df 2 6 12 20

MS 14.760 1.575 2.073

a. Fail to reject the null hypothesis because the F statistic does not exceed the critical cutoff. b. Fail to reject the null hypothesis and conclude there is no effect of the independent variable. c. Reject the null hypothesis and conclude there are differences among the levels of the independent variable. d. Reject the null hypothesis and conclude the independent variable has no effect. 58. The _____ sum of squares is unique to the within-subjects design. a. between-groups b. within-groups c. total d. subjects 59. A one-way within-groups ANOVA is used when there is(are) _____ independent variable(s) with at least _____ levels. a. 1; 2 b. 2; 2 c. 1; 3 d. 2; 3 60. _____ is (are) an additional assumption that must be evaluated in a within-groups ANOVA compared to a between-groups ANOVA. a. Random sampling b. Order effects c. Sample size d. Mean differences

Page 12

Name:

Class:

Date:

Chap_13_5e 61. For the data in the source table, what decision can be made about the hypotheses being tested? Source Between-groups Subjects Within-groups Total

SS 11.1 7.8 9.1 28

df 2 4 8 14

MS 5.55 1.95 1.14

a. Fail to reject the null hypothesis because the F statistic does not exceed the critical cutoff. b. Fail to reject the null hypothesis and conclude there is no effect of the independent variable. c. Reject the null hypothesis and conclude there are differences among the levels of the independent variable. d. Reject the null hypothesis and conclude the independent variable has no effect. 62. What is the correct formula for subjects degrees of freedom in a one-way within-groups ANOVA? a. Ngroups – 1 b. n – 1 c. Ntotal – 1 d. n – 2 63. The one-way within-groups design can be viewed as an extension of which other research design because of its ability to analyze data from more groups? a. independent-samples t test b. paired-samples t test c. Tukey HSD test d. between-groups ANOVA 64. A researcher wants to examine people's preference for pets by having 10 people act as "foster owners" for four different types of family pets: dogs, cats, birds, and fish. The participants will foster each type of pet for one week, and a scale measure will be used to assess preference. Which research design should be used? a. one-way between-groups ANOVA b. correlation c. one-way within-groups ANOVA d. paired-samples t tests

Page 13

Name:

Class:

Date:

Chap_13_5e 65. According to the conventions established by Jacob Cohen, an effect size of 0.067 for a one-way withingroups ANOVA is considered to be: a. small. b. medium. c. medium to large. d. large. 66. Musical preferences were examined for four different types of musical genres (pop, hip hop, country, and rock), using a one-way within-groups ANOVA with nine participants. The standard error was calculated as 0.21. Where are the significant effects for preference among these groups, given the following means for preference: Pop M = 4.43, Hip hop M = 4.28, Country M = 2.83, Rock M = 4.07? a. Preference for country music is significantly lower than preferences for the other three genres, and there are no other differences among the mean preferences. b. Pop and hip hop are preferred over country and rock, but there are no differences between the other genres. c. Country and rock music are less preferred than pop, but there are no other differences present. d. All post hoc tests were significant at the 0.05 level. 67. The benefits of the within-groups design are seen in the calculation of the F statistic, in that the: a. numerator is larger, resulting in a larger F statistic. b. numerator is smaller, resulting in a smaller F statistic. c. denominator is smaller, resulting in a larger F statistic. d. denominator is larger, resulting in a smaller F statistic. 68. To calculate the total sum of square for the one-way within-groups ANOVA, each score's deviation from the _____ is squared and summed. a. mean b. grand mean c. cell mean d. group mean

Page 14

Name:

Class:

Date:

Chap_13_5e 69. For the data in the source table, what is the effect size for this one-way within-groups ANOVA? Source Between-groups Subjects Within-groups Total

SS 13.16 7.67 21.42 42.25

df 3 5 15 23

MS 4.387 1.534 1.428

F 3.07 1.07

a. 0.311 b. 0.381 c. 0.506 d. 2.869 70. A paired-samples t test is used when there are _____ groups; a within-groups ANOVA is used when there are at least _____ groups. a. 3; 2 b. 2; 3 c. 1; 3 d. 1; 2 71. In addition to hypothesis testing, post hoc tests are required as a way to assess: a. the power of the statistical analysis. b. the number of subjects needed to correctly reject the null hypothesis. c. where the significant differences exist among groups. d. whether the significant differences found are large enough to matter. 72. What is the correct formula for between-groups degrees of freedom? a. Ngroups – 1 b. n – 1 c. Ntotal – 1 d. n – 2 73. Which degrees of freedom value is unique to the one-way within-groups ANOVA? a. between-groups b. subjects c. within-groups d. total

Page 15

Name:

Class:

Date:

Chap_13_5e 74. A recent criticism leveled against much of the psychological research literature has to do with WEIRD samples, and how these samples limit the generalizability of the research findings. Researchers are now being encouraged to specify the target population to which the study's results apply. This statement is known as a: a. population of interest (POI). b. conditions of generalizability (COG). c. population of concern (POC). d. constraint on generality (COG). 75. For the data in the source table, calculate the F statistic of interest to researchers. Source Between-groups Subjects Within-groups Total

SS 11.1 7.8 9.1 28

df 2 4 8 14

MS 5.55 1.95 1.14

a. 4.87 b. 2.85 c. 1.71 d. 0.21 76. Determine the critical cutoff for F at a p level of 0.05 if the following degrees of freedom were calculated for a one-way within-groups ANOVA: dfbetween = 4, dfsubjects = 7, dftotal = 39, dfwithin = 28. a. 2.61 b. 2.64 c. 2.72 d. 4.12 77. Determine the critical cutoff for F at a p level of 0.05 if the following degrees of freedom were calculated for a one-way within-groups ANOVA: dfbetween = 3, dfsubjects = 8, dftotal = 35, dfwithin = 24. a. 4.72 b. 4.07 c. 3.27 d. 3.01 78. If using a within-groups design results in a larger F statistic, this also means: a. there is less variability between groups. b. it is easier to reject the null hypothesis with this design. c. Type I errors are more likely. d. power is decreased. Copyright Macmillan Learning. Powered by Cognero.

Page 16

Name:

Class:

Date:

Chap_13_5e Enter the appropriate word(s) to complete the statement. 79. Order effects in a one-way within-groups ANOVA are avoided by using _______.

80. The symbol for effect size for one-way within-groups ANOVA is _______.

81. When using a one-way within-groups ANOVA, the _______ variable has more than two levels.

82. Another name for the one-way within-groups ANOVA is the _______ ANOVA.

83. An assumption of the one-way within-groups ANOVA that was not made for a one-way between-groups ANOVA is to avoid _______ effects.

84. _______ is the process of varying the order of levels of the independent variable across participants.

Page 17

Name:

Class:

Date:

Chap_13_5e 85. The one-way _______ ANOVA is used when there are three or more levels of the independent variable, and the same or matched participants experience all levels of the independent variable.

86. There is a new sum of squares for _______ when calculating the one-way within-groups ANOVA.

87. The one-way within-groups ANOVA source table includes two F statistics but we care about only the _______ F statistic.

88. To determine where differences lie between levels of the independent variable, the researcher can use the _______ test after calculating a significant one-way within-groups ANOVA.

89. The one-way within-groups ANOVA is used when there are three or more levels of the independent variable, a scale-dependent variable, and the same or matched participants are in _______ group.

90. The one-way within-groups ANOVA source table includes _______ F statistics.

Page 18

Name:

Class:

Date:

Chap_13_5e 91. The issue of generalizability has been raised in recent years because the majority of psychological research has been conducted in North America, Australia, Israel, and Europe. These sorts of samples are referred to as _______ samples.

92. The one-way within-groups ANOVA is used when there are _______ or more levels of the independent variable, and the same or matched participants experience all levels.

93. Lower within-groups variability means a smaller _______ for the F statistic.

Reading journal articles can be challenging for students, as such articles are often technical in nature. A high school that prides itself on preparing students for college wants to purchase journals that are written at a level accessible to students. The school librarian recruits four students with varying academic ability to read articles from four different journals and rate their readability from 1 (very difficult to read and understand) to 7 (very easy to read and understand). Some hypothetical data are shown in the table. Table: Journal and Readability Journal 1 2 1 3 2

Journal 2 3 3 4 5

Journal 3 4 3 5 3

Journal 4 5 5 6 7

94. (Table: Journal and Readability) Complete steps 5 and 6 of hypothesis testing. Be sure to complete the source table when calculating the F ratio.

Page 19

Name:

Class:

Date:

Chap_13_5e A researcher is interested in what impact self-focusing cues have on exercise behavior. She manipulates four stations at a local weight-loss and physical fitness center for women. One station has a large mirror installed, another includes a computer-feedback system, the third station has a personal trainer present at all times, and the fourth station is situated behind a small privacy wall. The women use the stations in whatever order they chose. Six women are assessed for the intensity of their efforts at each station. Hidden cameras are used. The intensity of the workout is scored from 50, meaning low intensity, to 100, high intensity. Some hypothetical data are shown in the table. Table: Workout Intensity Mirror 76 82 88 72 91 83

Computer Feedback 84 72 68 81 90 77

Personal Trainer 86 92 89 83 98 91

Privacy Wall 60 66 72 78 81 69

95. (Table: Workout Intensity) List each of the assumptions for performing the appropriate ANOVA, and for each assumption evaluate whether the described study and data meet the assumption.

96. Explain what a WEIRD sample is, how these samples impact how we interpret research results, and what suggestions have been offered to address this issue.

97. Explain the logic of a within-groups analysis of variance, paying special attention to how the variance is partitioned (i.e., how variance components are divided up). Be sure to mention how this design offers an advantage over a between-groups design.

Page 20

Name:

Class:

Date:

Chap_13_5e Researchers interested in the perception of three-dimensional shapes on computer screens decide to investigate which components of a square figure or cube are necessary for viewers to perceive details of the shape. They vary the stimuli to include fully rendered cubes, cubes drawn with corners but incomplete sides, and cubes with missing corner information. The viewers are trained on how to detect subtle deformations in the shapes, and then their accuracy rate is measured across the three figure conditions. Accuracy is reported as percent correct. Five participants are recruited for an intense study during which a large number of trials are required. The trials are presented in different orders for each participant using a random numbers table to determine unique sequences. The ANOVA source table and the group means follow. Table: Deformation Detection in Cubes Source Between-groups Subjects Within-groups Total

SS 30.346 25.943 18.472 74.761

df 2 4 8 14

MS 15.173 6.48575 2.309

F 6.571 2.809

Table: Deformation Detection in Cubes, Group Means Group Full cubes Corners, incomplete sides Missing corners

M 96.5 91.9 79.7

98. (Table: Deformation Detection in Cubes & Table: Deformation Detection in Cubes, Group Means) (a) Is it appropriate to run a post hoc analysis for this experiment? Why or why not? (b) If it is appropriate to run a post hoc analysis for this experiment, perform the analysis using Tukey's HSD with a p level of 0.05. Clearly indicate which groups differ significantly from one another.

Page 21

Name:

Class:

Date:

Chap_13_5e A researcher is interested in what reaction people have while crying. Five emergency room nurses were randomly selected to participate in a study where they were exposed to different types of people and emotions. Participants were simply asked to rate the likability of the person seen on a video clip from 1 (not very likable) to 10 (highly likable). The video clips of interest showed men crying, women crying, and children of both genders crying. These clips were randomly ordered among other video segments of different people showing different emotions, and a new random order was generated for each presentation. For each participant, ratings were averaged for images within the same category. Some hypothetical data are shown in the table. Table: Reaction to Crying Men 8 5 6 7 4

Women 7 6 6 8 9

Girls 9 6 8 8 7

Boys 8 7 6 7 8

99. (Table: Reaction to Crying) List each of the assumptions for performing the appropriate ANOVA, and for each assumption evaluate whether the described study and data meet the assumption.

Page 22

Name:

Class:

Date:

Computer Feedback 84 72 68 81 90 77

Personal Trainer 86 92 89 83 98 91

Privacy Wall 60 66 72 78 81 69

100. (Table: Workout Intensity) Complete steps 5 and 6 of hypothesis testing. Be sure to complete the source table when calculating the F ratio.

Page 23

Name:

Class:

Date:

Chap_13_5e A researcher performed an attractiveness study using still photos of men, with four different hair conditions: short hair, long hair, partially bald, and completely shaved. All photos were controlled for other factors that influence attractiveness. Eighteen heterosexual women from a random sample were recruited to rate the photos for attractiveness. A rating scale from 0 to 50, with higher scores indicating greater perceived attractiveness, was used and an alpha level of 0.01 was set by the researcher. The ANOVA source table and the group means are provided. Table: Hair and Attractiveness Source Between-groups Subjects Within-groups Total

SS 331.071 306.982 1548.254 2186.307

df 3 17 51 71

MS 110.357 18.0577647 30.3579216

F 3.64 0.59

Table: Hair and Attractiveness, Group Means Group Short Long Partially bald Completely shaved

M 30.175 31.108 29.127 31.862

101. (Table: Hair and Attractiveness & Table: Hair and Attractiveness, Group Means) (a) Is it appropriate to run a post hoc analysis for this experiment? Why or why not? If it is appropriate to run a post hoc analysis for this experiment, perform the analysis using Tukey's HSD with a p level of 0.01. Clearly indicate which groups differ significantly from one another. (b) Calculate the R2 measure of effect size for this ANOVA.

Page 24

Name:

Class:

Date:

Chap_13_5e Researchers are interested in the differential effects of reward and punishment on performance. They have subjects learn list of words under three different payoff conditions. In the reward condition, participants receive a monetary reward for each correctly remembered work. In contrast, in the punishment condition, participants lose money for each incorrectly remembered stimulus word. In the neutral condition participants are neither rewarded for accurate responses nor punished for inaccurate responses. Accuracy is reported as percent correct. Sixteen participants are recruited and tested in each of the three payoff conditions. The presentation of payoff conditions is counterbalanced to avoid order effects. The ANOVA source table and the group means follow. Table: Responses to Reward and Punishment Source Between-groups Subjects Within-groups Total

SS 137.201 95.046 380.164 612.411

df 2 15 30 47

MS 68.601 6.336 12.672

F 5.413 0.500

Table: Responses to Reward and Punishment, Group Means Group Neutral Reward Punishment

M 89.7 94.3 87.2

102. (Table: Responses to Reward and Punishment & Table: Responses to Reward and Punishment, Group Means) Use the information to calculate the R2 measure of effect size for this ANOVA.

Page 25

Name:

Class:

Date:

Chap_13_5e Reading journal articles can be challenging for students, as such articles are often technical in nature. A high school that prides itself on preparing students for college wants to purchase journals that are written at a level accessible to students. The school librarian recruits four students with varying academic ability to read articles from four different journals and rate their readability from 1 (very difficult to read and understand) to 7 (very easy to read and understand). Some hypothetical data are shown in the table. Table: Journal and Readability Journal 1 2 1 3 2

Journal 2 3 3 4 5

Journal 3 4 3 5 3

Journal 4 5 5 6 7

103. (Table: Journal and Readability) List each of the assumptions for performing the appropriate ANOVA, and for each assumption evaluate whether the described study and data meet the assumption.

Page 26

Name:

Class:

Date:

SS 30.346 25.943 18.472 74.761

df 2 4 8 14

MS 15.173 6.48575 2.309

F 6.571 2.809

Table: Deformation Detection in Cubes, Group Means Group Full cubes Corners, incomplete sides Missing corners

M 96.5 91.9 79.7

104. (Table: Deformation Detection in Cubes & Table: Deformation Detection in Cubes, Group Means) Use the information to calculate the R2 measure of effect size for this ANOVA.

Page 27

Name:

Class:

Date:

SS 137.201 95.046 380.164 612.411

df 2 15 30 47

MS 68.601 6.336 12.672

F 5.413 0.500

Table: Responses to Reward and Punishment, Group Means Group Neutral Reward Punishment

M 89.7 94.3 87.2

105. (Table: Responses to Reward and Punishment & Table: Responses to Reward and Punishment, Group Means) (a) Is it appropriate to run a post hoc analysis for this experiment? Why or why not? (b) If it is appropriate to run a post hoc analysis for this experiment, perform the analysis using Tukey's HSD with a p level of 0.05. Clearly indicate which groups differ significantly from one another.

Page 28

Name:

Class:

Date:

Women 7 6 6 8 9

Girls 9 6 8 8 7

Boys 8 7 6 7 8

106. (Table: Reaction to Crying) Complete steps 5 and 6 of hypothesis testing. Be sure to complete the source table when calculating the F ratio.

Page 29

Name:

Class:

Date:

Chap_13_5e Answer Key 1. False 2. True 3. True 4. False 5. True 6. False 7. True 8. True 9. False 10. True 11. True 12. False 13. False 14. False 15. True 16. False 17. False 18. False 19. False 20. True 21. True 22. True 23. c 24. d 25. a 26. a Copyright Macmillan Learning. Powered by Cognero.

Page 30

Name:

Class:

Date:

Chap_13_5e 27. d 28. a 29. c 30. c 31. b 32. a 33. c 34. c 35. c 36. d 37. d 38. c 39. c 40. b 41. b 42. a 43. d 44. a 45. a 46. d 47. c 48. d 49. b 50. c 51. c 52. b 53. c 54. b Copyright Macmillan Learning. Powered by Cognero.

Page 31

Name:

Class:

Date:

Chap_13_5e 55. d 56. c 57. c 58. d 59. c 60. b 61. c 62. b 63. b 64. c 65. b 66. a 67. c 68. b 69. b 70. b 71. c 72. a 73. b 74. d 75. a 76. c 77. d 78. b 79. counterbalancing 80. 81. independent 82. repeated-measures Copyright Macmillan Learning. Powered by Cognero.

Page 32

Name:

Class:

Date:

Chap_13_5e 83. order 84. Counterbalancing 85. within-groups 86. subjects 87. between-groups 88. Tukey HSD 89. every, each 90. two, 2 91. WEIRD 92. three, 3 93. denominator 94. Step 5: Calculate total sum of squares. The sum is 38.438. aaa Journal 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4

X 2 1 3 2 3 3 4 5 4 3 5 3 5 5 6 7

(X – GM) –1.813 –2.813 –0.813 –1.813 –0.813 –0.813 0.187 1.187 0.187 –0.813 1.187 –0.813 1.187 1.187 2.187 3.187

(X – GM)2 3.287 7.913 0.661 3.287 0.661 0.661 0.035 1.409 0.035 0.661 1.409 0.661 1.409 1.409 4.783 10.157

Calculate between-groups sum of squares. The sum is 28.188. Journal 1

X 2

Group Mean 2.00

(M – GM) –1.813

(M – GM)2 3.287 Page 33

Name:

Class:

Date:

Chap_13_5e 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4

1 3 2 3 3 4 5 4 3 5 3 5 5 6 7

2.00 2.00 2.00 3.75 3.75 3.75 3.75 3.75 3.75 3.75 3.75 5.75 5.75 5.75 5.75

–1.813 –1.813 –1.813 –0.063 –0.063 –0.063 –0.063 –0.063 –0.063 –0.063 –0.063 1.937 1.937 1.937 1.937

3.287 3.287 3.287 0.004 0.004 0.004 0.004 0.004 0.004 0.004 0.004 3.752 3.752 3.752 3.752

Calculate subjects sum of squares. The sum equals 5.688. Participant

Journal

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4

2 1 3 2 3 3 4 5 4 3 5 3 5 5 6 7

Participant Mean 3.5 3 4.5 4.25 3.5 3 4.5 4.25 3.5 3 4.5 4.25 3.5 3 4.5 4.25

(Mpart– GM)

–0.313 –0.813 0.687 0.437 –0.313 –0.813 0.687 0.437 –0.313 –0.813 0.687 0.437 –0.313 –0.813 0.687 0.437

(Mpart– GM)2 0.098 0.661 0.472 0.191 0.098 0.661 0.472 0.191 0.098 0.661 0.472 0.191 0.098 0.661 0.472 0.191

Calculate within-groups sum of squares. SSwithin = SStotal – SSbetween – SSsubjects SSwithin = 38.438 – 28.188 – 5.688 = 4.562 Source Copyright Macmillan Learning. Powered by Cognero.

F Page 34

Name:

Class:

Date:

Chap_13_5e Between-groups Subjects Within-groups Total

28.188 5.688 4.562 34.438

3 3 9 15

9.396 1.896 0.507

18.534 3.740

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. The readability of articles does differ between the journals, F(3, 9) = 18.534, p < 0.05. Post hoc tests are needed to determine where significant differences occurred. 95. This study requires a one-way within-groups ANOVA, which has four assumptions. The first assumption is that the participants are randomly selected from the population. The women assessed self-selected to be members of the fitness center. We would want to be careful in generalizing past this sample. The second assumption is that the dependent variable is normally distributed in the population. The sample size is small, so the sampling distribution may not be normal. The third assumption is that the group variances are homoscedastic. The variances are included in the following table, and the largest variance is just a little more than twice the smallest. Because the sample sizes are equal (this is a within-subjects design) and the largest variance is more than two times the smallest variance, the homogeneity of variance assumption may be violated and we should proceed with caution. Mirror 50.800

Computer Feedback 64.667

Personal Trainer 26.967

Privacy Wall 60.000

The final assumption relates to order effects. The stations were used in an order chosen by the women, so we do not know if the order was varied. 96. A WEIRD sample is a sample consisting of participants recruited from Western, Educated, Industrialized, Rich, and Democratic countries. Most published psychological research is conducted using participants from these countries. Because most of the world's population lives in countries that are not WEIRD, the results of such psychological research may not be generalizable to the larger world. Some researchers have suggested that scientific investigators include a statement of constraints on generality (COG) in their papers. This statement clearly specifies the target population to which the results can be generalized. 97. A within-groups analysis of variance uses the same participants, or matched participants, at each level of the independent variables. In other words, each participant experiences all levels. Four sources of variability are identified for a within-groups ANOVA, including total sum of squares, between-groups sum of squares, withingroups sum and squares, and a new sum of squares labeled for subjects. This new sum of squares accounts for variability for each participant across conditions by getting an average for each participant across conditions, subtracting the grand mean, squaring each deviation, and summing those values. Within-groups variability then excludes variability that can be accounted for by the subjects. This is the advantage of the within-groups ANOVA compared to the between-groups analysis; we can factor out additional variability by using the same participants across conditions. The calculation of the F statistic is then a ratio of between-groups variability to within-groups variability, after removing variability due to subjects. Because of the removal of subjects variability, the withingroups variability is smaller, resulting in a larger F statistic. This reduction in variability in the denominator, along with the larger F statistic, makes it easier to reject the null hypothesis. Copyright Macmillan Learning. Powered by Cognero.

Page 35

Name:

Class:

Date:

Chap_13_5e 98. (a) Yes, it is appropriate to perform post hoc tests because the F statistic exceeds the critical cutoff. (b) The standard error is 0.680. Full cubes versus corners = (96.5 – 91.9)/0.680 = 6.765 Full cubes versus missing corners = (96.5 – 79.7)/0.680 = 24.706 Corners versus missing corners = (91.9 – 79.7)/0.680 = 17.941 The critical cutoff for q is 4.04. All three post hoc tests are significant at the 0.05 level. Participants are significantly more accurate when figures are drawn as complete cubes (M = 96.5% accurate), compared to with corners but incomplete sides (M = 91.9), and both of those conditions result in higher accuracy than when cubes are rendered with missing corners (M = 79.7). 99. This study requires a one-way within-groups ANOVA, which has four assumptions. The first assumption is that the participants are randomly selected from the population. We know that these people were selected because they are nurses, and they were randomly selected from all emergency room nurses. The second assumption is that the dependent variable is normally distributed in the population. The data may be skewed toward higher likability ratings. The sample size is small, so the sampling distribution may not be normal. The third assumption is that the group variances are homoscedastic. The variances are Men = 2.50, Women = 1.7, Girls = 1.3, and Boys = 0.7. Because the largest variance is not more than two times the smallest variance, the homogeneity of variance assumption is not violated. The final assumption relates to order effects. The video segments were randomly presented and a unique random order was used for each presentation, so we can assume order effects were controlled through this procedure. 100. Step 5: Calculate total sum of squares. The sum is 2109.634. Condition Mirror Mirror Mirror Mirror Mirror Mirror Computer Computer Computer Computer Computer Computer Trainer Trainer Trainer Trainer Trainer Trainer Wall

X 76 82 88 72 91 83 84 72 68 81 90 77 86 92 89 83 98 91 60

(X – GM) –4.375 1.625 7.625 –8.375 10.625 2.625 3.625 –8.375 –12.375 0.625 9.625 –3.375 5.625 11.625 8.625 2.625 17.625 10.625 –20.375

(X – GM)2 19.141 2.641 58.141 70.141 112.891 6.891 13.141 70.141 153.141 0.391 92.641 11.391 31.641 135.141 74.391 6.891 310.641 112.891 415.141 Page 36

Name:

Class:

Date:

Chap_13_5e Wall Wall Wall Wall Wall

66 72 78 81 69

–14.375 –8.375 –2.375 0.625 –11.375

206.641 70.141 5.641 0.391 129.391

Calculate between-groups sum of squares. The sum is 1097.460. Condition Mirror Mirror Mirror Mirror Mirror Mirror Computer Computer Computer Computer Computer Computer Trainer Trainer Trainer Trainer Trainer Trainer Wall Wall Wall Wall Wall Wall

X 76 82 88 72 91 83 84 72 68 81 90 77 86 92 89 83 98 91 60 66 72 78 81 69

(M – GM) 1.625 1.625 1.625 1.625 1.625 1.625 –1.708 –1.708 –1.708 –1.708 –1.708 –1.708 9.458 9.458 9.458 9.458 9.458 9.458 –9.375 –9.375 –9.375 –9.375 –9.375 –9.375

Group Mean 82 82 82 82 82 82 78.667 78.667 78.667 78.667 78.667 78.667 89.833 89.833 89.833 89.833 89.833 89.833 71 71 71 71 71 71

(M – GM)2 2.641 2.641 2.641 2.641 2.641 2.641 2.917 2.917 2.917 2.917 2.917 2.917 89.454 89.454 89.454 89.454 89.454 89.454 87.891 87.891 87.891 87.891 87.891 87.891

Calculate subjects sum of squares. The sum equals 472.884. Participant 1 2 3 4 5

Condition Mirror Mirror Mirror Mirror Mirror

X 76 82 88 72 91

Participant Mean (Mpart– GM) (Mpart– GM)2 76.5 –3.875 15.016 78 –2.375 5.641 79.25 –1.125 1.266 78.5 –1.875 3.516 90 9.625 92.641 Page 37

Name:

Class:

Date:

Chap_13_5e 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6

Mirror Computer Computer Computer Computer Computer Computer Trainer Trainer Trainer Trainer Trainer Trainer Wall Wall Wall Wall Wall Wall

83 84 72 68 81 90 77 86 92 89 83 98 91 60 66 72 78 81 69

80 76.5 78 79.25 78.5 90 80 76.5 78 79.25 78.5 90 80 76.5 78 79.25 78.5 90 80

–0.375 –3.875 –2.375 –1.125 –1.875 9.625 –0.375 –3.875 –2.375 –1.125 –1.875 9.625 –0.375 –3.875 –2.375 –1.125 –1.875 9.625 –0.375

0.141 15.016 5.641 1.266 3.516 92.641 0.141 15.016 5.641 1.266 3.516 92.641 0.141 15.016 5.641 1.266 3.516 92.641 0.141

Calculate within-groups sum of squares. SSwithin = SStotal – SSbetween – SSsubjects SSwithin = 2109.634 – 1097.418 – 472.884 = 539.332 Source Between-groups Subjects Within-groups Total

SS 1097.418 472.884 539.332 2109.634

df 3 5 15 23

MS 365.806 94.577 35.955

F 10.174 2.630

Step 6: The F statistic exceeds the F critical value. Therefore, we reject the null hypothesis. The conditions of the workout that might influence self-focusing do influence the intensity of a workout, F(3, 15) = 10.174, p < 0.05. Post hoc tests are needed to see where significant differences exist between conditions. 101. (a) It is not appropriate to conduct post hoc tests because the ANOVA is not significant. (b) R2 = 331.071/(2186.307 – 306.982) = 0.176. This is a large effect size. While the ANOVA was not significant, the effect size indicates that there might be something to this hypothesis, encouraging future study. 102. R2 = 137.201/(612.411 – 95.046) = 0.265. This is a large effect size, indicating that 26.5% of the variability in memory performance is accounted for by payoff condition under which the participant is tested.

Page 38

Name:

Class:

Date:

Chap_13_5e 103. This study requires a one-way within-groups ANOVA, which has four assumptions. The first assumption is that the participants are randomly selected from the population. We know that these students were selected because of their varying academic ability, so we know they were not randomly selected. The second assumption is that the dependent variable is normally distributed in the population. The data do not appear to be badly skewed, but the sample size is small, so the sampling distribution may not be normal. The third assumption is that the group variances are homoscedastic. Just by looking at the data, we can see that the variability across conditions is highly similar. The variances are Journal 1 = 0.67 and Journals 2, 3, 4 = 0.92. Because the largest variance is not more than two times the smallest variance, the homogeneity of variance assumption is not violated. The final assumption relates to order effects. There is no indication that counterbalancing was done to control for the order of journals read, so order effects may be present. 104. R2 = 30.346/(74.761 – 18.472) = 0.622. This is a very large effect size, indicating that 62.2% of the variability in accuracy of detecting deformations is accounted for by the rendering of the cube. 105. (a) Yes, it is appropriate to perform post hoc tests because the F statistic exceeds the critical cutoff. (b) The standard error is 0.890. Neutral versus reward = (89.7 – 94.3)/0.890 = –5.169 Neutral versus punishment = (89.7 – 87.2)/0.890 = 2.809 Reward versus punishment = (94.3 – 87.2)/0.890 = 7.978 The critical cutoff for q is 3.49. Two of the three post hoc tests are significant at the 0.05 level. Participants are significantly more accurate when their responses are rewarded (M = 94.3% accurate), compared to conditions where they are punished (M = 87.2) or when there is no reward or punishment (M = 87.2), and there was no difference between the punishment and neutral conditions. 106. Step 5: Calculate total sum of squares. The sum is 32.000. Crier Man Man Man Man Man Woman Woman Woman Woman Woman Girl Girl Girl Girl Girl Boy Boy

X 8 5 6 7 4 7 6 6 8 9 9 6 8 8 7 8 7

(X – GM) 1 –2 –1 0 –3 0 –1 –1 1 2 2 –1 1 1 0 1 0

(X – GM)2 1.000 4.000 1.000 0.000 9.000 0.000 1.000 1.000 1.000 4.000 4.000 1.000 1.000 1.000 0.000 1.000 0.000 Page 39

Name:

Class:

Date:

Chap_13_5e Boy Boy Boy

6 7 8

–1 0 1

1.000 0.000 1.000

Calculate between-groups sum of squares. The sum is 7.200. Crier Man Man Man Man Man Woman Woman Woman Woman Woman Girl Girl Girl Girl Girl Boy Boy Boy Boy Boy

X 8 5 6 7 4 7 6 6 8 9 9 6 8 8 7 8 7 6 7 8

Group Mean 6 6 6 6 6 7.2 7.2 7.2 7.2 7.2 7.6 7.6 7.6 7.6 7.6 7.2 7.2 7.2 7.2 7.2

(M – GM) –1 –1 –1 –1 –1 0.2 0.2 0.2 0.2 0.2 0.6 0.6 0.6 0.6 0.6 0.2 0.2 0.2 0.2 0.2

(M – GM)2 1.000 1.000 1.000 1.000 1.000 0.040 0.040 0.040 0.040 0.040 0.360 0.360 0.360 0.360 0.360 0.040 0.040 0.040 0.040 0.040

Calculate subjects sum of squares. The sum equals 10.0. Participant 1 2 3 4 5 1 2 3 4 5 1 Copyright Macmillan Learning. Powered by Cognero.

Crier Man Man Man Man Man Woman Woman Woman Woman Woman Girl

8 5 6 7 4 7 6 6 8 9 9

Participant Mean 8 6 6.5 7.5 7 8 6 6.5 7.5 7 8

(Mpart– GM 1 –1 –0.5 0.5 0 1 –1 –0.5 0.5 0 1 Page 40

Name:

Class:

Date:

Chap_13_5e 2 3 4 5 1 2 3 4 5

Girl Girl Girl Girl Boy Boy Boy Boy Boy

6 8 8 7 8 7 6 7 8

6 6.5 7.5 7 8 6 6.5 7.5 7

–1 –0.5 0.5 0 1 –1 –0.5 0.5 0

Calculate within-groups sum of squares. SSwithin = SStotal – SSbetween – SSsubjects SSwithin = 32.0 – 7.2 – 10.0 = 14.8 Source Between-groups Subjects Within-groups Total

SS 7.2 10.0 14.8 32.0

df 3 4 12 42

MS 2.400 2.500 1.233

F 1.946 2.027

Step 6: The F statistic falls short of the F critical value. Therefore, we fail to reject the null hypothesis. We do not know if t relationship between characteristics of someone who is crying and the reaction of emergency room nurses, specifically ratin F(3, 12) = 1.946, p > 0.05.

Page 41

Name:

Class:

Date:

Chap_14_5e Indicate whether the statement is true or false. 1. An interaction is more likely to occur in a two-way ANOVA when at least one of the main effects is significant. a. True b. False 2. An effect size calculation of 0.08 for a two-way between-groups ANOVA is considered to be a small effect. a. True b. False 3. A two-way factorial experimental design yields no more information than two single-factor experiments. a. True b. False 4. In a two-way between-groups ANOVA, there are four sources of variability. a. True b. False 5. A 2 × 4 ANOVA will have 6 cells. a. True b. False 6. In a two-way between-groups ANOVA, there are three sources of variability. a. True b. False 7. "Factor" is another word to describe the dependent variable in a two-way ANOVA. a. True b. False 8. Analysis of variance is a statistical tool that can be applied to within-groups designs. a. True b. False 9. When performing a two-way ANOVA, there will be three critical values—one for each main effect and one for the interaction. a. True b. False 10. A significant main effect found in a two-way between-groups ANOVA trumps any significant interaction. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_14_5e 11. The variations on ANOVA, MANOVA and MANCOVA, both include more than one dependent variable. a. True b. False 12. Including a covariate in an analysis means the effects of the independent variables are examined after statistically adding the effect of another variable. a. True b. False 13. A two-way between-groups ANOVA has three hypotheses. a. True b. False 14. When looking up the critical values for F, the within-groups degrees of freedom is the same for each test in a two-way between-groups ANOVA. a. True b. False 15. The effect size measure for a two-way between-groups ANOVA is R2, the same as for a one-way ANOVA. a. True b. False 16. In a two-way ANOVA, differences among row means are used to assess the interaction effect. a. True b. False 17. Intersecting or eventually intersecting lines in a bar graph of the cell means are evidence of an interaction. a. True b. False 18. A two-way between-groups ANOVA has just two hypotheses. a. True b. False 19. A significant interaction found in a two-way between-groups ANOVA trumps any main effects. a. True b. False 20. A 2 × 4 ANOVA will have 8 cells. a. True b. False

Page 2

Name:

Class:

Date:

Chap_14_5e 21. "Factor" is another word to describe the independent variable in a two-way ANOVA. a. True b. False 22. The variations on ANOVA, ANCOVA and MANCOVA, both include a covariate in the design. a. True b. False 23. Parallel lines in a bar graph of the cell means are evidence of an interaction. a. True b. False 24. The variations on ANOVA, MANOVA and MANCOVA, both include a covariate in the design. a. True b. False Indicate the answer choice that best completes the statement or answers the question. The following table reflects the results of a study by Roediger and Karpicke (2006). The authors investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to the material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. Table: Test-Enhanced Learning

Study–study Study–test Mean

5 minutes 0.80 0.75 0.78

2 days 0.55 0.70 0.63

1 week 0.42 0.55 0.49

Mean 0.59 0.67

25. (Table: Test-Enhanced Learning) The cells of this study reflect an interaction. Which statement best describes the interaction? a. On average, people in the study–test condition had better memory compared to people in the study– study condition. b. People in the study–test condition performed more poorly on the 5-minute recall test than did those in the study–study condition, but at longer retention intervals (2 days and 1 week); those in the study– test condition performed better than did those in the study–study condition. c. On average, memory performance decreased as the retention interval increased. d. On average, memory performance increased as the retention interval increased.

Page 3

Name:

Class:

Date:

Chap_14_5e 26. Forys and Dahlquist (2007) investigated the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. Participants were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were next instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. What is (are) the main effect(s) being tested in the study? a. coping style b. coping style and cognitive strategy condition c. pain tolerance and coping style d. pain tolerance and cognitive strategy condition 27. Forys and Dahlquist (2007) investigated the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. Participants were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were next instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. How should the ANOVA used to analyze the data be labeled? a. 4 × 2 between-groups ANOVA b. 4 × 2 within-groups ANOVA c. 2 × 2 between-groups ANOVA d. 2 × 2 within-groups ANOVA 28. A hypothesis test with two nominal independent variables and a scale dependent variable is a: a. one-way ANOVA. b. two-way ANOVA. c. two-groups t test. d. two-groups z test.

Page 4

Name:

Class:

Date:

Chap_14_5e The following source table depicts the results of a fictional study investigating whether the number of hours of sleep a person gets varies with gender (male, female) and with the number of cups of coffee consumed in a day. Equal numbers of men and women were randomly assigned to drink 1, 2, or 3 cups of coffee during the course of a day and then record the number of hours they slept that night. Table: Coffee and Sleep Source Gender Cups of coffee Gender × Cups Within Total

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

29. (Table: Coffee and Sleep) For the data in the source table, what is the effect size for cups of coffee? a. small b. medium c. large d. too small to be of interest 30. A social psychologist wants to statistically test whether gender and political affiliation affect IQ scores. Which test should the researcher use? a. one-way ANOVA b. two-way ANOVA c. three-way ANOVA d. within-groups ANOVA 31. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. How should the ANOVA used to analyze the data be labeled? a. 4 × 2 between-groups ANOVA b. 4 × 2 within-groups ANOVA c. 2 × 2 between-groups ANOVA d. 2 × 2 within-groups ANOVA

Page 5

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

32. (Table: Coffee and Sleep) For the data in the source table, what is the effect size for the interaction? a. 0.01 b. 0.10 c. 0.31 d. 0.69 33. In a two-way ANOVA, the researcher makes decisions regarding _____ null hypotheses. a. three separate b. two separate c. three interdependent d. two interdependent 34. Roediger and Karpicke (2006) investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. What is (are) the main effect(s) being tested in the study? a. study condition b. retention interval condition and idea units recalled c. study condition and idea units recalled d. study condition and retention interval condition

Page 6

Name:

Class:

Date:

Chap_14_5e 35. A social psychologist wants to statistically test whether gender affects reading test scores. Which test should the researcher use? a. one-way ANOVA b. two-way ANOVA c. three-way ANOVA d. within-groups ANOVA The following table reflects the results of a study by Roediger and Karpicke (2006). The authors investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to the material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. Table: Test-Enhanced Learning

Study–study Study–test Mean

5 minutes 0.80 0.75 0.78

2 days 0.55 0.70 0.63

1 week 0.42 0.55 0.49

Mean 0.59 0.67

36. (Table: Test-Enhanced Learning) The cells of this study reflect an interaction. Is this a quantitative or qualitative interaction? Why? a. This is a quantitative interaction because the effect of the study condition does not reverse depending on the retention interval. b. This is a qualitative interaction because the effect of the study condition does not reverse depending on the retention interval. c. This is a quantitative interaction because the effect of the study condition reverses depending on the retention interval. d. This is a qualitative interaction because the effect of the study condition reverses depending on the retention interval.

Page 7

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

37. (Table: Coffee and Sleep) For the data in the source table, what is the effect size for the interaction? a. small b. medium c. large d. too small to be of interest

Page 8

Name:

Class:

Date:

Chap_14_5e The following figure reflects the results of a study by Roediger and Karpicke (2006). The authors investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to the material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. Figure: Testing and Memory

38. (Figure: Testing and Memory) The figure reflects an interaction. Is this interaction a quantitative or qualitative interaction? Why? a. The figure reflects a quantitative interaction; the effect of the study condition does not reverse depending on the retention interval. b. The figure reflects a qualitative interaction; the effect of the study condition does not reverse depending on the retention interval. c. The figure reflects a quantitative interaction; the effect of the study condition reverses depending on the retention interval. With the 5-minute retention interval, study–study is better than study–test, but with longer retention intervals, study–test is better than study–study. d. The figure reflects a qualitative interaction; the effect of the study condition reverses depending on the retention interval. With the 5-minute retention interval, study–study is better than study–test, but with longer retention intervals, study–test is better than study–study.

Page 9

Name:

Class:

Date:

Chap_14_5e 39. Main effects refer to the: a. combined effects of two dependent variables. b. combined effects of two independent variables. c. effect of one level of the independent variable on the dependent variable, disregarding other levels of the independent variable. d. effect of a single independent variable on the dependent variable, disregarding all other variables in the study. 40. An interaction occurs when: a. the dependent variable does not depend on any of the independent variables. b. two independent variables both influence the dependent variable. c. the effects of one independent variable depend on the level of the other independent variable. d. a single independent variable changes the dependent variable, disregarding all other variables in the study. The following table reflects the results of a study by Forys and Dahlquist (2007) investigating the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. They were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were then instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. Table: Coping with Pain Distraction Monitoring Avoiding Mean

84.5 136.8 110.7

Sensation Monitoring 93.3 85.6 89.5

Mean 88.9 111.2

41. (Table: Coping with Pain) Based on the cells of this study, which effects appear to be present? a. a main effect of coping style and an interaction between coping style and cognitive strategy b. an interaction between coping style and cognitive strategy c. a main effect of coping style and a main effect of cognitive strategy d. a main effect of coping style, a main effect of cognitive strategy, and an interaction between coping style and cognitive strategy

Page 10

Name:

Class:

Date:

Chap_14_5e 42. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher asks an equal number of men and women to watch 0, 3, 5, and 9 hours of the Syfy channel each week for 6 weeks. At the end of each week, the researcher measures the "geekiness" of the men and women on a scale from 1 to 10. How should the researcher label the ANOVA used to analyze the data? a. a 2 × 4 × 6 mixed-design ANOVA b. a 2 × 4 × 6 mixed-design MANOVA c. a 4 × 2 within-groups ANOVA d. a 2 × 4 between-groups ANOVA 43. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the null hypothesis for the main effect of gender? a. On average, men and women are equally geeky. b. On average, men are geekier than women. c. On average, watchers are equally geeky regardless of the hours of the Syfy channel they watch. d. On average, men are geekier than women when watching fewer hours but less geeky than women when watching more hours of the Syfy channel. 44. Brunoni et al. (2006) investigated whether the combination of low-level electrical brain stimulation, transcranial direct current stimulation (tDCS), and sertraline, a commonly used SSRI antidepressant, would be effective in ameliorating depression in a group of clinically depressed individuals. The researchers randomly assigned participants to one of two medication conditions (placebo or sertraline) and to one of two brain electrical stimulation conditions (sham tDCS or active tDCS). How many cells does this study have? a. 2 b. 4 c. 6 d. 8

Page 11

Name:

Class:

Date:

Chap_14_5e 45. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher asks an equal number of men and women to watch 0, 3, 5, and 9 hours of the Syfy channel each week for 6 weeks. At the end of each week, the researcher measures the "geekiness" of the men and women on a scale from 1 to 10. How many cells are in this design? a. 6 b. 8 c. 12 d. 48 The following table reflects the results of a study by Forys and Dahlquist (2007) investigating the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. They were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were then instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. Table: Coping with Pain Distraction Monitoring Avoiding Mean

84.5 136.8 110.7

Sensation Monitoring 93.3 85.6 89.5

Mean 88.9 111.2

46. (Table: Coping with Pain) The cells of this study reflect an interaction between coping style and cognitive strategy. Which statement best describes the interaction? a. People using both distraction and sensation-monitoring cognitive strategies were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People using a sensation-monitoring strategy were able to keep their hand in the ice water for longer than people using a distraction strategy. c. People using a distraction strategy were able to keep their hand in the ice water for longer than people using a sensation-monitoring strategy. d. The effect of cognitive strategy depended on the coping style of the participant. Those with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensationmonitoring strategy.

Page 12

Name:

Class:

Date:

Chap_14_5e The following figure reflects the results of a study by Forys and Dahlquist (2007) investigating the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. They were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were next instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. Figure: Strategies for Dealing with Pain

47. (Figure: Strategies for Dealing with Pain) The figure reflects an interaction. Is this a quantitative or qualitative interaction? Why? a. This is a quantitative interaction because the effect of coping style does not reverse depending on the cognitive strategy employed. b. This is a qualitative interaction because the effect of coping style does not reverse depending on the cognitive strategy employed. c. This is a qualitative interaction because the effect of coping style reverses depending on the cognitive strategy employed. d. This is a quantitative interaction because the effect of cognitive strategy reverses depending on the coping style.

Page 13

Name:

Class:

Date:

Chap_14_5e 48. (Figure: Strategies for Dealing with Pain) The figure reflects a main effect of cognitive strategy. Which statement best describes the main effect? a. People using both distraction and sensation-monitoring cognitive strategies were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People using a sensation-monitoring strategy were able to keep their hand in the ice water for longer than people using a distraction strategy. c. People using a distraction strategy were able to keep their hand in the ice water for longer than people using a sensation-monitoring strategy. d. The effect of cognitive strategy depended on the coping style of the participant. People with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensationmonitoring strategy. 49. A two-way ANOVA accounts for _____ sources of variability. a. two b. three c. four d. five 50. Roediger and Karpicke (2006) investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. How should the analysis of these data be labeled? a. 4 × 2 between-groups ANOVA b. 2 × 3 within-groups ANOVA c. 2 × 3 between-groups ANOVA d. 2 × 2 within-groups ANOVA 51. A researcher performs a 2 × 3 ANOVA with seven participants in each cell of the study design. How many cells are in this design? a. 7 b. 6 c. 3 d. 2

Page 14

Name:

Class:

Date:

Chap_14_5e 52. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the null hypothesis for the main effect of hours watched? a. On average, men and women are equally geeky. b. On average, men are geekier than women. c. On average, watchers are equally geeky regardless of the hours of the Syfy channel they watch. d. On average, men are geekier than women when watching fewer hours but less geeky than women when watching more hours of the Syfy channel. 53. In a two-way factorial ANOVA, between-groups variance is divided into: a. two main effects and a single interaction. b. one main effect and two interactions. c. two main effects and two interactions. d. one main effect and a single interaction. 54. In addition to assessing whether each independent variable has an effect on the dependent variable, a factorial ANOVA allows the researcher to: a. use multiple dependent measures in a single analysis. b. control for a third variable that might be related to the dependent measure, prior to investigating the independent variable of interest. c. determine whether the effects of one factor depend on the other factor. d. partition out the variability due to individual differences and the variability due to measurement error. 55. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the research hypothesis for the interaction effect? a. Men and women are equally geeky, regardless of the hours of the Syfy channel they watched. b. The effect of geekiness is dependent on hours of the Syfy channel watched. c. The effect of geekiness is dependent on the gender of the Syfy channel watcher. d. The effect of hours of the Syfy channel watched is dependent on gender.

Page 15

Name:

Class:

Date:

Chap_14_5e The following table reflects the results of a study by Forys and Dahlquist (2007) investigating the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. They were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were then instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. Table: Coping with Pain Distraction Monitoring Avoiding Mean

84.5 136.8 110.7

Sensation Monitoring 93.3 85.6 89.5

Mean 88.9 111.2

56. (Table: Coping with Pain) The cells of this study reflect a main effect of cognitive strategy. Which statement best describes the main effect? a. People using both distraction and sensation-monitoring cognitive strategies were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People using a sensation-monitoring strategy were able to keep their hand in the ice water for longer than people using a distraction strategy. c. People using a distraction strategy were able to keep their hand in the ice water for longer than people using a sensation-monitoring strategy. d. The effect of cognitive strategy depended on the coping style of the participant. Those with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensationmonitoring strategy. 57. (Table: Coping with Pain) The cells of this study reflect a main effect of coping style. Which statement best describes the main effect? a. People with monitoring and avoiding coping styles were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People with an avoiding coping style were able to keep their hand in the ice water for longer than people with a monitoring coping style. c. People with a monitoring coping style were able to keep their hand in the ice water for longer than people with an avoiding coping style. d. The effect of coping style depended on the cognitive strategy employed. Those with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensation-monitoring strategy. Copyright Macmillan Learning. Powered by Cognero.

Page 16

Name:

Class:

Date:

Chap_14_5e 58. What form of ANOVA should be used to analyze data when there is more than one dependent variable? a. two-way analysis of variance b. analysis of covariance c. multivariate analysis of variance d. mixed-design analysis of variance 59. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is (are) the main effect(s) being tested in the study? a. gender b. hours of Syfy watched c. gender and hours of Syfy watched d. hours of Syfy watched and "geekiness"

Page 17

Name:

Class:

Date:

Chap_14_5e The following figure reflects the results of a study by Brunoni et al. (2006) that investigated whether the combination of low-level electrical brain stimulation, transcranial direct current stimulation (tDCS), and sertraline, a commonly used SSRI antidepressant, would be effective in ameliorating depression in a group of clinically depressed individuals. The researchers randomly assigned participants to one of two medication conditions (placebo or sertraline) and to one of two brain electrical stimulation conditions (sham tDCS or active tDCS). Participants' levels of depression was assessed using the Montgomery–Asberg Depression Rating Scale (MADRS) after six weeks of treatment. Figure: Brain Stimulation and Medication

60. (Figure: Brain Stimulation and Medication) Based on the figure, which effects appear to be present? a. a main effect of medication and an interaction between medication and electrical brain stimulation b. an interaction between medication and electrical brain stimulation c. a main effect of medication and a main effect of electrical brain stimulation d. a main effect of medication, a main effect of electrical brain stimulation, and an interaction between medication and electrical brain stimulation 61. A researcher performs a 2 × 4 ANOVA with seven participants in each cell of the study design. What are the within-groups degrees of freedom? a. 56 b. 48 c. 21 d. 18

Page 18

Name:

Class:

Date:

Chap_14_5e 62. Brunoni et al. (2006) investigated whether the combination of low-level electrical brain stimulation, transcranial direct current stimulation (tDCS), and sertraline, a commonly used SSRI antidepressant, would be effective in ameliorating depression in a group of clinically depressed individuals. The researchers randomly assigned participants to one of two medication conditions (placebo or sertraline) and to one of two brain electrical stimulation conditions (sham tDCS or active tDCS). How should the ANOVA used to analyze the data be labeled? a. 4 × 2 between-groups ANOVA b. 4 × 2 within-groups ANOVA c. 2 × 2 between-groups ANOVA d. 2 × 2 within-groups ANOVA 63. The covariate in analysis of covariance must be measured as what type of variable? a. nominal b. ordinal c. scale d. integer

Page 19

Name:

Class:

Date:

64. (Figure: Strategies for Dealing with Pain) The figure reflects an interaction between coping style and cognitive strategy. Which statement best describes the interaction? a. People using both distraction and sensation-monitoring cognitive strategies were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People using a sensation-monitoring strategy were able to keep their hand in the ice water for longer than people using a distraction strategy. c. People using a distraction strategy were able to keep their hand in the ice water for longer than people using a sensation monitoring strategy. d. The effect of cognitive strategy depended on the coping style of the participant. People with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensationmonitoring strategy.

Page 20

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

65. (Table: Coffee and Sleep) Using a p level of 0.05, what are the significant effects? a. a main effect of gender and an interaction between gender and the number of cups of coffee b. a main effect of the number of cups of coffee and no other effects c. a main effect of the number of cups of coffee and an interaction between gender and the number of cups of coffee d. a main effect of gender and a main effect of cups of coffee. 66. The correct formula for calculating the sum of squares for within-groups variability is: a. Σ(X – GM)2. b. Σ(Mrow or column – GM)2. c. Σ(X – Mcell)2. d. Σ(Mcell – X)2.

Page 21

Name:

Class:

Date:

67. (Figure: Testing and Memory) The figure reflects an interaction. Which statement best describes the interaction? a. On average, people in the study–test condition had better memory compared to people in the study– study condition. b. People in the study–test condition performed more poorly on the 5-minute recall test than did those in the study–study condition, but at longer retention intervals (2 days and 1 week); those in the study– test condition performed better than did those in the study–study condition. c. On average, memory performance decreased as the retention interval increased. d. On average, memory performance increased as the retention interval increased.

Page 22

Name:

Class:

Date:

Chap_14_5e 68. Forys and Dahlquist (2007) investigated the effects of coping style and cognitive strategy on dealing with pain. Participants were first classified as having a monitoring or avoiding coping style. Participants were then randomly assigned to one of two cognitive strategy conditions, distraction or sensation monitoring. Participants were next instructed to use the cognitive strategy while submerging their hand in ice water. The researchers measured pain tolerance as the number of seconds that participants were able to keep their hand in the ice water. How many cells does this study have? a. 2 b. 4 c. 6 d. 8 69. An effect size of 0.03 for a two-way ANOVA is considered to be: a. small. b. medium. c. large. d. an error.

Page 23

Name:

Class:

Date:

70. (Figure: Testing and Memory) Based on the cells of this study, which effects appear to be present? a. a main effect of study condition and an interaction between study condition and retention interval b. an interaction between study condition and retention interval c. a main effect of study condition and a main effect of retention interval d. a main effect of study condition, a main effect of retention interval, and an interaction between study condition and retention interval

Page 24

Name:

Class:

Date:

71. (Figure: Brain Stimulation and Medication) The figure reflects an interaction. Is this a quantitative or qualitative interaction? Why? a. This is a quantitative interaction because the effect of electrical brain stimulation does not reverse depending on the type of medication used. b. This is a qualitative interaction because the effect of electrical brain stimulation does not reverse depending on the type of medication used. c. This is a qualitative interaction because the effect of electrical brain stimulation reverses depending on the type of medication used. d. This is a quantitative interaction because the effect of electrical brain stimulation reverses depending on the type of medication used.

Page 25

Name:

Class:

Date:

72. (Figure: Strategies for Dealing with Pain) Based on the figure, which effects appear to be present? a. a main effect of coping style and an interaction between coping style and cognitive strategy b. an interaction between coping style and cognitive strategy c. a main effect of coping style and a main effect of cognitive strategy d. a main effect of coping style, a main effect of cognitive strategy, and an interaction between coping style and cognitive strategy

Page 26

Name:

Class:

Date:

Chap_14_5e 73. Brunoni et al. (2006) investigated whether the combination of low-level electrical brain stimulation, transcranial direct current stimulation (tDCS), and sertraline, a commonly used SSRI antidepressant, would be effective in ameliorating depression in a group of clinically depressed individuals. The researchers randomly assigned participants to one of two medication conditions (placebo or sertraline) and to one of two brain electrical stimulation conditions (sham tDCS or active tDCS). What is (are) the main effect(s) being tested in the study? a. type of brain electrical stimulation b. level of depression c. type of medication and depression d. type of brain electrical stimulation and type of medication 74. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the research hypothesis for the main effect of hours watched? a. On average, men and women differ in their geekiness levels. b. On average, watchers differ in their geekiness levels across the various hours of the Syfy channel watched. c. On average, watchers are equally geeky regardless of the hours of the Syfy channel they watch. d. On average, men are geekier than women when watching fewer hours but less geeky than women when watching more hours of the Syfy channel. 75. The correct formula for calculating the sum of squares for the between-groups variables, represented as rows or columns, is: a. Σ(X – GM)2. b. Σ(Mrow or column – GM)2. c. Σ(X – Mcell)2. d. Σ(Mcell – X)2. 76. Which design has more than one dependent variable? a. mixed-design ANOVA b. ANCOVA c. MANCOVA d. two-way ANOVA

Page 27

Name:

Class:

Date:

Chap_14_5e 77. Within-groups degrees of freedom in a two-way ANOVA is calculated by: a. multiplying together the degrees of freedom associated with each of the main effects. b. adding together the degrees of freedom associated with each of the main effects. c. subtracting the degrees of freedom for the first main effect from the degrees of freedom for the second main effect. d. subtracting a single participant from each cell of the study and then adding the results for all the cells. 78. A social psychologist researcher wants to statistically test whether gender affects math test scores. Which type of ANOVA design should the researcher use? a. between-groups b. within-groups c. mixed-groups d. repeated-measures

Page 28

Name:

Class:

Date:

79. (Figure: Strategies for Dealing with Pain) The figure reflects a main effect of coping style. Which statement best describes the main effect? a. People with monitoring and avoiding coping styles were able to keep their hand in the ice water for longer than 60 seconds, on average. b. People with an avoiding coping style were able to keep their hand in the ice water for longer than people with a monitoring coping style. c. People with a monitoring coping style were able to keep their hand in the ice water for longer than people with an avoiding coping style. d. The effect of coping style depended on the cognitive strategy employed. People with an avoiding coping style kept their hand in the ice water longer when using a distraction strategy, but those with a monitoring coping style kept their hand in the ice water longer when using a sensation-monitoring strategy.

Page 29

Name:

Class:

Date:

80. (Figure: Brain Stimulation and Medication) The figure reflects a main effect of electrical brain stimulation. Which statement best describes the main effect? a. People receiving active tDCS had scores on the MADRS of 16, on average. b. People treated with active tDCS had lower depression scores compared to people treated with sham tDCS. c. People treated with sertraline had lower depression scores compared to individuals treated with placebo. d. The reduction in depression scores as a result of active tDCS was greater for those individuals who received active medication as compared to those individuals who only received placebo. 81. The degrees of freedom for the interaction in a two-way ANOVA is calculated by: a. multiplying together the degrees of freedom associated with each of the main effects. b. adding together the degrees of freedom associated with each of the main effects. c. subtracting the degrees of freedom for the first main effect from the degrees of freedom for the second main effect. d. subtracting a single participant from each cell of the study and then adding the results for all the cells.

Page 30

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

82. (Table: Coffee and Sleep) What was the sample size for the entire study? a. 30 b. 35 c. 36 d. 40 83. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become :geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the research hypothesis for the main effect of gender? a. On average, men and women are equally geeky. b. On average, men and women differ in their geekiness levels. c. On average, men are geekier when they watch more hours of the Syfy channel than fewer hours. d. On average, men are geekier than women when watching fewer hours but less geeky than women when watching more hours of the Syfy channel.

Page 31

Name:

Class:

Date:

84. (Figure: Testing and Memory) The figure reflects a main effect of study condition. Which statement best describes the main effect? a. On average, people in the study–test condition had better memory compared to people in the study– study condition. b. People in the study–test condition performed more poorly on the 5-minute recall test than did those in the study–study condition, but at longer retention intervals (2 days and 1 week); those in the study– test condition performed better than did those in the study–study condition. c. On average, memory performance decreased as the retention interval increased. d. On average, people in the study–study condition had better memory compared to people in the study–test condition. 85. An effect size of 0.07 for a two-way ANOVA is considered to be: a. small. b. medium. c. large. d. an error.

Page 32

Name:

Class:

Date:

86. (Figure: Brain Stimulation and Medication) The figure reflects a main effect of medication. Which statement best describes the main effect? a. People receiving active tDCS had scores on the MADRS of 16, on average. b. People treated with sham tDCS had lower depression scores compared to people treated with active tDCS. c. People treated with sertraline had lower depression scores compared to individuals treated with placebo. d. The reduction in depression scores as a result of active tDCS was greater for those individuals who received active medication as compared to those individuals who only received placebo.

Page 33

Name:

Class:

Date:

Chap_14_5e The following table reflects the results of a study by Roediger and Karpicke (2006). The authors investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to the material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. Table: Test-Enhanced Learning

Study–study Study–test Mean

5 minutes 0.80 0.75 0.78

2 days 0.55 0.70 0.63

1 week 0.42 0.55 0.49

Mean 0.59 0.67

87. (Table: Test-Enhanced Learning) The cells of this study reflect a main effect of study condition. Which statement best describes the main effect? a. On average, people in the study–test condition had better memory compared to people in the study– study condition. b. People in the study–test condition performed more poorly on the 5-minute recall test than did those in the study–study condition, but at longer retention intervals (2 days and 1 week); those in the study– test condition performed better than did those in the study–study condition. c. On average, memory performance decreased as the retention interval increased. d. On average, people in the study–study condition had better memory compared to people in the study–test condition. 88. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. How many cells does this study have? a. 2 b. 4 c. 6 d. 8

Page 34

Name:

Class:

Date:

Chap_14_5e 89. If at least one independent variable is between-groups and at least one is within-groups, the design is referred to as: a. an interaction. b. one-way. c. mixed. d. two-way. 90. An effect size of 0.21 for a two-way ANOVA is considered to be: a. small. b. medium. c. large. d. an error. The following source table depicts the results of a fictional study investigating whether the number of hours of sleep a person gets varies with gender (male, female) and with the number of cups of coffee consumed in a day. Equal numbers of men and women were randomly assigned to drink 1, 2, or 3 cups of coffee during the course of a day and then record the number of hours they slept that night. Table: Coffee and Sleep Source Gender Cups of coffee Gender × Cups Within Total

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

91. (Table: Coffee and Sleep) For the data in the source table, what is the size of the effect for gender? a. small b. medium c. large d. too small to be of interest 92. (Table: Coffee and Sleep) Using a p level of 0.01, the critical value for the main effect of gender is _____, and the critical value for the main effect of cups of coffee is _____. a. 4.17; 3.32 b. 4.17; 5.39 c. 5.39; 7.56 d. 7.56; 5.39

Page 35

Name:

Class:

Date:

Chap_14_5e 93. (Table: Coffee and Sleep) For the data in the source table, what is the effect size for gender? a. 0.01 b. 0.10 c. 0.31 d. 0.69 94. For a two-way ANOVA, _____ is the measure of effect size. a. Cohen's d b. R2 c. r d. R – 1 95. A researcher performs a 2 × 4 ANOVA with seven participants in each cell of the study design. What are the degrees of freedom for the interaction? a. 2 b. 3 c. 4 d. 6 96. An effect size of 0.134 for a two-way ANOVA is considered to be: a. small. b. medium. c. large. d. too small to be of interest. 97. The correct formula for calculating the total sum of squares is: a. Σ(N – 1)/2. b. Σ(X – GM)2. c. Σ(X – GM)/(N – 1). d. Σ(X – GM)/N. 98. _____ is a term used to describe an independent variable in a study with more than one independent variable. a. Interaction b. Subjects effect c. Effect d. Factor

Page 36

Name:

Class:

Date:

Chap_14_5e 99. Analysis of covariance involves: a. analyzing more than one dependent variable. b. analyzing the residuals in an interaction. c. determining all relevant variables that are correlated with the dependent variable prior to collecting the data. d. statistically controlling for a third variable that is associated with the dependent measure. 100. The owners of the Syfy channel are interested in whether watching the Syfy channel causes people to become "geeky," and if so, whether any such effects depend on a person's gender. They hire a researcher to design and carry out the appropriate study. The researcher randomly assigns an equal number of men and women to watch 0, 3, 5, or 9 hours of the Syfy channel each week for 6 weeks. What is the null hypothesis for the interaction effect? a. Men and women are equally geeky, regardless of the hours of the Syfy channel they watched. b. The effect of geekiness is not dependent on hours of the Syfy channel watched. c. The effect of geekiness is not dependent on the gender of the Syfy channel watcher. d. The effect of hours of the Syfy channel watched is not dependent on gender. 101. Roediger and Karpicke (2006) investigated whether the test-enhanced learning effect (the demonstration that repeated testing improves memory for material) was due merely to repeated exposure to material. They randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. How many cells does this study have? a. 2 b. 4 c. 6 d. 10

Page 37

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

102. (Table: Coffee and Sleep) Using a p level of 0.05, the critical value for the main effect of gender is _____, and the critical value for the main effect of cups of coffee is _____. a. 3.32; 3.32 b. 4.13; 3.28 c. 4.13; 4.13 d. 4.17; 3.32 103. If an analysis of variance includes both within-groups factors and between-groups factors, it is called a _____ ANOVA. a. two-way b. complex-design c. between-within d. mixed-design 104. In a two-way ANOVA, if there is a significant interaction: a. at least one of the main effects is significant. b. the interpretation of significant main effects is independent of the interaction. c. the interpretation of any significant main effects should be ignored. d. both of the main effects are also significant. 105. A(n) _____ occurs when _____ independent variable(s) have an effect on the dependent variable in combination with one another. a. interaction; two or more b. interaction; two or less c. main effect; two or more d. main effect; two or less Copyright Macmillan Learning. Powered by Cognero.

Page 38

Name:

Class:

Date:

106. (Figure: Brain Stimulation and Medication) The figure reflects an interaction between medication and electrical stimulation. Which statement best describes the interaction? a. People receiving active tDCS had scores on the MADRS of 16, on average. b. People treated with sham tDCS had lower depression scores compared to people treated with active tDCS. c. People treated with sertraline had lower depression scores compared to individuals treated with placebo. d. The reduction in depression scores as a result of active tDCS was greater for those individuals who received active medication as compared to those individuals who received placebo. 107. School administrators in Boston want to assess the efficacy of a new reading program they wish to put in place. They randomly assign some schools in the city to institute the new reading program and other schools to remain on the traditional reading program. When performing the analysis to determine whether students using the new reading program perform better, the educators control for socioeconomic status, a variable known to be related to educational outcomes. What kind of analysis have the educators performed? a. two-way ANOVA b. MANOVA c. ANCOVA d. one-way ANOVA Copyright Macmillan Learning. Powered by Cognero.

Page 39

Name:

Class:

Date:

SS 0.38 75.38 3.66 34.51 113.93

df 1 2 2 30 35

MS 0.38 37.69 1.83 1.15

F 0.33 32.76 1.59

108. (Table: Coffee and Sleep) For the data in the source table, what is the effect size for cups of coffee? a. 0.01 b. 0.10 c. 0.31 d. 0.69 109. In a table depicting the cells of a two-way ANOVA, the marginal means reflect: a. the interaction. b. the two main effects. c. one of the main effects. d. within-groups variance.

Page 40

Name:

Class:

Date:

Study–study Study–test Mean

5 minutes 0.80 0.75 0.78

2 days 0.55 0.70 0.63

1 week 0.42 0.55 0.49

Mean 0.59 0.67

110. (Table: Test-Enhanced Learning) Based on the cells of this study, which effects appear to be present? a. a main effect of study condition and an interaction between study condition and retention interval b. an interaction between study condition and retention interval c. a main effect of study condition and a main effect of retention interval d. a main effect of study condition, a main effect of retention interval, and an interaction between study condition and retention interval

Page 41

Name:

Class:

Date:

84.5 136.8 110.7

Sensation Monitoring 93.3 85.6 89.5

Mean 88.9 111.2

111. (Table: Coping with Pain) The cells of this study reflect an interaction. Is it a quantitative or qualitative interaction? Why? a. This is a quantitative interaction because the effect of coping style does not reverse depending on the cognitive strategy employed. b. This is a qualitative interaction because the effect of coping style does not reverse depending on the cognitive strategy employed. c. This is a qualitative interaction because the effect of cognitive strategy reverses depending on the coping style. d. This is a quantitative interaction because the effect of cognitive strategy reverses depending on the coping style. Enter the appropriate word(s) to complete the statement. 112. An effect size statistic of 0.08 is considered to be _______ for a two-way between-groups ANOVA, according to Cohen's conventions.

113. A 3 × 2 ANOVA would have _______ cells.

Page 42

Name:

Class:

Date:

Chap_14_5e 114. _____ is the effect size statistic for a two-way between-groups ANOVA.

115. An analysis of variance with multiple dependent variables is called a(n) _______.

116. A statistical _______ occurs when two or more independent variables have an effect on the dependent variable in combination.

117. An analysis of variance with multiple dependent variables and a covariate is called a(n) _______.

118. A box that depicts the unique combination of levels of the independent variables in a factorial design is called a(n) _______.

119. A 3 × 3 ANOVA would have _______ cells.

Page 43

Name:

Class:

Date:

Chap_14_5e 120. A(n) _______ interaction occurs when the effect of one independent variable is strengthened or weakened at one or more levels of the other independent variable, but the direction of the initial effect does not change.

121. A quantitative interaction occurs when the effect of one independent variable is _______ at one or more levels of the other independent variable, but the direction of the initial effect does not change.

122. In an ANOVA with more than one independent variable, the independent variables may sometimes be referred to as _______.

123. When an interaction is NOT present, the bar graph for a two-way ANOVA will have lines that are _______.

124. In a two-way ANOVA, _______ F statistics are calculated.

125. An analysis of variance with three independent variables is a(n) _______ ANOVA.

Page 44

Name:

Class:

Date:

Chap_14_5e 126. If a study includes both between-groups factors and within-groups factors, a(n) _______ ANOVA is used to analyze the data.

127. In a two-way ANOVA, a(n) _______ occurs when one of the independent variables has an influence on the dependent variable.

128. _______ is the mean of a row or a column in a two-way ANOVA.

129. A qualitative interaction occurs when the direction of the effect of one independent variable _______ depending on the level of the other independent variable.

130. A(n) _______ interaction occurs when the direction of the effect of one independent variable reverses depending on the level of the other independent variable.

131. When an interaction is present, the bar graph for a two-way ANOVA will have lines that _______ at some point.

Page 45

Name:

Class:

Date:

Chap_14_5e 132. An analysis of variance with one dependent variable and a covariate is called a(n) _______.

133. In an ANOVA with more than one independent variable, the _______variables may sometimes be referred to as factors.

134. An effect size statistic of 0.17 is considered to be _______ for a two-way between-groups ANOVA, according to Cohen's conventions.

135. An analysis of variance with two independent variables is a(n) _______ ANOVA.

136. The Harvard School of Public Health reviewed studies investigating the relation between alcohol consumption and health. One of these studies investigated genetic factors as a moderator of the influence of alcohol on individual health. Half the participants had a slow-acting version of the ADH3 enzyme, and the other half had a fast-acting version of the enzyme. The researchers found that, compared to people who did not drink at all, people who drank a moderate amount of alcohol (1 or 2 drinks per day) and who had the slow-acting version of the enzyme showed better cardiovascular health than people who either did not drink at all or drank moderately and had the fast-acting version of the enzyme. (a) Identify the independent variables and the level of each independent variable. (b) What is the dependent variable? (c) What kind of ANOVA would be used to analyze these data? (d) Draw a table depicting the cells of the study. (e) Interpret the results of the study. What were the significant main effects or interactions?

Page 46

Name:

Class:

Date:

Chap_14_5e 137. A team of researchers are interested in investigating impairments in driving performance as a result of cannabis and alcohol consumption. Participants are tested in a driving simulator with varying doses of alcohol and cannabis. Participants consumed orange juice either with or without alcohol and then smoked cannabis cigarettes containing no THC, 1.8% THC, or 3% THC. Participants were then tested in a driving simulator, and the number of driving errors committed during a 30-minute test drive. (a) Identify the independent variables and the level of each independent variable. (b) What is the dependent variable? (c) What kind of ANOVA would be used to analyze these data? (d) Draw a table depicting the cells of the study.

138. The Syfy channel is interested in whether watching Syfy causes people to become geeky and whether such an effect depends on the viewer being male or female. Researchers randomly assigned 12 women and 12 men to watch either 0 hours or 5 hours of the Syfy channel each week for 6 weeks, and then asked all participants to complete a personality inventory designed to detect "geeky" personalities. The data for the experiment follow. Higher scores on the personality inventory indicate a greater amount of "geekiness." Female, 0 hours: 3, 7, 2, 3, 4, 5 Female, 5 hours: 7, 7, 3, 5, 8, 6 Male, 0 hours: 8, 4, 6, 8, 4, 6 Male, 5 hours: 10, 14, 10, 15, 12, 11 (a) Draw a table depicting the cells of the study, and include the appropriate means and standard deviations in each cell. (b) Conduct the six steps of hypothesis testing for this set of data. (c) Graph and interpret any significant effects.

139. Dr. Harwick administered tests to a classroom of students to assess gender differences on the test scores. The students completed a math test and a reading test. Dr. Harwick computed total scores for both groups of subjects and hypothesized that female students would perform higher on math and reading compared to male students. Should Dr. Harwick use a one-way or two-way ANOVA to test his hypothesis? Explain your answer.

Page 47

Name:

Class:

Date:

Chap_14_5e 140. A study by Jackman-Cram, Dobson, and Martin (2006) investigated the communication behavior of married couples during problem-solving discussions. The researchers first classified participant couples according to their level of marital distress (distressed, nondistressed). They then classified the couples on the basis of whether one of the spouses was suffering depression (depressed, nondepressed). All couples were videotaped during problem-solving discussions, and all videotapes were then coded by the researchers using the Marital Interaction Coding System. Among the dependent measures was the proportion of communications that were "facilitative" in nature. The following fictional data reproduce the pattern of results obtained for the women partners in the original research. Distressed, depressed: 0.45, 0.15, 0.22, 0.30, 0.25, 0.18 Nondistressed, depressed: 0.60, 0.32, 0.49, 0.39, 0.41, 0.45 Distressed, nondepressed: 0.31, 0.18, 0.42, 0.20, 0.17, 0.22 Nondistressed, nondepressed: 0.52, 0.45, 0.61, 0.36, 0.42, 0.46 (a) Draw a table depicting the cells of the study and include the appropriate means in each cell. (b) Conduct steps 4 through 6 of hypothesis testing for this set of data. (c) Graph and interpret any significant effects.

141. A developmental psychologist is interested in whether the types of video games children have played affect the number of violent images the children report in their dreams and stories, and whether this effect varies if the children are hyperactive. To test her hypothesis, she obtains a sample of 10 hyperactive children and 10 nonhyperactive children. She randomly assigns the hyperactive and nonhyperactive children to play either a violent video game (Grand Theft Auto) or a prosocial game (Soccer) for an hour every day for a month, and then she codes the number of violent images in the children's dreams, stories, and pictures. The data follow: Hyperactive, violent game: 10, 13, 12, 10, 11 Hyperactive, prosocial game: 4, 6, 3, 4, 3 Nonhyperactive, violent game: 8, 11, 7, 9, 10 Nonhyperactive, prosocial game: 5, 3, 6, 5, 4 (a) Draw a table depicting the cells of the study, and include the appropriate means and standard deviations in each cell. (b) Conduct steps 4 through 6 of hypothesis testing for this set of data. (c) Graph and interpret any significant effects.

Page 48

Name:

Class:

Date:

Chap_14_5e 142. (Table: Test-Enhanced Learning and Memory) Roediger and Karpicke (2006) investigated whether the testenhanced learning effect (repeated testing improves memory for material) was due merely to repeated exposure to the material. The researchers randomly assigned participants to one of two study conditions (study–study or study–test) and to one of three retention interval conditions (final test at a delay of 5 minutes, 2 days, or 1 week). The dependent variable was the proportion of idea units recalled from an encyclopedia passage. The statistics reported by the researchers are as follows, and the mean number of idea units recalled appears in the table shown. Effect of study condition, F(1, 117) = 36.39, p < 0.05 Effect of retention interval, F(2, 117) = 50.34, p < 0.05 Condition by retention interval, F(2, 117) = 32.10, p < 0.05 Table: Test-Enhanced Learning and Memory

Study–study Study–test

5 minutes 0.80 0.75

2 days 0.55 0.70

1 week 0.42 0.55

(a) What significant effects did the authors find? (b) Use the cell means to graph and then interpret the results of the study.

Page 49

Name:

Class:

Date:

Chap_14_5e 143. A developmental psychologist is interested in whether the types of video games children have played affect the number of violent images the children report in their dreams and stories, and whether this effect varies if the children are hyperactive. To test her hypothesis, she obtains a sample of 10 hyperactive children and 10 nonhyperactive children. She randomly assigns the hyperactive and nonhyperactive children to play either a violent video game (Grand Theft Auto) or a prosocial game (Soccer) for an hour every day for a month, and then she codes the number of violent images in the children's dreams, stories, and pictures. The data follow: Hyperactive, violent game: 10, 13, 12, 10, 11 Hyperactive, prosocial game: 4, 6, 3, 4, 3 Nonhyperactive, violent game: 8, 11, 7, 9, 10 Nonhyperactive, prosocial game: 5, 3, 6, 5, 4 Table: Video Games and Violent Images 2

Use the information provided to calculate effect size measures for each of the three effects. Interpret the meaning of each R2 value.

144. Do the health benefits of alcohol consumption depend on the type of alcohol and the amount consumed? The Harvard School of Public Health reviewed studies investigating the relation between alcohol consumption and health. One study investigated whether different types of alcoholic beverages lead to different health outcomes. Based on self-report measures, the researchers classified a sample of participants who were older than age 30 as either moderate drinkers (1 or 2 drinks per day) or heavy drinkers (more than 2 drinks per day). The researchers also classified participants by type of alcoholic beverage typically consumed (beer, white wine, red wine, or hard liquor). The researchers then measured participants' cardiovascular health. Regardless of the type of alcoholic beverage typically consumed, participants who were moderate drinkers had, on average, better cardiovascular health than participants who were heavy drinkers. (a) Identify the independent variables and the level of each independent variable. (b) What is the dependent variable? (c) What kind of ANOVA would be used to analyze these data? (d) Draw a table depicting the cells of the study. (e) Interpret the results of the study. What were the significant main effects or interactions? What additional analyses would be needed to specify results in detail?

Page 50

Name:

Class:

Date:

Chap_14_5e 145. A researcher is interested in investigating whether the number of hours of sleep a person gets varies with gender (male, female) and with the number of cups of coffee the person consumes in a day. Equal numbers of men and women were asked to drink 1, 2, and 3 cups of coffee during the course of a day (on different days) and then record the number of hours they slept that night. (a) Identify the independent variables and the level of each independent variable. (b) What is the dependent variable? (c) What kind of ANOVA would be used to analyze these data? (d) Draw a table depicting the cells of the study.

146. In a two-way ANOVA, three assumptions must be met for appropriate conclusions to be drawn: (1) participants must be randomly selected, (2) the distribution must be normal, and (3) there must be equality of variance. What are some possible dangers from not assessing for normality in a two-way ANOVA?

Page 51

Name:

Class:

Date:

Chap_14_5e 147. A study by Jackman-Cram, Dobson, and Martin (2006) investigated the communication behavior of married couples during problem-solving discussions. The researchers first classified participant couples according to their level of marital distress (distressed, nondistressed). They then classified the couples on the basis of whether one of the spouses was suffering depression (depressed, nondepressed). All couples were videotaped during problem-solving discussions, and all videotapes were then coded by the researchers using the Marital Interaction Coding System. Among the dependent measures was the proportion of communications that were "facilitative" in nature. The following fictional data reproduce the pattern of results obtained for the women partners in the original research. Distressed, depressed: 0.45, 0.15, 0.22, 0.30, 0.25, 0.18 Nondistressed, depressed: 0.60, 0.32, 0.49, 0.39, 0.41, 0.45 Distressed, nondepressed: 0.31, 0.18, 0.42, 0.20, 0.17, 0.22 Nondistressed, nondepressed: 0.52, 0.45, 0.61, 0.36, 0.42, 0.46 Table: Communication and Marital Distress 2 Source Marital distress Depression Distress × Depression Within Total

SS 0.246 0.001 0.002

df 1 1 1

MS 0.246 0.001 0.002

0.188 0.437

20 23

0.009

F 26.144 0.054 0.195

Use the information provided to calculate effect size measures for each of the three effects. Interpret the meaning of each R2 value.

148. A researcher wants to statistically test whether there are gender differences in participant responses to stress. The researcher exposed all participants to five stress conditions (no stress, public speaking stress, heat stress, cold stress, noise stress). The researcher hypothesized that participants' stress levels, as measured by salivary cortisol levels, would be affected by type of stress and by gender differences. What are the independent and dependent variables in the study? How many levels does the independent variable have? Which test should the researcher use to test her hypothesis? Explain your answer.

Page 52

Name:

Class:

Date:

Chap_14_5e 149. Using a sample of eighth-grade students, Yenlimez, Sungur, and Tekkaya (2006) investigated the influence of reasoning ability on students' achievement test scores while controlling for their grades in science class. Identify the covariate used in the described analysis and explain why you think the researchers chose to use that particular covariate.

150. (Table: Sertraline and Electrical Current Therapy) Brunoni et al. (2006) investigated whether the combination of low-level electrical brain stimulation, transcranial direct current stimulation (tDCS), and sertraline, a commonly used SSRI antidepressant, would be effective in ameliorating depression in a group of clinically depressed individuals. The researchers randomly assigned participants to one of two medication conditions (placebo or sertraline) and to one of two brain electrical stimulation conditions (sham tDCS or active tDCS). The dependent variable was the individual's depression score on the Montgomery–Asberg Depression Rating Scale (MADRS) after six weeks of treatment. The statistics reported by the researchers are as follows, and the average MADRS scores appear in the table shown. Effect of medication, F(1, 116) = 5.15, p < 0.05 Effect of tDCS, F(1, 116) = 12.85, p < 0.05 Medication by tDCS, F(1, 116) = 0.51, p > 0.05 Table: Sertraline and Electrical Current Therapy

Sham tDCS Active tDCS

Placebo 24.73 21.67

Sertraline 19.07 13.17

(a) What significant effects did the authors find? (b) Use the cell means to graph and then interpret the results of the study.

Page 53

Name:

Class:

Date:

Chap_14_5e Answer Key 1. False 2. False 3. False 4. True 5. False 6. False 7. False 8. True 9. True 10. False 11. True 12. False 13. True 14. True 15. True 16. False 17. True 18. False 19. True 20. True 21. True 22. True 23. False 24. False 25. b 26. b Copyright Macmillan Learning. Powered by Cognero.

Page 54

Name:

Class:

Date:

Chap_14_5e 27. c 28. b 29. c 30. b 31. a 32. b 33. a 34. d 35. a 36. d 37. b 38. d 39. d 40. c 41. d 42. a 43. a 44. b 45. d 46. d 47. c 48. c 49. c 50. c 51. b 52. c 53. a 54. c Copyright Macmillan Learning. Powered by Cognero.

Page 55

Name:

Class:

Date:

Chap_14_5e 55. d 56. c 57. b 58. c 59. c 60. d 61. b 62. c 63. c 64. d 65. b 66. c 67. b 68. b 69. a 70. d 71. a 72. d 73. d 74. b 75. b 76. c 77. d 78. a 79. b 80. b 81. a 82. c Copyright Macmillan Learning. Powered by Cognero.

Page 56

Name:

Class:

Date:

Chap_14_5e 83. b 84. a 85. b 86. c 87. a 88. d 89. c 90. c 91. d 92. d 93. a 94. b 95. b 96. c 97. b 98. d 99. d 100. d 101. c 102. d 103. d 104. c 105. a 106. d 107. c 108. d 109. b 110. d Copyright Macmillan Learning. Powered by Cognero.

Page 57

Name:

Class:

Date:

Chap_14_5e 111. c 112. medium 113. six, 6 114. 115. MANOVA 116. interaction 117. MANCOVA 118. cell 119. nine, 9 120. quantitative 121. strengthened or weakened 122. factors 123. parallel 124. three, 3 125. three-way 126. mixed-design 127. main effect 128. Marginal mean 129. reverses 130. qualitative 131. intersect 132. ANCOVA 133. independent 134. large 135. two-way

Page 58

Name:

Class:

Date:

Chap_14_5e 136. (a) The first independent variable was a version of the ADH3 enzyme (slow acting, fast acting), and the second independent variable was the number of drinks a person consumed per day (none, moderate number). (b) The dependent variable was cardiovascular health. (c) A 2 × 2 between-groups ANOVA would be used to analyze these data. (d) (Table: Drinking and ADH3 Enzyme) The table depicts the cells of the study. Table: Drinking and ADH3 Enzyme Slow acting Slow, none Slow, moderate

No drinks Moderate drinker

Fast acting Fast, none Fast, moderate

(e) The authors found no main effects, but did discover a significant interaction between the type of enzyme a participant had and the number of drinks the participant consumed. Moderate drinkers who had the slow-acting enzyme showed better cardiac health than did nondrinkers with the same type of enzyme. For participants with the fast-acting enzyme, however, the amount they consumed did not affect their cardiovascular health. 137. (a) The first independent variable is alcohol (no alcohol, alcohol), and the second independent variable is the dose of cannabis (0%, 1.8%, 3%). (b) The dependent variable is the number of driver errors. (c) A 2 × 3 betweengroups ANOVA would be used to analyze these data. (d) (Table: Alcohol and Cannabis) The table depicts the cells of this study. Table: Alcohol and Cannabis

No alcohol Alcohol

0% THC No alcohol, 0% THC Alcohol, 0% THC

1.8% THC No alcohol, 1.8% THC Alcohol, 1.8% THC

3% THC No alcohol, 3% THC Alcohol, 3% THC

138. (a) (Table: Gender and Syfy 1) The answer should look like the table. Table: Gender and Syfy 1

Male Female

0 hours M = 6.0, SD = 1.79 M = 4.0, SD = 1.79

5 hours M = 12.0, SD = 2.10 M = 6.0, SD = 1.79

(b) Step 1: The four populations are (1) men who have not watched the Syfy channel, (2) men who have watched the Syfy channel for 5 hours per week, (3) women who have not watched the Syfy channel, and (4) women who have watched the Syfy channel for 5 hours per week. The comparison distribution will be an F distribution, and we will perform a 2 × 2 between-groups ANOVA. Some, but not all, of our assumptions are met. Participants were not randomly selected. It is unclear whether the distribution of the population is normal, but the homogeneity of variance assumption is met because the sample sizes are equal, and the largest variance is not more than five times the smallest variance. Copyright Macmillan Learning. Powered by Cognero.

Page 59

Name:

Class:

Date:

Chap_14_5e Step 2: Main effect of gender Null hypothesis: Levels of geekiness do not vary as a function of gender. Research hypothesis: Levels of geekiness vary as a function of gender. Main effect of hours watched Null hypothesis: Levels of geekiness do not vary as a function of the number of hours a person watches the Syfy channel. Research hypothesis: Levels of geekiness vary as a function of the number of hours a person watches the Syfy channel. Interaction of gender and hours watched Null hypothesis: The effect on geekiness of gender or hours watched do not depend on the level of the other variable. Research hypothesis: The effect on geekiness of gender or hours watched depend on the level of the other variable. Step 3: There will be three comparison distributions: (1) Main effect of gender: The comparison distribution is an F distribution with df = (1, 20). (2) Main effect of hours watched: The comparison distribution is an F distribution with df = (1, 20). (3) Interaction: The comparison distribution is an F distribution with df = (1, 20). Step 4: Because all the effects we are testing have the same comparison distribution, if we choose a p level of 0.05, the F critical value for all three hypothesis tests is 4.35. Step 5: (Table: Gender and Syfy 2) The answer should look like the table. Table: Gender and Syfy 2 Source Gender Hours watched Gender × Hours Within Total

SS 96.00 96.00 24.00 70.00 286.00

df 1 1 1 20 23

MS 96.00 96.00 24.00 3.50

F 27.429 27.429 6.857

Step 6: All three effects are significant. There is a main effect of gender, F(1, 20) = 27.429, p < 0.05; a main effect of the number of hours of Syfy watched, F(1, 20) = 27.429, p < 0.05; and an interaction between gender and the number of hours of Syfy watched, F(1, 20) = 6.857, p < 0.05. (c) (Figure: Gender and Syfy) The graph indicates that watching the Syfy channel caused both men and women to become more geeky, but this effect was stronger for men than for women. Figure: Gender and Syfy

Page 60

Name:

Class:

Date:

Chap_14_5e

139. Dr. Harwick should not use either a one-way or a two-way ANOVA. To use a one-way ANOVA, there must be only one independent variable and one dependent variable. To use a two-way ANOVA, there must be two independent variables and one dependent variable. Dr. Nelson's study has one independent variable (gender) and two dependent variables (math and reading scores). When there is more than one dependent variable, the statistic that should be used is a MANOVA. 140. (a) (Table: Communication and Marital Distress 1) The answer should look like the table. Table: Communication and Marital Distress 1

Distressed Nondistressed

Depressed M = 0.26, SD = 0.11 M = 0.44, SD = 0.10

Nondepressed M = 0.25, SD = 0.10 M = 0.47, SD = 0.09

(b) Step 4: The comparison distribution for all three hypothesis tests is an F distribution with df = (1, 20). Therefore the F critical value for all three hypothesis tests is 4.35. Step 5: (Table: Communication and Marital Distress 2) The answer should look like the table. Table: Communication and Marital Distress 2 Source Marital distress Depression Distress × Depression Within Total

SS 0.246 0.001 0.002

df 1 1 1

MS 0.246 0.001 0.002

0.188 0.437

20 23

0.009

F 26.144 0.054 0.195

Step 6: The only significant effect is the main effect of marital distress, F(1, 20) = 26.144, p < 0.05. (c) (Figure: Communication and Marital Distress) The graph reveals that wives in couples experiencing marital Copyright Macmillan Learning. Powered by Cognero.

Page 61

Name:

Class:

Date:

Chap_14_5e distress engage in fewer facilitative communications than do wives in couples that are not experiencing marital distress. Figure: Communication and Marital Distress

141. (a) (Table: Video Games and Violent Images 1) The answer should look like the table. Table: Video Games and Violent Images 1

Hyperactive Nonhyperactive

Prosocial game M = 4.00, SD = 1.22 M = 4.60, SD = 1.14

Violent game M = 11.20, SD = 1.30 M = 9.0, SD = 1.58

(b) Step 4: The comparison distribution for all three hypothesis tests is an F distribution with df = (1, 16). Therefore the F critical value for all three hypothesis tests is 4.49. Step 5: (Table: Video Games and Violent Images 2) The answer should look like the table. Table: Video Games and Violent Images 2 Source Game type Hyperactivity Copyright Macmillan Learning. Powered by Cognero.

SS 168.20 3.20

df 1 1

MS 168.20 3.20

F 96.11 1.83 Page 62

Name:

Class:

Date:

Chap_14_5e Game × Hyperactivity Within Total

9.80

28.00 209.20

16 19

1.75

5.60

Step 6: There is a significant main effect of game type, F(1, 16) = 96.11, p < 0.05, and a significant interaction between game type and hyperactivity, F(1, 16) = 5.60, p < 0.05. (c) (Figure: Video Games and Violent Images) The graph reveals that playing the violent video game produced an increase in the number of violent images reported and that this effect was larger for hyperactive children than for nonhyperactive children. Figure: Video Games and Violent Images

Page 63

Name:

Class:

Date:

Chap_14_5e 142. (a) The authors found a significant main effect of study condition, a significant main effect of retention interval, and a significant interaction between study condition and retention interval. (b) (Figure: Test-Enhanced Learning and Memory) The graph reveals that at the 5-minute retention interval, the group who studied twice performed better on the memory test, but at longer retention intervals, the group who studied and tested performed better on the memory test. Figure: Test-Enhanced Learning and Memory

143. For the main effect of game type, R2 = 168.20 /(209.20 – 3.20 – 9.80) = 0.86, which is a very large effect. It makes sense that a very large effect would be statistically significant, even with limited sample size. For the main effect of hyperactivity, R2 = 3.20 /(209.20 – 168.20 – 9.80) = 0.10, which is a medium to large effect. However, we did not find a significant effect, which means there might be an issue with the calculations of F, too small of a sample size, or the need for replicating the study. For the interaction, R2 = 9.80/(209.20 – 168.20 – 3.20) = 0.26, which is a large effect and consistent with the significant F. If the researcher really thinks something is going on with these variables, she might want to collect more data to assess the three effects with greater confidence.

Page 64

Name:

Class:

Date:

Chap_14_5e 144. (a) The first independent variable is type of alcohol (beer, white wine, red wine, hard liquor). The second independent variable is type of drinker (moderate, heavy). (b) The dependent variable is cardiovascular health. (c) A 2 × 4 between-groups ANOVA would be used to analyze these data. (d) (Table: Drinking and Cardiovascular Health) The table depicts the cells of the study. Table: Drinking and Cardiovascular Health

Moderate Heavy

Beer White wine Red wine Moderate, beer Moderate, white Moderate, red Heavy, beer Heavy, white Heavy, red

Liquor Moderate, liquor Heavy, liquor

(e) The authors found a significant main effect of type of drinker but no other effects. Moderate drinkers had better cardiovascular health than heavy drinkers, and it did not matter what type of alcohol a drinker consumed. To capture where the between-group differences occurred, post hoc tests would be needed for type of alcohol consumed. 145. (a) The first independent variable is gender (male, female), and the second independent variable is the number of cups of coffee consumed (1, 2, 3). (b) The dependent variable is the number of hours slept. (c) A 2 × 3 mixeddesign ANOVA would be used to analyze these data. (d) (Table: Gender and Coffee) The table depicts the cells of this study. Table: Gender and Coffee

Male Female

1 cup Male, 1 cup Female, 1 cup

2 cups Male, 2 cups Female, 2 cups

3 cups Male, 3 cups Female, 3 cups

146. (Students should be encouraged to discuss their ideas.) One issue with not assessing for normality is that the data could be positively or negatively skewed. Also, not assessing for normality can threaten the study's validity, reliability, and generalizability. If data from one sample or group is skewed, group comparisons may lead to inaccurate and false results. 147. For the main effect of marital distress, R2 = 0.246/(0.437 – 0.001 – 0.002) = 0.57, which is a large effect. It makes sense that a large effect would be statistically significant, even with a limited sample size. For the main effect of depression, R2 = 0.001/(0.437 – 0.246 – 0.002) = 0.00, which indicates that there is essentially no effect for this variable. For the interaction, R2 = 0.002/(0.437 – 0.246 – 0.001) = 0.01, which is a small effect. If the researcher really thinks something is going on with these variables, he or she might want to collect more data to assess the three effects with greater confidence.

Page 65

Name:

Class:

Date:

Chap_14_5e 148. The independent variable gender has two levels (male and female). The independent variable stress has five levels (no stress, public speaking stress, heat stress, cold stress, and noise stress). The dependent variable is salivary cortisol levels. The researcher should use a mixed-groups (between-within) ANOVA. The within-groups factor is stress condition. The between-groups factor is gender. This test would allow the researcher to examine the impact of both gender and stress condition on salivary cortisol levels. 149. The covariate is the students' grades in science class. The researchers may be interested in the effect of reasoning ability on achievement after taking into account the students' scientific ability. In other words, they are interested in the independent effects that reasoning ability has on achievement test performance. 150. (a) The authors found a significant main effect of medication condition and a significant main effect of electrical brain stimulation, but no significant interaction between medication and brain electrical stimulation. (b) (Figure: Test-Enhanced Learning and Memory) The graph reveals that participants who received sertraline had lower depression scores than participants who were treated with placebo; that participants who were treated with active tDCS had lower depression scores than those individuals who were treated with sham tDCS; and that those treated with both medication and active tDCS had the lowest reported depression scores at the end of the six weeks of treatment. Figure: Sertraline and Electrical Current Therapy

Page 66

Name:

Class:

Date:

Chap_15_5e Indicate whether the statement is true or false. 1. A correlation coefficient higher than 0.50 is extremely unusual in behavioral science research. a. True b. False 2. A partial correlation describes a situation in which the complete correlation calculation has not been performed. a. True b. False 3. When performing a hypothesis test for correlation, the statistician tests the research hypothesis that ρ ≠ 1.0. a. True b. False 4. A correlation coefficient higher than 0.05 is extremely unusual in behavioral science research. a. True b. False 5. Correlation coefficients at or near –1.00 or 1.00 are relatively common. a. True b. False 6. Validity is difficult to assess and is sometimes not considered as a result. a. True b. False 7. When performing a hypothesis test for correlation, the statistician tests the null hypothesis that ρ ≠ 0.0. a. True b. False 8. When performing a hypothesis test for correlation, the statistician tests the null hypothesis that ρ = –1.0. a. True b. False 9. In calculating the correlation coefficient, the denominator corrects for variability and sample size. a. True b. False 10. When performing a hypothesis test for correlation, the statistician tests the null hypothesis that ρ = 1.0. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 1

Name:

Class:

Date:

Chap_15_5e 11. Coefficient alpha can easily be computed by hand. a. True b. False 12. When performing a hypothesis test for correlation, the statistician tests the research hypothesis that ρ ≠ –1.0. a. True b. False 13. If two variables are negatively correlated, then low scores on one variable would be associated with low scores on the other variable. a. True b. False 14. When performing a hypothesis test for correlation, the statistician tests the research hypothesis that r = 0.0. a. True b. False 15. If two variables are negatively correlated, then high scores on one variable would be associated with low scores on the other variable. a. True b. False 16. The strength or size of a correlation coefficient is dependent on its sign. a. True b. False 17. When performing a hypothesis test for correlation, the statistician tests the null hypothesis that ρ = 0.0. a. True b. False 18. The strength or size of a correlation coefficient is independent of its sign. a. True b. False 19. Test–retest reliability is intended to measure a person's (not a test's) consistency over time. a. True b. False 20. If two variables are positively correlated, then high scores on one variable would be associated with low scores on the other variable. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 2

Name:

Class:

Date:

Chap_15_5e 21. The first step in calculating a correlation is to examine the data using a visual display. a. True b. False 22. If two variables are negatively correlated, they cannot be causally related. a. True b. False 23. When performing a hypothesis test for correlation, the statistician tests the research hypothesis that ρ = 1.0. a. True b. False 24. Psychometrics is a field that is saturated with qualified professionals. a. True b. False 25. Correlation coefficients at or near ±1.00 are considered perfect correlations. a. True b. False 26. If two variables are correlated, it means that they are causally related in some way. a. True b. False 27. A correlation coefficient of 0.05 is considered a large effect by Cohen (1988). a. True b. False 28. If two variables are positively correlated, then low scores on one variable would be associated with low scores on the other variable. a. True b. False 29. If two variables are positively correlated, then high scores on one variable would be associated with high scores on the other variable. a. True b. False 30. Correlation coefficients at or near 0.00 are considered perfect correlations. a. True b. False

Page 3

Name:

Class:

Date:

Chap_15_5e 31. When performing a hypothesis test for correlation, the statistician tests the research hypothesis that ρ ≠ 0.0. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 32. According to guidelines published by Cohen (1988), a correlation value of 0.32 would be considered: a. small. b. medium. c. large. d. very large. 33. A researcher discovers that length of time spent following a Mediterranean diet is negatively correlated with risk of developing cancer. Which statement logically follows from this information? a. Eating a Mediterranean diet increases the risk of developing cancer. b. Eating a Mediterranean diet reduces the risk of developing cancer. c. People who ate a Mediterranean diet for longer periods were less likely to have cancer. d. People who develop cancer don't enjoy eating a Mediterranean diet. 34. If a positive relation exists between two variables, then high scores on one variable will be associated with _____ scores on the other variable. a. low b. high c. both high and low d. neither high nor low 35. What is the formula for the Pearson correlation coefficient? a. b. c. d. 36. Why do correlation coefficients greater than 0.50 rarely occur in the social sciences? a. Human behavior is the product of many interacting variables; any single variable will be limited in its association with a behavior. b. Social scientists fail to construct experiments carefully enough to detect larger correlations. c. The highest value that a correlation coefficient can take on is 0.60. d. Social scientists have not yet discovered the variables that are the best predictors of human behavior. Copyright Macmillan Learning. Powered by Cognero.

Page 4

Name:

Class:

Date:

Chap_15_5e 37. If a researcher wishes to assess the relation between race and access to high school literacy, it may be important to take into account the overlapping variability that race and high school literacy share with household income. If a correlation between race and high school literacy is found after controlling for household income, this result indicates that: a. high school literacy is not associated with race or household income. b. race, high school literacy, and household income are perfectly correlated with one another. c. there are racial inequities in high school literacy that cannot be accounted for by differences in household income. d. any differences in high school literacy are attributable to differences in household income. 38. What is the research hypothesis when testing for significance using the Pearson correlation coefficient? a. r ≠ 0 b. r = 0 c. ρ ≠ 0 d. ρ = 0 39. Researchers who study the process of reading in children have discovered that better readers make smoother eye movements across the page. If a scatterplot of the data is drawn, the points on the graph would likely run from _____ left to _____ right. a. upper; lower b. lower; upper c. upper; upper d. lower; lower 40. If a researcher wishes to assess the relation between an agricultural county's average temperature over the growing season and the crop yield, it may be important to take into account the overlapping variability that crop yield and temperature share with rainfall. To do so, the researcher would compute a: a. split-half reliability. b. partial correlation. c. concurrent validity. d. coefficient alpha. 41. The average of all possible split-half correlations is: a. coefficient alpha, a measure of validity. b. coefficient alpha, a measure of reliability. c. the test–retest reliability. d. the test–retest validity.

Page 5

Name:

Class:

Date:

Chap_15_5e The relation between the number of class sessions attended and the grade obtained on the final exam for students enrolled in a statistics course during the fall 2018 semester is shown in the figure. Figure: Attendance and Exam Grade

42. (Figure: Attendance and Exam Grade) Based on the scatterplot, it appears that as the number of classes a student attends increases, the grade on the final exam: a. increases. b. decreases. c. remains the same. d. is unrelated. 43. Researchers who study the process of reading in children have discovered that better readers watch fewer hours of television. If a scatterplot of the data is drawn, the points on the graph would likely run from _____ left to _____ right. a. upper; lower b. lower; upper c. upper; upper d. lower; lower 44. A correlation is computed using data from 27 people. What is (are) the critical cutoff(s) for a two-tailed hypothesis test with a p level of 0.05? a. 0.317 b. 0.374 c. –0.381 and 0.381 d. –0.374 and 0.374

Page 6

Name:

Class:

Date:

Chap_15_5e 45. Psychometricians are concerned with: a. developing high-quality tests and measures. b. fixing psychological issues in people. c. studying illness and the onset of psychological illness. d. statistics and computers. This scatterplot, created from 1995 U.S. News & World Report data on approximately 1300 U.S. colleges and universities, depicts the relation between the student–faculty ratio at the school and the number of admissions applications the school received. Figure: Student–Faculty Ratio

46. (Figure: Student–Faculty Ratio) Based on the scatterplot, what is the relation between an institution's student–faculty ratio and the number of applications it receives? a. perfect positive b. positive c. negative d. no relation 47. In a reanalysis of published studies, Twenge and Im (2007) found that during 1958–2001, the need for social approval of people in the United States was negatively correlated with the U.S. divorce rate during the same period (the correlation coefficient was –0.38). This correlation means that: a. the need for social approval prevented people from seeking divorce. b. the need for social approval spurred people to divorce. c. when the need for social approval was high, divorce rates were low. d. when the need for social approval was high, divorce rates were high.

Page 7

Name:

Class:

Date:

48. (Figure: Attendance and Exam Grade) Based on the scatterplot, what is the relation between a student's class attendance and grade on the final exam? a. perfect positive b. positive c. negative d. no relation 49. What kind of correlation would you expect to find between the amount of snowfall and rates of attendance at college classes? a. positive b. zero c. negative d. perfect negative 50. When conducting a hypothesis test for the Pearson correlation coefficient, the research hypothesis states that there is _______ between the two variables. a. either a positive or a negative correlation b. no correlation c. a positive correlation d. a negative correlation

Page 8

Name:

Class:

Date:

Chap_15_5e 51. The correlation coefficient between class attendance and number of points missed on a statistics exam is – 0.64. Which statement regarding this finding is correct? a. If you start attending class more regularly, the number of points you miss on the next exam is certain to increase. b. There is no causal relationship between the two variables. c. If you attend class regularly, you are more likely to do well on the exam than someone who does not attend class regularly. d. The correlation demonstrates causality. 52. A Pearson correlation coefficient is calculated for 38 individuals. What value of df should be used to determine statistical significance in hypothesis testing? a. 38 b. 37 c. 36 d. 35 53. If a negative relation exists between two variables, then high scores on one variable will be associated with _____ scores on the other variable. a. low b. high c. both high and low d. neither high nor low 54. Before calculating the correlation coefficient, it is advisable to create a _____ as a way of displaying the association between the two variables. a. scatterplot b. line graph c. histogram d. polygon 55. Which of these correlation coefficients allows a perfect prediction of scores on one variable from knowledge of scores on the other variable? a. –1.00 b. 0 c. 0.50 d. 2.00

Page 9

Name:

Class:

Date:

Chap_15_5e 56. One assumption for using hypothesis testing for Pearson correlation is that one variable should vary equally at each level of the other variable. What is the easiest way to determine whether this assumption has been met? a. Review the value of r; if it is high, then the assumption has been met. b. Conduct a post hoc test following the calculation of r. c. Draw a scatterplot to see whether the range of values is equal across all values of the other variable. d. Calculate the cross-products of the deviation scores. If the result is positive, then the assumption has been met. 57. If all the points on a scatterplot fall on a single line: a. there is no relation between the variables. b. the relation between the variables is perfect. c. the variables are causally related. d. the relation between the variables is positive. 58. The numerator (top half) of the Pearson correlation coefficient formula includes the: a. sum of the product of the deviations for each variable. b. square root of the product of the two sums of squares. c. difference between the two sample means. d. correction for variability and sample size. 59. Which of these values of the correlation coefficient indicates the weakest relationship between two variables? a. –0.83 b. –0.51 c. 0.32 d. 0.04 60. Test–retest reliability is determined by: a. administering the same measure to the same sample at two different points in time and calculating the correlation between an individual's performance on the two administrations. b. administering the same measure to two different samples at two different points in time and calculating the correlation between an individual's performance at the two different times. c. correlating the odd-numbered items of a measure with an individual's performance on the evennumbered items of that same measure. d. correlating the odd-numbered items of a measure with an individual's performance on the evennumbered items of a different measure.

Page 10

Name:

Class:

Date:

Chap_15_5e 61. The Pearson correlation coefficient is symbolized by: a. r. b. x. c. c. d. t. 62. If a test actually measures what it is intended to measure, then that test is: a. reliable. b. valid. c. free of restriction of range problems. d. internally consistent. 63. What is the correlation between a person's weight as measured in grams and a person's weight as measured in pounds? a. –1.00 b. 0.00 c. +0.50 d. +1.00 64. The most widely used measure of reliability is: a. the split-half correlation, in which the odd-numbered and even-numbered items of a measure are correlated to assess internal consistency. b. coefficient alpha, the average of all possible split-half correlations. c. test–retest reliability. d. criterion-related reliability. 65. According to guidelines published by Cohen (1988), a correlation value of –0.47 would be considered: a. small. b. medium. c. large. d. very large. 66. Which of these values of the correlation coefficient indicates the weakest relationship between two variables? a. –0.93 b. 0.61 c. –0.69 d. 0.92

Page 11

Name:

Class:

Date:

Chap_15_5e 67. A correlation is computed using data from 18 people. What is (are) the critical cutoff(s) for a two-tailed hypothesis test with a p level of 0.05? a. –0.389 or 0.389 b. –0.456 and 0.456 c. –0.468 and 0.468 d. –0.590 or 0.590 68. When conducting a hypothesis test for the Pearson correlation coefficient, a researcher is testing the null hypothesis that the correlation value is: a. 0. b. 1.00. c. –1.00. d. either 1.00 or –1.00. 69. A technique that quantifies the degree of association between two variables after statistically removing the association of a third variable with both of those two variables is: a. split-half reliability. b. partial correlation. c. concurrent validity. d. coefficient alpha. 70. Which of these represents a perfect correlation? a. –1.00 only b. 1.00 only c. 0 d. –1.00 or 1.00 71. According to guidelines published by Cohen (1988), a correlation value of –0.011 would be considered: a. small. b. medium. c. large. d. very large. 72. What is the relationship between correlation and causation? a. There is no relationship between correlation and causation. b. Correlation is necessary for causation. c. Correlation is sufficient for causation. d. Correlation is both necessary and sufficient for causation.

Page 12

Name:

Class:

Date:

Chap_15_5e 73. Using the Pearson correlation coefficient to analyze the relationship between two variables is appropriate only if the variables are: a. scale variables. b. ordinal variables. c. linearly related. d. linearly related and scale variables. 74. Based on research with her patients, Dr. Colburn knows that the correlation coefficient between scores on an anxiety scale and comfort at a social gathering is –0.35. If the critical value for r is 0.349, what should she conclude? a. The null hypothesis should be retained. b. Scores on the anxiety scale are significantly related to feelings of comfort in a social gathering. c. Scores on the anxiety scale are not significantly related to feelings of comfort in a social gathering. d. Scores on the anxiety scale are causally related to feelings of comfort in a social gathering. 75. A correlation ranges in value between _____ and _____. a. 0; 1.00 b. –1.00; 1.00 c. –1.00; 0 d. negative infinity; positive infinity 76. Which of these values indicates the strongest relationship between two variables? a. 0.49 b. 0.35 c. –0.23 d. –0.47 77. A positive correlation between the number of classes a student attends and grade on the statistics final exam indicates that: a. smarter students enjoy learning and so come to more classes than less successful students do. b. teachers are more lenient graders for those students who attend class regularly. c. the more classes a student attends, the greater the likelihood that the student will do well on the final exam. d. attending class regularly results in students understanding the material better and doing better on the final exam.

Page 13

Name:

Class:

Date:

Chap_15_5e 78. A positive correlation between head size and foot size indicates that: a. having a large head causes a person to have large feet. b. having large feet causes a person to have a large head. c. people with larger feet also tend to have larger heads. d. people with larger feet tend to have smaller heads. 79. What is the null hypothesis when testing for significance using the Pearson correlation coefficient? a. µ1 = µ2 b. r = 0 c. r ≠ 0 d. ρ = 0 80. What kind of correlation would you expect to find between levels of family income and household spending on consumer goods? a. positive b. zero c. negative d. perfect positive 81. The denominator of the Pearson correlation equation corrects for _____ and _____ issues present in the numerator. a. sample size; variability b. sample size; non-normality c. negative values; variability d. negative values; nonlinear data 82. Which of the following statements is an assumption that should be met before hypothesis testing with the correlation coefficient? a. The two variables are significantly related to each other. b. At least one of the variables is ordinal or nominal. c. The variability in one variable is not the same as the variability in the other variable. d. Each variable varies equally, regardless of the magnitude of the other variable. 83. According to guidelines published by Cohen (1988), a correlation value of _____ is considered small, _____ is medium, and _____ is strong. a. 0.10; 0.30; 0.50 b. 0.25; 0.50; 0.75 c. 0.40; 0.60; 0.80 d. 1.00; 2.00; 3.00

Page 14

Name:

Class:

Date:

Chap_15_5e 84. According to guidelines published by Cohen (1988), a correlation value of –0.53 would be considered: a. small. b. medium. c. large. d. very large. 85. When a positive relation exists between two variables, what will be true in the calculation of the Pearson correlation coefficient? a. The product of most pairs of deviations will be positive. b. Most deviation scores will be positive. c. Most deviation scores will be negative. d. The product of most pairs of deviations will be negative. 86. In a reanalysis of published studies, Twenge and Im (2007) found that during 1958–2001, the need for social approval of people in the United States was negatively correlated with the U.S. violent crime rate during the same period (the correlation coefficient was –0.31). This correlation means that: a. the need for social approval prevented people from committing violent crimes. b. the need for social approval spurred people to commit violent crimes. c. as the need for social approval went up, the number of violent crimes also increased. d. as the need for social approval went up, the number of violent crimes decreased. 87. If a positive relation exists between two variables, then low scores on one variable will be associated with _____ scores on the other variable. a. low b. high c. both high and low d. neither high nor low 88. When conducting a hypothesis test for the Pearson correlation coefficient, the null hypothesis states that there is _______ between the two variables. a. either a positive or a negative correlation b. no correlation c. a positive correlation d. a negative correlation 89. Which of these values indicates the strongest relationship between two variables? a. 0.62 b. 0.53 c. –0.27 d. –0.71 Copyright Macmillan Learning. Powered by Cognero.

Page 15

Name:

Class:

Date:

Chap_15_5e 90. A _____ is a graphical representation of the relation between two variables. a. correlation coefficient b. scatterplot c. histogram d. polygon 91. A perfect linear relationship will yield a Pearson correlation value of: a. 1.00 only. b. –1.00 only. c. 0. d. 1.00 or –1.00. 92. The results of a study find a positive correlation between gum chewing and life expectancy. Which statement would be a statistically appropriate response to the results of the study? a. You purchase a lifetime supply of gum because chewing gum is good for your health. b. You bemoan the possibility of living so long that you will have to chew lots of gum. c. You become curious about what third variables might cause both increases in gum chewing and increases in life expectancy. d. You tell all your friends and family members to chew gum because it is good for their health. 93. What kind of correlation would you expect to find between the age of a car (nonclassic) and its value? a. positive b. zero c. negative d. perfect positive

Page 16

Name:

Class:

Date:

Chap_15_5e Nietfeld and Ender (2003) performed a study investigating the relation between the intelligence of student teachers and their efficacy in the classroom. This relation is depicted in the scatterplot, which shows fictional data that replicate the pattern of performance observed by the researchers. Figure: Teaching Efficacy

94. (Figure: Teaching Efficacy) Based on the scatterplot, what kind of relation between intelligence and teaching efficacy did Nietfeld and Ender find? a. perfect positive b. positive c. negative d. no relation 95. "The incidence of depression among college students is negatively correlated with the number of sunny days each year in the locale where they attend college." This statement means that: a. the chances of a college student being depressed tend to increase with fewer sunny days. b. a student attending college in a locale with very few sunny days will become depressed. c. the chances of a college student being depressed tend to increase as the number of sunny days increases. d. a depressed student who moves to a locale with more sunny days will become less depressed. 96. What kind of correlation would you expect to find between the results of two dice thrown simultaneously? a. positive b. zero c. negative d. perfect positive

Page 17

Name:

Class:

Date:

Chap_15_5e 97. Researchers who study the process of reading in children have discovered that better readers make smoother eye movements across the page (i.e., there is a positive correlation between reading skill and the smoothness of the eye movements). Which statement is a possible cause of the correlation between smooth eye movements and reading ability? a. The ability to make smooth eye movements improves reading ability. b. Having good reading skills allows a child to make smooth eye movements. c. Some third factor might cause both smooth eye movements and improved reading ability. d. All of these options are possible causes of the correlation. 98. In a reanalysis of published studies, Twenge and Im (2007) found that during 1958–2001, the need for social approval of people in the United States was positively correlated with changes in the Dow Jones Industrial Average during that same period (the correlation coefficient was 0.10). This means that: a. the need for social approval caused people to invest more money in the stock market. b. the need for social approval prevented people from investing money in the stock market. c. when the need for social approval was high, the Dow Jones Industrial Average was also high. d. when the need for social approval was high, the Dow Jones Industrial Average was low. 99. The Pearson correlation coefficient is a statistic that measures: a. the causal association between scale variables. b. an association among scale, ordinal, and nominal variables. c. a linear relation between two scale variables. d. relatedness in terms of variability between variables. 100. The first step involved in calculating the Pearson correlation coefficient is to: a. multiply each raw score in one variable by the corresponding raw score in the second variable. b. square all the scores. c. calculate the deviation of each score from its mean. d. calculate the standard deviation for each variable. 101. If a negative relation exists between two variables, then low scores on one variable will be associated with _____ scores on the other variable. a. low b. high c. both high and low d. neither high nor low

Page 18

Name:

Class:

Date:

Chap_15_5e 102. When conducting a hypothesis test for the Pearson correlation coefficient, degrees of freedom is calculated as: a. Σ(X – MX)2. b. (N1 – 1)(N2 – 1). c. N – 1. d. N – 2. 103. Based on research with her patients, Dr. Colburn knows that the correlation coefficient between scores on an anxiety scale and comfort at a social gathering is –0.47. According to guidelines established by Cohen, how would the strength of this relationship be characterized? a. small b. medium c. large d. very large 104. The denominator (bottom half) of the Pearson correlation coefficient includes the: a. sum of the product of the deviations for each variable. b. square root of the product of the two sum of squares. c. difference between the sample means. d. square root of the deviations for each variable. 105. In an investigation of the psychometric properties of the Criminal Sentiments Scale (CSS), Witte et al. (2006) administered the scale to 72 sex offenders. They found that scores on the scale correlated positively with recidivism for nonsexual crimes, but not for sexual crimes. This research suggests that the CSS is: a. a valid measure of sentiment to commit sexual crimes, but not a valid measure of sentiment to commit nonsexual crimes. b. a valid measure of sentiment to commit nonsexual crimes, but not a valid measure of sentiment to commit sexual crimes. c. a valid measure of sentiment to commit both nonsexual crimes and sexual crimes. d. not a reliable measure. 106. Which of the following is NOT an assumption when conducting hypothesis testing for r? a. The population distributions should be normal. b. To ensure generalizability, the data should be randomly selected. c. The variability of scores on one variable should be equal across all levels of the second variable. d. The variables being studied are positively correlated.

Page 19

Name:

Class:

Date:

Chap_15_5e Enter the appropriate word(s) to complete the statement. 107. The distribution against which the Pearson correlation is compared is the _______ distribution.

108. _______ is used to calculate reliability either through test–retest reliability or through a measure of internal consistency, such as coefficient alpha.

109. The denominator of the Pearson correlation formula corrects for sample size and _______.

110. The _______ of a correlation is indicated by how close to "perfect" its data points are.

111. Correlation is used to calculate _______, often by associating a new measure with an existing measure known to assess the variable of interest.

112. The first step in assessing data for a correlation is to create a(n) _______ as a visual display.

Page 20

Name:

Class:

Date:

Chap_15_5e 113. If two variables are negatively correlated, people who have high scores on one variable will tend to have _______ scores on the other variable.

114. According to Cohen's (1988) guidelines, an r of 0.12 would be considered a(n) _______ correlation.

115. Another name for coefficient alpha is _______.

116. If two variables are _______ correlated, people who have low scores on one variable will tend to have low scores on the other variable.

117. The branch of statistics used in the development of tests and measures is _______.

118. A correlation value of _______ would indicate that there was no association between the two variables.

119. Coefficient alpha is a commonly used estimate of a test's or measure's _______.

Page 21

Name:

Class:

Date:

Chap_15_5e 120. When a positive correlation is observed, the sum of the products of deviations will be _______.

121. If two variables are positively correlated, people who have high scores on one variable will tend to have _______ scores on the other variable.

122. If two variables are _______ correlated, people who have high scores on one variable will tend to have high scores on the other variable.

123. According to Cohen's (1988) guidelines, an r of –0.55 would be considered a(n) _______ correlation.

124. If two variables are _______ correlated, people who have high scores on one variable will tend to have low scores on the other variable.

125. Degrees of freedom are calculated as _______ for the Pearson correlation coefficient.

Page 22

Name:

Class:

Date:

Chap_15_5e 126. Correlation coefficients of _____ and _____ indicate that there is a perfect association between the two variables.

127. The technique of controlling for a third variable when evaluating the strength of association between two variables of interest is called _______.

128. According to Cohen's (1988) guidelines, an r of –0.28 would be considered a(n) _______ correlation.

129. Calculating a correlation coefficient is appropriate only when there is a(n) _______ relation between two variables.

130. The direction of a correlation is indicated by its _______.

131. Correlation is not _______.

Page 23

Name:

Class:

Date:

Chap_15_5e 132. (Table: Devils Games Played and Penalty Minutes) The data in the table represent the number of games played and the penalty minutes accrued by each of 18 New Jersey Devils hockey players during the 2012– 2013 season. The mean number of penalty minutes was 41.78 with a standard deviation of 28.25. The mean number of games played was 74.39 with a standard deviation of 9.095. (a) Calculate the Pearson correlation coefficient between the number of games played and penalty minutes. (b) Assess the significance of the correlation coefficient assuming a two-tailed hypothesis test with a p level of 0.05. Table: Devils Games Players and Penalty Minutes Penalty Minutes 114 36 20 69 25 42 14 38 36 92 61 16 64 34 30 18 8 35

Games Played 48 62 66 69 71 72 74 75 75 76 76 80 82 82 82 82 82 85

Page 24

Name:

Class:

Date:

Chap_15_5e 133. (Figure: Percent Metropolitan and Crime Rate) The figure depicts, for each of the 50 states and Washington, D.C., the relation between the percentage of the state's population living in a metropolitan area and the violent crime rate for the state. (a) Based on the scatterplot, what is the relation between the percentage of a state's population living in a metropolitan area and the violent crime rate? (b) Identify and describe the outlier in the scatterplot. (c) Describe how an outlier analysis might be useful in this case and what kind of information it might yield. Figure: Percent Metropolitan and Crime Rate

Page 25

Name:

Class:

Date:

Chap_15_5e 134. (Table: Agreeableness) Psychologists studied a sample of 104 adolescents in a midwestern town. The researchers assessed a personality variable called "agreeableness" using a standard personality questionnaire. The questionnaire is a measure of how nice a person is to be around. The questions ask about how cheerful, stubborn, polite, bossy, and cooperative the person is. Each adolescent's average of those items is reported. The researchers also created a measure of behavior problems. The adolescents self-reported on various behaviors in the last 6 months, including cheating, swearing, stealing, and fighting. Each adolescent's sum of behavior problems is reported. The table shows a representative sample of 10 participants. For this sample of 10 participants, (a) draw the scatterplot, (b) compute the Pearson correlation coefficient, and (c) assess the significance of the correlation coefficient assuming a two-tailed hypothesis test with a p level of 0.05. Table: Agreeableness Participant 2 14 23 31 44 52 63 78 90 100

Agreeableness Rating 4.3 3 3.4 3.1 2.9 4 4.7 1.8 2.9 4.7

Behavior Problems 5 22 10 12 28 21 9 31 14 6

135. Research coming out of the Developmental Labs, Inc., of Huntington, New York, established that reading skill was positively correlated with smooth movements of the eyes during reading (i.e., fewer pauses and fewer retracks). A friend suggests that the local school district invest money in equipment that teachers could use to help students better move their eyes so that reading skill can be improved. How would you respond to this friend's recommendation?

Page 26

Name:

Class:

Date:

Chap_15_5e 136. The correlation between age and penalty minutes for all 28 New Jersey Devils hockey players is 0. But the partial correlation between age and penalty minutes, when controlling for the number of games played, is – 0.241. Why do these two correlations differ? What relation could exist among the three variables that gives rise to these differences?

Page 27

Name:

Class:

Date:

Chap_15_5e 137. (Figure: Student–Faculty Ratio) The following scatterplot, created from 1995 U.S. News & World Report data on approximately 1300 U.S. colleges and universities, depicts the relation between the student–faculty ratio at the school and the number of admissions applications that the school received. (a) Based on the data depicted in the scatterplot, what is the relation between a school's student–faculty ratio and the number of applications the school receives? (b) (Table: Student–Faculty Ratio) A subset of this large data set appears in the table. For this set of six schools, calculate the Pearson correlation coefficient. (c) Is the relation for these six schools (as determined by the Pearson correlation coefficient) similar to the relation observed in the larger data set (as determined by the scatterplot)? Is this sample of six schools representative of the larger data set from which it was drawn? Figure: Student–Faculty Ratio

Table: Student–Faculty Ratio School Alaska Pacific University University of Arkansas at Pine Bluff Arizona State University, main campus California State University at Long Beach Pomona College Connecticut College

Applications 193 1450 12,809 10,278 3037 3035

Student–Faculty Ratio 11.9 16.2 18.9 20.3 8.9 10.7

Page 28

Name:

Class:

Date:

Chap_15_5e 138. A systematic review of the literature by Monzani and colleagues (2019) documented a positive correlation between skipping breakfast and obesity among children and adolescents, such that those who skip breakfast tend to be overweight or obese. State three different hypotheses about the causal relationship creating this correlation, one for each possible causal relationship in the ABC model: A → B, B → A, or C → A and C → B.

Page 29

Name:

Class:

Date:

Chap_15_5e 139. (Figure: TV and Life Expectancy) The figure depicts the relation between the number of people per television set and life expectancy in 40 countries as reported in the 1993 World Almanac and Book of Facts. (a) What is the relation between the number of people per television set and life expectancy as depicted in the scatterplot? (b) (Table: TV and Life Expectancy) A subset of the data appears in the table. For this subset of data, calculate the Pearson correlation coefficient. Figure: TV and Life Expectancy

Table: TV and Life Expectancy Country Bangladesh Kenya North Korea United States Italy

Life Expectancy 53.5 61.0 70.0 75.5 78.5

People/TV 315 96 90 1 4

Page 30

Name:

Class:

Date:

Chap_15_5e 140. (Table: Scatterplot A) Draw a scatterplot and compute the Pearson correlation coefficient for the data in the table. Table: Scatterplot A X 3 2 1 4 6

Y 4 5 9 7 2

141. For the New Jersey Devils hockey players who played more than 40 games in the 2012–2013 season, there is a negative correlation between the number of games played and the number of minutes spent in the penalty box. Players who played in more games spent less time in the penalty box. State three different hypotheses about the causal relationship creating this correlation, one for each possible causal relationship in the ABC model: A → B, B → A, or C → A and C → B.

142. The Dynamic Indicator of Basic Early Literacy Skills (DIBELS) is a measure developed to assess the literature skill of young children just entering school (kindergarten and first grade). (a) What could be done to determine whether DIBELS is a reliable measure? (b) What could be done to examine the validity of DIBELS?

Page 31

Name:

Class:

Date:

Chap_15_5e 143. Data from the 2008 General Social Survey–National Death Index data set reveal a negative correlation between the number of hours spent watching television and life expectancy in the American public: Life expectancy is lower among individuals who report watching more television. State three different hypotheses about the causal relationship creating this correlation, one for each possible causal relationship in the ABC model: A → B, B → A, or C → A and C → B.

144. (Table: Scatterplot B) Draw a scatterplot and compute the Pearson correlation coefficient for the data in the table. Table: Scatterplot B X 4 2 3 6 5

Y 3 2 5 9 6

145. Why might a financial advisor recommend investing in stocks whose performances are negatively correlated?

Page 32

Name:

Class:

Date:

Chap_15_5e Answer Key 1. True 2. False 3. False 4. False 5. False 6. True 7. False 8. False 9. True 10. False 11. False 12. False 13. False 14. False 15. True 16. False 17. True 18. True 19. False 20. False 21. True 22. False 23. False 24. False 25. True 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 33

Name:

Class:

Date:

Chap_15_5e 27. False 28. True 29. True 30. False 31. True 32. b 33. c 34. b 35. a 36. a 37. c 38. c 39. b 40. b 41. b 42. a 43. a 44. c 45. a 46. d 47. c 48. b 49. c 50. a 51. c 52. c 53. a 54. a Copyright Macmillan Learning. Powered by Cognero.

Page 34

Name:

Class:

Date:

Chap_15_5e 55. a 56. c 57. b 58. a 59. d 60. a 61. a 62. b 63. d 64. b 65. c 66. b 67. c 68. a 69. b 70. d 71. a 72. b 73. d 74. b 75. b 76. a 77. c 78. c 79. d 80. a 81. a 82. d Copyright Macmillan Learning. Powered by Cognero.

Page 35

Name:

Class:

Date:

Chap_15_5e 83. a 84. c 85. a 86. d 87. a 88. b 89. d 90. b 91. d 92. c 93. c 94. b 95. a 96. b 97. d 98. c 99. c 100. c 101. b 102. d 103. c 104. b 105. b 106. d 107. r 108. Correlation 109. variability 110. strength Copyright Macmillan Learning. Powered by Cognero.

Page 36

Name:

Class:

Date:

Chap_15_5e 111. validity 112. scatterplot 113. low 114. small 115. Cronbach's alpha 116. positively 117. psychometrics 118. 0.00, zero 119. reliability 120. positive 121. high 122. positively 123. large 124. negatively 125. N - 2 126. 1.00; -1.00 127. partial correlation 128. medium 129. linear 130. sign 131. causation 132. (a) r = –0.52. (b) The critical cutoffs based on 16 degrees of freedom are –0.468 and 0.468. The correlation coefficient exceeds that cutoff, so we would be able to reject the null hypothesis and conclude there is a statistically significant correlation between these variables. 133. (a) The scatterplot depicts a positive relation between the percentage of a state's population living in a metro area and the violent crime rate for that state. (b) One state stands out. For this state, 100% of the population lives in a metro area and the crime rate is nearly triple that of every other state. (c) It would be useful to know what state the outlier was. For example, the outlier might be Washington, D.C., which is not really a state. Also, it would be useful to look at other changes taking place in the outlier state during the year the data were collected to gain insight into why the crime rate was so high. Copyright Macmillan Learning. Powered by Cognero.

Page 37

Name:

Class:

Date:

Chap_15_5e 134. (a) (Figure: Agreeableness) The figure contains a scatterplot of the data. Figure: Agreeableness

(b) r = –0.77. (c) The critical cutoffs based on 8 degrees of freedom are –0.632 and 0.632. The correlation coefficient exceeds that cutoff, so we would be able to reject the null hypothesis and conclude there is a statistically significant correlation between these variables. 135. Students' answers will vary. This answer is just one example. I would tell the friend that given just this correlational information, we do not know the causal mechanism underlying the relation. Perhaps being able to move your eyes smoothly improves reading ability, but the causal relation could also be in the opposite direction: Children might be skilled readers because they do not need to make lots of stops and retracing of sentences when reading. We should perform an experiment to determine the causal direction before investing lots of money in training programs nationwide. 136. Age may be positively correlated with the number of games played: More experienced players probably get more opportunities to play than rookies do. Because of that, the older players have, overall, more time on the ice and a greater opportunity to incur penalties. Thus, once we take into account the number of games in which the player played (thus controlling for the opportunities for penalties), we are able to see the negative correlation between age and penalty minutes. 137. (a) Given the scatterplot, there appears to be no relation between a school's student–faculty ratio and the number of applications it receives. (b) r = 0.75. (c) Although there is no relation in the larger data set, there is a large correlation in the small sample of six schools. This small sample is not representative of the larger set of schools from which it was drawn. 138. Students' answers will vary. This answer is just one example. It is possible that skipping breakfast actually causes people to gain weight. Perhaps missing nutrients and calories early in the day leads to overeating later in the day, or perhaps eating breakfast results in less snacking so that people who eat breakfast are less likely to gain weight. Alternatively, perhaps being overweight or obese results in individuals who are not hungry when they first get up and, therefore, are more likely to skip breakfast. Finally, some third variable, such as exercise and lifestyle choices, could cause both skipping breakfast and weight gain. 139. (a) The scatterplot depicts a negative relation between the number of people per television set and life expectancy. As the number of people per television set increases, life expectancy decreases. (b) r = –0.92. Copyright Macmillan Learning. Powered by Cognero.

Page 38

Name:

Class:

Date:

Chap_15_5e 140. (Figure: Scatterplot A) r = –0.74 Figure: Scatterplot A

141. Students' answers will vary. This answer is just one example. Perhaps playing more games gives players more experience so that they incur fewer penalties. Or perhaps players who incur lots of penalties are prevented from playing in future games during the season by their coaches. Conversely, a third variable, such as experience, might cause players to both play more often and incur fewer penalties. 142. (a) To determine whether DIBELS is a reliable measure, we could administer it to a sample of kindergarteners and first graders and then calculate the coefficient alpha for the test. The coefficient alpha would be calculated as the average of all possible split-half correlations of the test. (b) As discussed in the text, a lot can be learned from a thorough review of the items on a test. Are the items asking about the issue you intended to examine and assess? Not only should we consider the items within a given measure, but we should also ask ourselves if we are operationalizing our variables in a way that is valid. The text uses the example of researchers operationalizing success using several different perspectives. 143. Students' answers will vary. This answer is just one example. It is possible, given this correlation, that watching TV results in an earlier death as a consequence of a more sedentary lifestyle. Alternatively, perhaps individuals with shorter lifespans are motivated to watch more television because they know that they won't be around for long. Finally, another possible explanation is that the poor health and/or numerous medical problems are causally related to both the amount of time a person spends in front of a television set and life expectancy.

Page 39

Name:

Class:

Date:

Chap_15_5e 144. (Figure: Scatterplot B) Figure: Scatterplot B

r = 0.87 145. When performance of stocks is negatively correlated, if one of the stocks is performing really poorly, the other stock is likely to be performing really well. It can be helpful to have money invested in different stocks that respond differently to the market so that the stocks' gains or losses even each other out. This way total investment is steady rather than being a roller coaster of ups and downs.

Page 40

Name:

Class:

Date:

Chap_16_5e Indicate whether the statement is true or false. 1. Unlike the calculation of effect size for ANOVA, r2 is considered an unbiased estimate of effect size for simple linear regression. a. True b. False 2. Correlation coefficient and proportionate reduction in error are inversely related; that is, as the correlation coefficient increases, proportion reduction in error decreases. a. True b. False 3. The slope is the amount that X is predicted to increase for an increase of one unit in Y. a. True b. False 4. Regression is typically used when analyzing the results of an experiment. a. True b. False 5. It is impossible for the regression line to do a poorer job than the mean of predicting the dependent variable. a. True b. False 6. Like correlation, regression cannot prove causal direction of a relation between two variables. a. True b. False 7. The intercept is the predicted value for X when Y is equal to 0. a. True b. False 8. The simple linear regression equation uses the following formula: Ŷ = a + b(X). a. True b. False 9. The predicted z score on the independent variable will always be closer to its mean than the z score for the dependent variable is to its mean. a. True b. False

Page 1

Name:

Class:

Date:

Chap_16_5e 10. The intercept is the predicted value for Y when X is equal to 1. a. True b. False 11. The intercept is the amount that Y is predicted to increase for an increase of one unit in X. a. True b. False 12. The slope is the predicted value for Y when X is equal to 0. a. True b. False 13. The intercept is the predicted value for Y when X is equal to 0. a. True b. False 14. The predicted z score on the dependent variable will always be closer to its mean than the z score for the independent variable is to its mean. a. True b. False 15. The slope is the amount that Y is predicted to increase for an increase of one unit in X. a. True b. False 16. The simple linear regression equation uses the following formula: Ŷ = a(X) + b. a. True b. False 17. The standard error of estimate is a measure of how accurately we predict using the regression equation or line of best fit. a. True b. False 18. Similar to the calculation of effect size for ANOVA, r2 is considered a biased estimate of effect size for simple linear regression. a. True b. False

Page 2

Name:

Class:

Date:

Chap_16_5e 19. The predicted z score on the dependent variable will always be closer to the z score for the independent variable than it is to its mean. a. True b. False 20. Correlation involves relating variables, whereas regression involves prediction. a. True b. False 21. Simple linear regression is a statistical technique that includes two or more predictor variables in a prediction equation. a. True b. False 22. Regression capitalizes on correlation by using what is known about the relation between two variables to make predictions beyond those variables. a. True b. False 23. Unlike correlation, regression can prove causal direction of a relation between two variables. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 24. Data from 147 colleges from 1995 to 2005 (Lee, 2008) were used to predict the endowments (in billions of dollars) for a college from the average SAT score of students attending the college. The resulting regression equation was Ŷ = –20.46 + 4.06(X). This regression indicates that: a. most colleges have very high endowments. b. for every one-point increase in SAT scores, a college can expect $4.06 billion more in endowments. c. for every one-dollar increase in endowments, the college can expect a half-point increase in SAT scores. d. for every one-point increase in SAT scores, a college can expect $20.46 billion fewer in endowments.

Page 3

Name:

Class:

Date:

Chap_16_5e The scatterplot and regression line on the left depict the relation between a state's expenditure per student and the average SAT scores for students in the state. The scatterplot and regression line on the right depict the relation between students' SAT Verbal and SAT Quantitative scores. Figure: Standard Error Comparisons 25. (Figure: Standard Error Comparisons) For which prediction is the standard error of the estimate greater? a. state expenditure per student from composite SAT scores b. composite SAT scores from the state's expenditure per student c. SAT Quantitative scores from SAT Verbal scores d. SAT Verbal scores from SAT Quantitative scores 26. As the correlation coefficient becomes stronger, proportionate reduction in error: a. becomes smaller. b. becomes larger. c. is unaffected. d. becomes more variable. 27. Which is the correct formula for calculating the adjusted r2? a. b. c. d. 28. As r2 increases, the standard error of the estimate: a. increases. b. decreases. c. stays the same. d. gets closer to 1. 29. Proportionate reduction in error can be symbolized by: a. r2. b. d. c. r. d. b. Copyright Macmillan Learning. Powered by Cognero.

Page 4

Name:

Class:

Date:

Chap_16_5e 30. Simple linear regression allows the statistician to: a. determine the relation among four or more variables. b. predict an individual's score on a dependent variable from her score on multiple independent variables. c. predict an individual's score on the dependent variable from her score on the independent variable. d. infer the direction of causal relations. 31. Which is the correct formula for the proportionate reduction in error? a. (b) b. c. a + bX d. 32. Which statistic quantifies the improvement in ability to predict a person's score when using the regression line rather than the mean? a. standard error of the estimation b. standard deviation c. slope d. proportionate reduction in error 33. In the formula Ŷ = a + b(X), a is the: a. slope. b. intercept. c. predicted value for the dependent variable. d. observed value on the independent variable. 34. The standardized regression coefficient expresses the: a. relation between the independent and dependent variable in terms of squared units. b. strength of the correlation between the two variables that are now incorporated into a regression analysis. c. predicted change in the dependent variable in terms of standard deviations for an increase of 1 standard deviation in the independent variable. d. likelihood of rejecting the null hypothesis with a regression analysis.

Page 5

Name:

Class:

Date:

Chap_16_5e 35. The correlation between the number of ounces of candy a child consumes weekly and the number of cavities the child has at the age 13 years is 0.74. If a child has a z score of –0.82 on the variable "candy consumed," the z score on the variable "number of cavities" would be predicted to be: a. –0.61. b. –0.78. c. 0.61. d. 0.78. 36. The predicted z score for the dependent variable will always be _____ the individual's z score for the independent variable. a. more than b. the same as c. two times d. less than 37. The tendency of scores that are particularly high or low to drift toward the mean over time is called: a. simple linear regression. b. standard error of the mean. c. regression to the mean. d. standard error of the estimate. 38. A statistics professor finds that he can predict a student's grade on the final exam based on the number of classes the student attended over the course of the semester. What can the professor say about the causal relationship between these two variables? a. The relationship likely is the result of some third variable that accounts for both a student's attendance in class and performance on the exam. b. Students who do well on exams come to class because they enjoy the material. c. Attendance in classes causes students to understand the material and do well on the exams. d. We cannot say anything definitive about the causal relationships, as all three of these explanations are possible. 39. As the standard error of estimate becomes larger, predictions become: a. less accurate. b. more accurate. c. larger. d. smaller.

Page 6

Name:

Class:

Date:

Chap_16_5e 40. A researcher wants to be able to predict first-semester grade point average with as much accuracy as possible, so she would like to use both high school grade point average and SAT score as predictor variables. Which technique would be most appropriate to make this prediction? a. simple linear regression b. proportionate reduction in error c. multiple regression d. standardized regression coefficient 41. A simple way to calculate proportionate reduction in error is by: a. taking the square root of the correlation coefficient. b. squaring SStotal. c. squaring the correlation coefficient. d. adding the slope and the y intercept. 42. In the equation for a regression line, the intercept is the: a. value for X when Y is equal to 0. b. predicted value for Y when X is equal to 0. c. amount that Y is predicted to increase for a one-unit increase in X. d. z score of the amount that Y is predicted to increase as X increases. 43. In the formula Ŷ = a + b(X), b is the: a. slope. b. intercept. c. predicted value for the dependent variable. d. observed value on the independent variable. 44. The standardized regression coefficient expresses a predicted change in the dependent variable in terms of: a. standard deviation units. b. slope. c. a one-unit change in the independent variable. d. error units. 45. Which statement does NOT indicate a limitation of using regression? a. The presence of confounding variables may limit confidence in the findings. b. The data used are rarely from a true experiment. c. Regression is appropriate only with linear data. d. Regression allows you to determine an equation for a straight line.

Page 7

Name:

Class:

Date:

Chap_16_5e 46. If two variables, independently, can help predict the outcome of a third variable, they are: a. autonomous. b. orthogonal. c. standardized. d. proportionate. 47. The standard error of estimate could be thought of as the: a. standard deviation of the data points around the regression line. b. standard deviation of the independent variable on the regression line. c. standard deviation of the dependent variable. d. amount of error made in random selection. 48. For the equation Ŷ = 125 + 4.2(X1) + 3.2(X2), which statement is true? a. The slope of the line is 125. b. This is a simple linear regression equation. c. There are two slopes. d. The y intercept is 7.5. 49. If the correlation between the number of beers consumed over a semester and the grade point average for the semester is –0.53, what would be the predicted z score for a person who has a z score of –0.92 for the number of beers consumed? a. 0.57 b. –0.57 c. 0.49 d. –0.49 50. In the computer printouts presented in the text, information for the multiple regression equation can be found in the column labeled: a. standardized coefficients. b. B under unstandardized coefficients. c. beta under standardized coefficients. d. sig. 51. What is the formula for predicting an individual's z score on the dependent variable from the z score on the independent variable using the correlation coefficient? a. b. c. d. Copyright Macmillan Learning. Powered by Cognero.

Page 8

Name:

Class:

Date:

Chap_16_5e 52. (Table: Coefficients) What is the y intercept for this problem? a. 1.134 b. 23.054 c. 30.331 d. 0.262 53. The proportionate reduction in error is a measure of the: a. amount of variance in the dependent variable explained by the independent variable. b. correlation between two variables. c. amount of variance in the independent variable explained by the dependent variable. d. slope of a regression line. The table includes information for creating a regression equation to predict students' attitudes toward statistics from their attitudes toward Ed Sheeran and beer. Table: Coefficentsa Unstandardized Coefficients

Standardized Coefficients Beta

Model B Std. Error t Sig. Constant 23.054 30.331 1.134 0.262 attSheeran 0.309 2.277 0.016 0.136 0.892 attBeer 0.468 0.102 0.548 4.588 0.000 a Dependent variable: Attitude toward statistics. 54. (Table: Coefficients) What is the slope for predicting attitude toward statistics from attitude toward beer? a. 0.468 b. 0.309 c. 0.548 d. 4.588 55. Regression is to _____ as correlation is to _____. a. association; causation b. causation; association c. relation; prediction d. prediction; relation

Page 9

Name:

Class:

Date:

Chap_16_5e 56. In the equation for a regression line, the slope is the: a. value for X when Y is equal to 0. b. predicted value for Y when X is equal to 0. c. amount that Y is predicted to increase for a one-unit increase in X. d. z score of the amount that Y is predicted to increase as X increases. 57. A man and a woman who are both tall (he is 6 feet tall and she is 5 feet 10 inches tall) have four children. Which child represents regression to the mean for height? a. Janet, who is 5 feet tall b. Amy, who is 5 feet 1 inch tall c. Laura, who is 5 feet 5 inches tall d. John, who is 6 feet 2 inches tall 58. Data from the World Health Organization in 2013 were used to predict the life expectancy for men in a country from the life expectancy of women in the same country. The resulting regression equation was Ŷ = 3.73 + 0.88(X). This regression equation implies that: a. when a woman's life expectancy increases by 1 year, a man's life expectancy increases by 3.73 years. b. women live 0.88 times as long as men do. c. when a woman's life expectancy increases by 1 year, a man's life expectancy increases by 0.88 of a year. d. the average life expectancy for men in some countries is 3.73. 59. In a study designed to predict blood cholesterol levels from amount of daily saturated fat consumed (in grams; X1) and number of hours of daily exercise (X2), the slope of X1 is 4, the slope of X2 is –3, and the y intercept is 140. Which formula is the regression equation for these data? a. Ŷ = 140 + 4(X1) – 3(X2) b. Ŷ = 140 – 3(X1) – 4(X2) c. Ŷ = 140 + 4(X1) + 3(X2) d. Ŷ = 140 + 1(X) 60. Every year it seems as though last season's baseball rookie of the year fails to live up to expectations for his sophomore season. What might explain this phenomenon? a. regression to the mean b. overestimation of effect size c. standard error of the estimation d. proportionate reduction in error

Page 10

Name:

Class:

Date:

Chap_16_5e 61. Data from the World Health Organization in 2013 were used to predict the life expectancy for men in a country from the life expectancy of women in the same country. The resulting regression equation was Ŷ = 3.73 + 0.88(X). Using the regression equation, what would be the predicted life expectancy of men in a country in which the life expectancy for women is 80 years? a. 66.67 years b. 67.12 years c. 70.40 years d. 74.13 years 62. The regression line is the line that: a. minimizes error in predicting scores on the dependent variable. b. is the mean of the dependent variable. c. minimizes error in predicting scores on the independent variable. d. minimizes the correlation coefficient. 63. In a study designed to predict blood cholesterol levels from amount of daily saturated fat consumed (in grams; X1) and number of hours of daily exercise (X2), the slope of X1 is 4, the slope of X2 is –3, and the y intercept is 140. If someone reports that she typically eats 15 grams of saturated fat daily and exercises 1 hour daily, what would be her predicted cholesterol level? a. 57 b. 99 c. 165 d. 197 64. An independent variable that makes a unique contribution to the prediction of a dependent variable is a(n) _____ variable. a. orthogonal b. latent c. manifest d. unique

Page 11

Name:

Class:

Date:

Chap_16_5e The scatterplot and regression line on the left depict the relation between a state's expenditure per student and the average SAT scores for students in the state. The scatterplot and regression line on the right depict the relation between students' SAT Verbal and SAT Quantitative scores. Figure: Standard Error Comparisons 65. (Figure: Standard Error Comparisons) Based on the scatterplots, for which prediction is r2 greater? a. state expenditure per student from composite SAT scores b. composite SAT scores from the state's expenditure per student c. SAT Quantitative scores from SAT Verbal scores d. SAT Verbal scores from SAT Quantitative scores 66. In a 2008 article by Hsiu-Ling Lee, data from 147 colleges from 1995 to 2005 were used to predict endowments for a college from the average SAT score of students attending the college, among other variables. The resulting regression equation for just these variables was Ŷ = –20.46 + 4.06(X). Using the regression equation, what would be predicted as the endowments (in billions of dollars) for a college whose students' average SAT score is 1080? a. 3.86 b. 4364.34 c. 4405.26 d. 22,092.74 67. If the standard error of the estimate is zero, the relation between two variables is: a. curvilinear. b. imperfect. c. perfect. d. unknown. 68. Which statistical tool allows you to predict a dependent score based on information about an independent variable? a. t test b. correlation c. regression d. standardization

Page 12

Name:

Class:

Date:

Chap_16_5e 69. For a simple linear regression, the standardized regression coefficient is: a. the square of the r statistic. b. equal to the Pearson correlation coefficient. c. the square root of the slope. d. unrelated to the correlation value. 70. The standard error of the estimate indicates: a. how far two regression lines are from each other. b. how far, on average, the regression line is from the mean. c. how much error there is in any single prediction made from a given regression equation. d. the typical distance between the regression line and each of the observed data points. 71. Which statement about the slope of the simple linear regression line is true? a. It has the same sign as the correlation. b. It is expressed as a standardized value. c. The square of the slope is the proportion of variation in Y explained by X. d. It is the value of Y when X is zero. 72. To predict a single dependent variable from more than one independent variable, which statistical technique would you use? a. multiple regression b. structural equation modeling c. simple linear regression d. correlation 73. Y is the symbol for a(n) _____, and Ŷ is the symbol for a(n) _____. a. predicted score on the independent variable; observed score on the independent variable b. observed score on the dependent variable; predicted score on the dependent variable c. observed score on the independent variable; predicted score on the dependent variable d. predicted score on the dependent variable; observed score on the independent variable 74. In the equation Ŷ = 125 + 4.2(X1) + 3.3(X2), what is the y intercept? a. 125 b. 7.5 c. 4.2 d. 3.3

Page 13

Name:

Class:

Date:

Chap_16_5e 75. A high standard error of the estimate indicates that the: a. mean is not a good representation of the sample data. b. observed Ys will vary greatly from the predicted Ys. c. sample mean is not a good representation of the population mean. d. observed Ys will cluster closely around the regression line. 76. What information do the slopes in a multiple regression equation provide about the correlation coefficient? a. The slopes tell us nothing about the correlation coefficient. b. The sign of the slope (positive or negative) tells us the direction of the correlation. c. The slope sign is inversely related to the direction of the correlation. d. The magnitude of the slope tells us how strong the correlation coefficient is. The table includes information for creating a regression equation to predict students' attitudes toward statistics from their attitudes toward Ed Sheeran and beer. Table: Coefficentsa Unstandardized Coefficients

Standardized Coefficients Beta

Model B Std. Error t Sig. Constant 23.054 30.331 1.134 0.262 attSheeran 0.309 2.277 0.016 0.136 0.892 attBeer 0.468 0.102 0.548 4.588 0.000 a Dependent variable: Attitude toward statistics. 77. (Table: Coefficients) What is the slope for predicting attitude toward statistics from attitude toward Ed Sheeran? a. 0.016 b. 0.309 c. 0.892 d. 2.277 78. In the equation Ŷ = 78 + 4.10(X1) + 7.60(X2), what is (are) the slope(s)? a. 78 b. 78 and 4.10 c. 4.10 d. 4.10 and 7.60

Page 14

Name:

Class:

Date:

Chap_16_5e 79. Regression cannot prove causation, but it can: a. provide specific quantitative predictions that help explain relations among variables. b. provide stronger evidence for association than does correlation. c. predict people's behaviors on variables that may seem impossible to measure. d. serve as a substitute for good experimental design. 80. Under what circumstance, unlikely as it might be, would the standard error of estimate be zero? a. The proportionate reduction in error is zero. b. The correlation coefficient is zero. c. The correlation coefficient is either 1.00 or –1.00. d. The proportionate reduction in error is 1. The table includes information for creating a regression equation to predict students' attitudes toward statistics from their attitudes toward Ed Sheeran and beer. Table: Coefficentsa Unstandardized Coefficients

Standardized Coefficients Beta

Model B Std. Error t Constant 23.054 30.331 1.134 attSheeran 0.309 2.277 0.016 0.136 attBeer 0.468 0.102 0.548 4.588 a Dependent variable: Attitude toward statistics. 81. (Table: Coefficients) Is either variable a significant predictor for attitude toward statistics? a. No, neither is a significant predictor. b. Yes, both are significant predictors. c. Attitude toward Ed Sheeran is a significant predictor but attitude toward beer is not. d. Attitude toward beer is a significant predictor but attitude toward Ed Sheeran is not.

Sig. 0.262 0.892 0.000

Enter the appropriate word(s) to complete the statement. 82. The measure of effect size for linear regression r2 is considered to be a(n) _______ estimator.

Page 15

Name:

Class:

Date:

Chap_16_5e 83. The _______ of the regression line tells how much the dependent variable is predicted to change for every one-unit change in the independent variable.

84. If a student gets 58% on his first statistics exam, regression to the mean would predict that he will get a(n) _______ score on his second statistics exam.

85. If a student gets 93% on her first statistics exam, regression to the mean would predict that she will get a(n) _______ score on her second statistics exam.

86. The proportionate reduction in error is also called the _______.

87. The line of best fit is also known as the _______.

88. _______ variables are independent variables that make separate but distinct contributions in the prediction of a dependent variable, as compared with the contributions of other variables.

Page 16

Name:

Class:

Date:

Chap_16_5e 89. The line of _______ is another name for the regression line.

90. _______ regression enables you to predict an individual's score on the dependent variable from his or her score on a single independent variable.

91. The point at which the regression line crosses the y-axis is referred to as the _______.

92. The point at which the regression line crosses the _______ is referred to as the intercept.

93. The _______ is also called the coefficient of determination.

94. The tendency for scores that are particularly high or low to drift toward the mean over time is called _______.

Page 17

Name:

Class:

Date:

Chap_16_5e 95. The _____ is the total amount of error if the mean of the dependent variable were used to predict every score on the dependent variable.

96. _______ regression enables prediction of an individual's score on the dependent variable from two or more independent variables.

97. The _______, symbolized by β, is the predicted change in the dependent variable in terms of standard deviations for a 1 standard deviation increase in the independent variable.

98. _____ represents the error that would occur if Y was predicted using the regression equation.

Page 18

Name:

Class:

Date:

Chap_16_5e 99. (Table: Twin Data Coefficients) The following SPSS analysis was conducted using a set of data collected by a team of five interviewers at the 16th Annual Twins Day Festival in Twinsburg, Ohio, in August 1991. A booth was set up at the festival's main entrance, and an ad inviting all adult twins to participate in the survey was placed in the festival program. In addition, the interviewers roamed the festival grounds, approaching all adult twins for an interview, and almost every pair of twins accepted. (a) What is the likely relation between the number of years of education completed by the individuals and their twin siblings? (b) The SPSS output for a linear regression analysis appears in the table. What is the independent variable in this analysis? (c) What is the dependent variable? (d) Use the table to generate the regression equation. Table: Twin Data Coefficientsa

a Dependent variable: Education Twin 2

Page 19

Name:

Class:

Date:

Chap_16_5e 100. (Table: Graduation Rate and Expenditure per Student) The table contains a subset of data published by the American Institutes for Research in 2013. The data specify the graduation rate of five different U.S. colleges and the instructional expenditure per student at each of the colleges. Use the data to answer the questions. (a) Generate the regression equation for predicting graduation rates from the expenditure per student. (b) Using the regression equation, predict the graduation rate for a school that spends $6800 per student. (c) Use the regression equation to determine how much a school would need to spend to obtain a 100% graduation rate. Table: Graduation Rate and Expenditure per Student College Alaska Pacific University Arizona State University, main campus California State University at Long Beach Pomona College Connecticut College

Expenditure/Student Graduation Rate 10,540 46.2 8836 57.5 6527 54.0 28,418 95.3 17,734 83.5

101. A study published in the European Journal of Public Health (Elmen et al., 1996) reported that birth weight (standardized for gestational age and sex) was negatively correlated with current adult mortality rates in the population (r = –0.58). (a) Given this correlation, if a newborn's z score on the birth weight measure is 0.66, what will be the z score for adult mortality? (b) What is r2 for predicting adult mortality rates from birth weight? (c) What, if any, is the causal relation suggested by this correlation and r2? Speculate on the nature of the causal relation underlying the correlation.

Page 20

Name:

Class:

Date:

Chap_16_5e 102. (Table: Twin Data Correlations) The following analysis was conducted using a set of data collected by a team of five interviewers at the 16th Annual Twins Day Festival in Twinsburg, Ohio, in August 1991. A booth was set up at the festival's main entrance, and an ad inviting all adult twins to participate in the survey was placed in the festival program. In addition, the interviewers roamed the festival grounds, approaching all adult twins for an interview, and almost every pair of twins accepted. Use the table exploring the correlation between the number of years of education completed by Twin 1 and Twin 1's annual income to answer the questions. (a) What is the correlation between years of education completed and income? (b) What is the income z score of a person who has an education z score of 0.63? (c) Draw a graph with the regression line for predicting z scores on income from z scores on education. Be sure to actually calculate values, using what you know about correlation, z scores, and regression to place points on your graph. Table: Twin Data Correlations

Page 21

Name:

Class:

Date:

Chap_16_5e 103. (Table: Residents Living in Poverty Coefficients) A multiple regression analysis was used to predict the percentage of a state's residents living in poverty from the percentage of the residents who graduated from high school and the percentage of the residents who are single parents. The SPSS output is provided in the table. (a) Write the regression equation for predicting percent poverty from percent high school graduates and percent single parents. (b) If for a given state 64% of its residents graduated from high school and 29% are single parents, what would you predict as the percentage of residents living in poverty? Table: Residents Living in Poverty Coefficientsa

a Dependent variable: Percent poverty

104. Why is the regression line sometimes called the line of best fit?

105. Your roommate is convinced that it is possible to determine a person's age based on the number of movies the person sees annually. To test this idea, your roommate collects data from a sample of people for whom the average age is 44 and the average number of movies watched per week is 2.5. The correlation between age and number of movies viewed per week is r = 0. Using this correlation, write the regression equation for predicting a person's age from the number of movies the person sees in a week.

Page 22

Name:

Class:

Date:

Chap_16_5e 106. Analysis of data for 193 countries reported by Bulled and Sosis (2010) revealed that the correlation between the adult literacy rate in a country and the life expectancy in the country was r = 0.70. (a) What is the adjusted r2 for predicting life expectancy from the literacy rate? (b) Explain what this adjusted r2 means for this set of data. (c) What, if any, is the causal relation suggested by this correlation and adjusted r2? Speculate on the nature of the causal relation underlying the correlation.

107. (Table: Students' Final Exam Grades) A multiple regression analysis was used to predict the grades students received on their final statistics exam based on their attendance over the course of the semester and their grade on the first midterm. The SPSS output is provided in the table. (a) Write the regression equation for predicting final test grades from percent exam 1 grades and percent classes attended. (b) If a given student earned a 74% on the first exam and attended 81% of the class sessions, what would you predict as that student's final exam grade? Table: Students' Final Exam Gradesa Unstandardized Coefficients

Standardized Coefficients t

Model B Std. Error (Constant) –15.316 10.630 Percent exam 1 0.725 0.134 Percent classes 0.413 0.069 attended a Dependent variable: Percent final exam grade

Beta

Sig.

0.477 0.527

–1.441 5.418 5.994

0.156 0.000 0.000

Page 23

Name:

Class:

Date:

Chap_16_5e 108. (Table: Physicians per 10,000 Population and Life Expectancy) The table contains a subset of data reported in 2016 by the World Health Organization. Use these data to generate the regression equation for predicting life expectancy in a country from the number of physicians per 10,000 people in that country. To help you get started, the mean and the standard deviation for both variables have been calculated. Table: Physicians per 10,000 Population and Life Expectancy Country Bangladesh Cambodia United States Egypt France Japan Mean SD

Physicians/10,000 People 4.8 1.7 25.9 8.1 32.3 24.1 16.15 12.81

Life Expectancy 72.7 69.4 78.5 70.5 82.9 84.2

76.37 6.40

109. Data collected by a psychology professor over a four-year period examined the relationship between class attendance and average test grades in an Introductory Psychology course. Results revealed that the number of classes attended was positively correlated with the students' average test grade over the course of the semester (r = 0.35). (a) What is the proportionate reduction in error, or r2 , for predicting average test grade from the number of classes attended? (b) Explain what this r2 means for this set of data.

110. You believe that you can determine a person's IQ from the number of grammatical errors the individual makes in a job application. (a) What type of analysis would you perform to test your idea? (b) What would be the independent variable in this analysis? (c) What would be the dependent variable in this analysis?

Page 24

Name:

Class:

Date:

Chap_16_5e 111. Your roommate is convinced that it is possible to determine a person's age based on the number of movies the person sees annually. (a) What type of analysis would your roommate perform to test this idea? (b) What would be the independent variable in this analysis? (c) What would be the dependent variable in this analysis?

Page 25

Name:

Class:

Date:

Chap_16_5e Answer Key 1. False 2. False 3. False 4. False 5. True 6. True 7. False 8. True 9. False 10. False 11. False 12. False 13. True 14. True 15. True 16. False 17. True 18. True 19. False 20. True 21. False 22. False 23. False 24. b 25. b 26. b Copyright Macmillan Learning. Powered by Cognero.

Page 26

Name:

Class:

Date:

Chap_16_5e 27. b 28. b 29. a 30. c 31. d 32. d 33. b 34. c 35. a 36. d 37. c 38. d 39. a 40. c 41. c 42. b 43. a 44. a 45. d 46. b 47. a 48. c 49. c 50. b 51. c 52. b 53. a 54. a Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_16_5e 55. d 56. c 57. c 58. c 59. a 60. a 61. d 62. a 63. d 64. a 65. c 66. b 67. c 68. c 69. b 70. d 71. a 72. a 73. b 74. a 75. b 76. b 77. b 78. d 79. a 80. c 81. d 82. biased Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_16_5e 83. slope 84. higher 85. lower 86. coefficient of determination 87. regression line 88. Orthogonal 89. best fit 90. Simple linear 91. intercept, y intercept 92. y-axis or y-axis 93. proportionate reduction in error 94. regression to the mean 95. 96. Multiple 97. standardized regression coefficient 98. 99. (a) The relation is likely to be positive. (b) education of Twin 1 (c) education of Twin 2 (d) Ŷ = 5.29 + 0.62(X) 100. (a) Ŷ = 35.907 + 0.002(X) (b) 49.51 (c) $32,046.50 101. (a) –0.38 (b) 0.34 (c) No specific causal mechanism can be implied by a correlation or regression. It is likely that both increases in birth weight and decreases in adult mortality are associated with an overall healthy population, which could be caused by any number of things, such as a stable economy, no wars, or ease of access to health care. 102. (a) r = 0.26 (b) 0.16 (c) Students' graphs may differ slightly in the specific points used to generate the line. To graph the line, they will need to compute at least one other pair of z scores [beyond that computed for part (b)]. 103. (a) Ŷ = 46.29 – 0.57 (HS grad) + 0.85 (single parent) (b) 34.46% of residents would be predicted to live in poverty within that state. 104. The regression line is the line that minimizes the amount of error between the prediction line and the actual points used to create the prediction line. What this means is that of any lines we may choose to draw through the data in an attempt to predict the dependent variable from the independent variable, the regression line is the one that will produce the least amount of prediction error. Thus, it is the line that best fits the data. Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_16_5e 105. Ŷ = 44 + 0(X) or Ŷ = 44 106. (a) r2 is 0.49. (b) This means that 49% of the variability in life expectancy is accounted for by the literacy rate in the country. (c) The correlation does not imply any particular causal relation. It is likely that a third variable, such as wealth of the nation, causes both a longer life expectancy and a higher literacy rate. 107. (a) Ŷ = –15.32 + 0.73 (exam 1) + 0.41 (classes) (b) 71.91% would be the predicted final exam grade for this student. 108. The correct answer is 68.934 +0.460(X). However, if students round everything to three decimal places during their calculations the answer will be slightly different: Ŷ = 68.940 + 0.454(X) 109. (a) r2 would be 0.12. (b) This means that 12% of the variability in average test grades is accounted for by the number of classes a student attends over the course of the semester. It tells us how much better our regression predictions were compared to using the mean as a predictor. 110. (a) simple linear regression (b) number of grammatical errors (c) the person's IQ 111. (a) simple linear regression (b) number of movies seen annually (c) the person's age

Page 30

Name:

Class:

Date:

Chap_17_5e Indicate whether the statement is true or false. 1. Adjusted standardized residuals are measures of effect size for chi-square. a. True b. False 2. When working with two nominal variables, you would use a chi-square test for independence. a. True b. False 3. When working with two nominal variables, each with two levels, a chi-square test for independence would be the appropriate test to use. a. True b. False 4. When graphing the results of a chi-square analysis, you would plot frequencies rather than percentages. a. True b. False 5. The degrees of freedom for a chi-square test of independence are calculated as N – 2. a. True b. False 6. When the assumptions of a parametric test are met, researchers are in greater danger of making a Type II error with a nonparametric test than with a parametric test. a. True b. False 7. When working with one nominal variable, a chi-square test for independence would be the appropriate test to use. a. True b. False 8. The suggested minimum for expected frequencies in each cell of the chi square is 10. a. True b. False 9. When working with one nominal variable with four categories, you would use a chi-square test for independence. a. True b. False

Page 1

Name:

Class:

Date:

Chap_17_5e 10. The degrees of freedom for a chi-square test of independence are calculated as k – 2. a. True b. False 11. Delucchi (1983) suggested there should be at least five times as many participants as cells in a chi-square study. a. True b. False 12. The null hypothesis for the chi-square test for independence is that the two variables have a dependent relationship. a. True b. False 13. Relative risk or likelihood is easy to interpret and understand, and does not require base-rate information. a. True b. False 14. When the assumptions of a parametric test are met, nonparametric tests have less statistical power than parametric tests do. a. True b. False 15. When the assumptions of a parametric test are met, researchers are in greater danger of making a Type I error with a nonparametric test than with a parametric test. a. True b. False 16. When graphing the results of a chi-square analysis, you would plot percentages rather than frequencies. a. True b. False 17. When working with one nominal variable with three categories, a chi-square test for goodness of fit would be the appropriate test to use. a. True b. False 18. When the assumptions of a parametric test are met, nonparametric tests have more statistical power than parametric tests do. a. True b. False

Page 2

Name:

Class:

Date:

Chap_17_5e 19. When working with one nominal variable, a chi-square test for goodness of fit would be the appropriate test to use. a. True b. False 20. When working with two nominal variables, a chi-square test for goodness of fit would be the appropriate test to use. a. True b. False 21. Even with scale data, a statistician may choose to run a nonparametric test because there is a large sample size and the data are normally distributed. a. True b. False 22. The suggested minimum for expected frequencies in each cell of the chi square is 5. a. True b. False 23. Delucchi (1983) suggested there should be at least 10 times as many participants as cells in a chi-square study. a. True b. False 24. Even with scale data, a statistician may choose to run a nonparametric test because there is a small sample size and/or the data are skewed. a. True b. False 25. The effect-size measure for chi square is called Cramér's phi. a. True b. False 26. The research hypothesis for the chi-square test for independence is that the two variables have an independent relationship. a. True b. False 27. When working with two nominal variables, each with two levels, a chi-square test for goodness of fit would be the appropriate test to use. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 3

Name:

Class:

Date:

Chap_17_5e 28. The research hypothesis for the chi-square test for independence is that the two variables have a dependent relationship. a. True b. False 29. The null hypothesis for the chi-square test for independence is that the two variables have an independent relationship. a. True b. False 30. When deciding on the critical value for the chi-square test, you must first decide whether to use a one-tailed or two-tailed hypothesis test. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 31. Sit and colleagues (2018) investigated the efficacy of adjunctive bright-light therapy in patients with bipolar disorder. Patients were randomly assigned to receive either bright-light therapy or dim-light therapy along with their regular medication. After six weeks, patients' clinical status was coded as improved or not improved. The researchers found that 15 of 23 patients treated with bright light were improved after six weeks and 4 of 23 patients treated with dim light were improved after six weeks. They also found that the relative likelihood of improvement with bright light was 3.75. This means that: a. bipolar patients treated with bright light were 3.75 times more likely to show clinical improvement than patients treated with dim light. b. bipolar patients treated with dim light were 3.75 times more likely to show clinical improvement than patients treated with bright light c. for every bipolar patient who improves with bright light, only 0.27 bipolar patients treated with dim light will show clinical improvement. d. for every bipolar patient who improves with dim light, only 0.27 bipolar patients treated with bright light will show clinical improvement. 32. If an adjusted standardized residual is larger than _____, the observed frequency in that cell differs significantly from the expected frequency. a. 0.05 b. 0.5 c. 1.0 d. 2.0

Page 4

Name:

Class:

Date:

Chap_17_5e 33. The adjusted standardized residual tells: a. whether there is an interaction between two nominal variables. b. the expected value of a cell when the null hypothesis is false. c. the expected value of a cell when the null hypothesis is true. d. the number of standard errors that an observed frequency is from its expected frequency. 34. The values in the cells of a chi-square test for goodness of fit are: a. proportions. b. medians. c. means. d. counts or frequencies. 35. For a study, a chi-square statistic of 3.42 was calculated. There were 67 participants in total, and 2 degrees of freedom for the first variable and 1 degree of freedom for the second variable. What is the effect size? a. 0.226 b. 0.160 c. 0.051 d. 0.015 36. For a study, a chi-square statistic of 2.82 was calculated. There were 40 participants in total, and 2 degrees of freedom for the first variable and 3 degrees of freedom for the second variable. What is the effect size? a. 0.970 b. 0.266 c. 0.188 d. 0.153 37. The chi-square test to use for two nominal variables is called the chi-square test for independence because we are assessing: a. an assumption of the chi square that the two variables are independent of each other. b. whether the dependent variable is related to the independent variable. c. whether the effect of each variable is independent of the other variable. d. whether the distribution in the sample is independent of the distribution in the population. 38. When there is one nominal variable, the appropriate statistical test to use is a chi-square test for: a. goodness of fit. b. independence. c. empowerment. d. association.

Page 5

Name:

Class:

Date:

Chap_17_5e 39. The adjusted standardized residual is calculated as the difference between the observed and expected frequencies in a given cell: a. divided by the degrees of freedom for the row. b. divided by the standard error. c. minus the total sample size. d. minus the total for the row and minus the total for the column. 40. In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summation of a case involving a battered woman who had shot and killed her husband. Of the 80 participants, 27 assigned a verdict of guilty, 40 a verdict of not guilty by reason of self-defense, and 13 a verdict of not guilty by reason of insanity. What type of chi-square test should the researchers use to determine whether the distribution of verdicts differed from what could have been expected by chance? a. association b. dissociation c. goodness of fit d. independence 41. Corsi et al. (2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program. In their sample of 491 drug users, 38.1% were single, 27.5% were married or had a partner, and 34.4% were divorced, separated, or widowed. When considering the drug users who actually entered a treatment program, 67 were single, 66 were married or had a partner, and 61 were divorced, widowed, or separated. What kind of chi-square test should Corsi et al. use to analyze the data? a. goodness of fit b. independence c. association d. dissociation 42. Using Cohen's conventions for effect size, how would you label a Cramér's V = 0.29, if the row df = 3 and the column df = 3? a. small b. medium c. large d. too small to be of interest

Page 6

Name:

Class:

Date:

Chap_17_5e 43. Sit and colleagues (2018) investigated the efficacy of adjunctive bright-light therapy in patients with bipolar disorder. Patients were randomly assigned to receive either bright-light therapy or dim-light therapy along with their regular medication. After six weeks, patients' clinical status was coded as improved or not improved. The researchers found that 15 of 23 patients treated with bright light were improved after six weeks and 4 of 23 patients treated with dim light were improved after six weeks. What is the relative likelihood of entering a treatment program if a bipolar patient is treated with bright light compared to dim light? a. 0.48 b. 0.65 c. 1.35 d. 3.75 44. For two nominal variables, the appropriate statistical test to use is a chi-square test for: a. goodness of fit. b. independence. c. orthogonality. d. dissociation. 45. When graphing the results of a chi-square test it is more useful to graph _____ rather than _____. a. frequencies; conditional proportions b. proportions or percentages; frequencies c. means; frequencies d. frequencies; means 46. In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband. Of the 80 participants, 27 assigned a verdict of guilty, 49 a verdict of not guilty by reason of self-defense, and 4 a verdict of not guilty by reason of insanity. What are the degrees of freedom associated with this chi square? a. 2 b. 3 c. 78 d. 79 47. Post hoc tests are carried out for chi square by: a. calculating correlations between the observed and expected frequencies in each cell. b. performing all possible two-group chi-square tests. c. calculating the adjusted standardized residual for each cell. d. using the Bonferroni correction for the p level.

Page 7

Name:

Class:

Date:

Chap_17_5e 48. A chi-square test for goodness of fit for which there are 5 categories and 40 participants would have _____ degrees of freedom. a. 195 b. 39 c. 7 d. 4 49. Which statement does NOT represent an assumption of the chi-square test? a. Participants are randomly selected from the population. b. The dependent variable is nominal. c. Each observation is independent of all other observations. d. The dependent variable is scale. 50. With a p level of 0.05, a chi-square test for goodness of fit for which there are 4 categories would have _____ degrees of freedom, and the critical value for the chi square would be _____. a. 2; 5.992 b. 3; 7.815 c. 4; 7.780 d. 4; 9.488 51. A researcher decides to examine whether people prefer milk chocolate or dark chocolate. A chi-square test for goodness of fit is conducted. What is the critical cutoff for a test with a p level of 0.05? a. 3.841 b. 5.992 c. 6.252 d. 7.815 52. The minimum typically suggested for expected frequencies in each cell of the chi square is: a. 2. b. 5. c. 10. d. 15. 53. For a study, a chi-square statistic of 1.45 was calculated. There were 50 participants in total, and 2 degrees of freedom for the first variable and 1 degree of freedom for the second variable. What is the effect size? a. 0.725 b. 0.170 c. 0.120 d. 0.029

Page 8

Name:

Class:

Date:

Chap_17_5e 54. What is the formula for calculating expected frequencies in the chi-square test for independence? a. b. c. d. 55. A chi-square test for independence is calculated on two variables, each with 2 degrees of freedom. An effect size of 0.24 is calculated. What size effect is this? a. small b. medium c. medium to large d. large 56. For a study, a chi-square statistic of 3.75 was calculated. There were 55 participants in total, and 3 degrees of freedom for the first variable and 2 degrees of freedom for the second variable. What is the effect size? a. 0.625 b. 0.185 c. 0.151 d. 0.068 57. A chi-square test for goodness of fit is an appropriate statistical test to use when the study has: a. one nominal dependent variable. b. one scale independent variable. c. two nominal dependent variables. d. two nominal independent variables. 58. A chi-square test for goodness of fit is conducted on a nominal variable with three levels. What is the critical cutoff for a test with a p level of 0.05? a. 4.605 b. 5.992 c. 6.252 d. 7.815

Page 9

Name:

Class:

Date:

Chap_17_5e 59. Legal researchers were interested in whether a rape defendant's perceived guilt depends on the type of argument introduced by his defense attorney. Researchers classified each defense by whether it introduced the idea that the female victim was asking for the encounter—the "she shouldn't have worn that short skirt" defense. In a simulated juror situation, the researchers counted the number of guilty verdicts "returned" by the 20 participating mock jurors. This number is a _____ dependent variable, which means that the data can be analyzed using a _____ test. a. nominal; nonparametric b. scale; nonparametric c. nominal; parametric d. scale; parametric 60. In which situation would it be inappropriate to use a nonparametric test? a. The sample size is small and the population distribution might be skewed. b. The dependent variable is scale. c. The dependent variable is nominal. d. The dependent variable is ordinal. 61. Chi-square tests for independence assess whether: a. two scale variables are associated. b. the means of two samples differ significantly from each other. c. two nominal variables are independent of each other. d. the distributions of two variables are normally distributed. 62. Corsi et al. (2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program. One of the factors they explored was marital status. The researchers found that 66 of 135 married (or partnered) drug users entered a treatment program and 67 of 185 single drug users entered a treatment program. They also found that the relative likelihood of entering a treatment program if a drug user is married is 1.35. This means that: a. married drug users are 1.35 times more likely to enter a treatment program than are single drug users. b. single drug users are 1.35 times more likely to enter a treatment program than are married drug users. c. for every married drug user who enters a treatment program, only 0.35 single drug users enter a treatment program. d. for every single drug user who enters a treatment program, only 0.35 married drug users enter a treatment program.

Page 10

Name:

Class:

Date:

Chap_17_5e 63. In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband. In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical). The dependent variable was the number of guilty verdicts. What is the critical cutoff for this test, given a p level of 0.05? a. 3.841 b. 5.992 c. 7.815 d. 9.488 64. The statistic for reporting the effect size of a chi square is: a. Cramér's V or ϕ. b. Cohen's d. c. R2. d. proportionate reduction in error. 65. A chi-square test for independence is an appropriate statistical test to use when the study has: a. one nominal dependent variable. b. one scale independent variable. c. two nominal dependent variables. d. two nominal independent variables. 66. When we calculate relative risk, we are essentially determining the ratio of two: a. observed frequencies. b. conditional proportions. c. relative likelihoods. d. expected frequencies. 67. A college instructor is interested in whether assigning mandatory prelecture quizzes affects students' grades. Since she is currently teaching two sections of the same class, the instructor assigns quizzes to one of the classes and just recommends the quiz problems to the second class. Then she determines the number of people in each class who will receive an A, B, C, D, or F on the final exam. What type of statistical test should the instructor use to analyze her data? a. chi-square test for goodness of fit b. chi-square test for independence c. analysis of variance d. independent-samples t test

Page 11

Name:

Class:

Date:

Chap_17_5e 68. When using a nonparametric test (in contrast to a parametric hypothesis test), the: a. range of variables available for statistical analysis is reduced. b. sample size should be very large. c. testing power is increased. d. risk of committing a Type II error is increased. 69. Degrees of freedom for the chi-square test of goodness of fit are determined by: a. subtracting 2 from the number of cells (k – 2). b. subtracting 1 from the number of cells (k – 1). c. multiplying the number of rows minus 1, times the number of columns minus 1. d. subtracting 1 from the number of participants (N – 1). 70. Sit and colleagues (2018) investigated the efficacy of adjunctive bright-light therapy in patients with bipolar disorder. Patients were randomly assigned to receive either bright-light therapy or dim-light therapy along with their regular medication. After six weeks, patients' clinical status was coded as improved or not improved. What type of chi-square test should the researchers use to analyze these data? a. association b. dissociation c. goodness of fit d. independence 71. The expected frequencies across all cells of the chi square should: a. always add up to the size of the sample. b. always add up to the size of the population. c. be five times the number of participants. d. be three times the number of participants.

Page 12

Name:

Class:

Date:

Chap_17_5e 72. A college instructor is interested in whether assigning mandatory prelecture quizzes affects students' grades. Since she is currently teaching two sections of the same class, the instructor assigns quizzes to one of the classes and just recommends the quiz problems to the second class. Then she determines the number of people in each class who will receive an A, B, C, D, or F on the final exam. What would be the appropriate statistical advice to give to the instructor? a. It is unnecessary for her to analyze her data because you know from experience that mandatory homework improves your own grades in college classes. b. Unfortunately, she will not be able to analyze the data because the dependent variable is not a scale variable. If she instead used percentages or points on the final exam as the dependent variable, she would be able to analyze the data. c. It is unnecessary for her to analyze the data because you know from experience that mandatory homework does not improve your own grades in college classes. d. She would have more power to detect an effect of homework if she used the students' percentage grades (or point grades) and performed a parametric statistical test. 73. The same formula is used to calculate the chi-square statistic in the chi-square test for goodness of fit and the chi-square test of independence. Which calculation differs along the way for these two tests? a. observed frequencies b. expected frequencies c. difference between observed and expected frequencies d. sample size 74. What is the correct formula for the chi-square statistic? a.

b. c. d.

Page 13

Name:

Class:

Date:

Chap_17_5e 75. A researcher wonders whether supplemental bright-light therapy can be useful in treating bipolar depression, and conducts a study in which one group of bipolar patients receives bright-light therapy and another group of patients receives dim-light therapy. After six weeks, the researcher administers a depression inventory and records the number of patients who have recovered. This number is a _____ dependent variable, which means that the data can be analyzed using a _____ test. a. nominal; nonparametric b. scale; nonparametric c. nominal; parametric d. scale; parametric 76. When graphing the results of a chi-square study, it is important for the y-axis to depict: a. only the range of values observed in the data. b. values of the dependent variable. c. the scale of proportions from 0 to 1.0. d. values of the independent variable. 77. The appropriate statistic for analyzing data in a situation in which both the independent and dependent variables are nominal is a(n): a. nominal; nonparametric b. scale; nonparametric c. nominal; parametric d. scale; parametric 78. One drawback of many nonparametric tests is that: a. there is no way to calculate effect sizes. b. the APA does not allow data analyzed with nonparametric tests to be published in APA journals. c. it is very difficult to meet the assumptions necessary to use the tests. d. such tests have more power than their parametric equivalent. 79. In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays. One question in which the researchers were interested was whether a person's "outness" depended on his or her class in school (freshman, sophomore, junior, or senior). All participants were classified as closeted or as out. What is the critical cutoff for this test, given a p level of 0.05? a. 3.841 b. 7.815 c. 14.067 d. 15.507

Page 14

Name:

Class:

Date:

Chap_17_5e 80. Chi-square tests assess whether: a. two interval variables are associated. b. the means of two samples differ significantly from each other. c. the observed data differ from the pattern expected given the null hypothesis. d. the distributions of two variables are normally distributed. 81. Corsi et al. (2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program. One of the factors explored was marital status. The researchers found that 66 of 135 married (or partnered) drug users entered a treatment program and 67 of 185 single drug users entered a treatment program. What is the relative likelihood of entering a treatment program if a drug user is married compared to if he or she is single? a. 0.36 b. 0.50 c. 1.35 d. 1.55 82. The table constructed for a chi-square test of independence is a _____ table. a. contingency b. source c. frequency d. periodic 83. One danger of reporting relative risk or relative likelihood information rather than frequencies is that: a. people may not understand what the relative risk information means. b. the base rate occurrence of an event is not reported, and the data can be misleading. c. people may be frightened on hearing the relative risk for a positive event. d. the proportion of events is not reported. 84. In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays. One question in which the researchers were interested was whether a person's "outness" depended on his or her gender. All participants were classified as closeted or as out. What type of chi-square test should the researchers use to answer the question of whether outness varies with gender? a. association b. dissociation c. goodness of fit d. independence

Page 15

Name:

Class:

Date:

Chap_17_5e 85. In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays. One question in which the researchers were interested was whether a person's "outness" depended on his or her class in school (freshman, sophomore, junior, or senior). All participants were classified as closeted or as out. What are the degrees of freedom associated with the chi-square analysis of these data? a. 8 b. 7 c. 3 d. 1 86. Another name for relative risk is: a. observed frequency. b. chi-square statistic. c. relative likelihood. d. Cramér's V. 87. In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband. In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical). The dependent variable was the number of guilty verdicts. What are the degrees of freedom associated with the chi-square analysis of these data? a. 4 b. 3 c. 2 d. 1 88. Legal researchers were interested in whether a rape defendant's perceived guilt depends on the type of argument introduced by his defense attorney. Researchers classified each defense by whether it introduced the idea that the female victim was asking for the encounter—the "she shouldn't have worn that short skirt" defense. In a simulated juror situation, the researchers asked each of the 20 participating mock jurors to rate the guilt of the defendant on a scale from 1 to 10. This score is a _____ dependent variable, which means that the data can be analyzed using a _____ test. a. nominal; nonparametric b. scale; nonparametric c. nominal; parametric d. scale; parametric

Page 16

Name:

Class:

Date:

Chap_17_5e 89. For a chi-square test, the minimum number of participants typically recommended is: a. 5. b. 10. c. three times the number of cells. d. five times the number of cells. 90. Degrees of freedom for the chi-square test of independence are determined by: a. subtracting 2 from the number of cells (k – 2). b. subtracting 1 from the number of cells (k – 1). c. multiplying the number of rows minus 1 times the number of columns minus 1. d. subtracting 1 from the number of participants (N – 1). 91. Which of these correctly states the results of a chi-square test reported in APA format? a. χ2(N = 1012) = 11.85, df = 1, p < 0.05 b. χ2(1, N = 1012) = 11.85, p < 0.05 c. χ2(1) = 11.85, p < 0.05 d. χ2(1) = 11.85, N = 1012, p < 0.05 92. A chi-square test for goodness of fit has a sample size of 35. What are the degrees of freedom for this chi square? a. k – 1 b. 33 c. 34 d. k – 2 Enter the appropriate word(s) to complete the statement. 93. If 5 deaths result from 100 car accidents in which a person is wearing a seat belt and 33 deaths result from 90 car accidents in which a person is not wearing a seat belt, a person is _______ times more likely to die if not wearing a seat belt.

94. If both the independent and dependent variables are measured as scale variables, then you can use either correlation or _______ to analyze the data.

Page 17

Name:

Class:

Date:

Chap_17_5e 95. To construct a graph of the results of the chi-square test for independence, you would first calculate _______.

96. With one nominal variable, you should use a(n) _______ to analyze the data.

97. If 5 deaths result from 100 car accidents in which a person is wearing a seat belt and 33 deaths result from 90 car accidents in which a person is not wearing a seat belt, the conditional probability of dying given that a person is not wearing a seat belt is _______.

98. Using a nonparametric test increases the risk of making a(n) _______ error.

99. With two nominal variables, you should use a(n) _______ to analyze the data.

100. Post hoc tests for the chi square involve calculating the _______.

Page 18

Name:

Class:

Date:

Chap_17_5e 101. The difference between the observed and expected frequencies in a cell divided by the standard error is the _______.

102. When graphing chi-square results, _____ or _____ are used rather than frequencies.

103. The critical cutoff value for a chi-square test for independence is _______ for a design with two variables with four and two levels, respectively, and a p level of 0.05.

104. If an adjusted standardized residual for a cell is more extreme than _______, then we can conclude that the observed frequency differs significantly from the expected frequency.

105. The effect size for chi square is _____.

106. The size of an effect with chi-square can be quantified through _______.

Page 19

Name:

Class:

Date:

Chap_17_5e 107. _______ is calculated by taking the ratio of two conditional probabilities.

108. The degrees of freedom are equal to the number of _______ minus 1 for the chi-square test for goodness of fit.

109. It would be inappropriate to use a parametric hypothesis test if the dependent variable was measured on a(n) _____ or a(n) _____ scale.

110. If both the independent and dependent variables are measured as _______ variables, then you can use either correlation or regression to analyze the data.

111. With _______ nominal variable(s), you should use a chi-square test for independence to analyze the data.

112. With _______ nominal variable(s), you should use a chi-square test for goodness of fit to analyze the data.

Page 20

Name:

Class:

Date:

Chap_17_5e 113. If 5 deaths result from 100 car accidents in which a person is wearing a seat belt and 33 deaths result from 90 car accidents in which a person is not wearing a seat belt, the conditional probability of dying given that a person is wearing a seat belt is _______.

114. (Table: Social Media Usage by Millennials) In a 2019, the Web site Civic Science (civicscience.com) surveyed millennials about their online behaviors as part of a project comparing older millennials, 30–34 years of age, to younger millennials, 25–29 years of age. One question asked about their daily usage of social media. Social media usage was coded as "Never or less than 1 hour per day," "1–2 hours per day," or "More than 2 hours per day." The table depicts the number of younger and older millennials in each response category. (a) Perform the appropriate hypothesis test to determine whether social media usage differs among older and younger millennials. (b) Calculate the appropriate measure of effect size for these data. Is this a small, medium, or large effect? (c) Graph these data. Table: Social Media Usage by Millennials

Younger millennials (25– 29 years old) Older millennials (30–34 years old)

Never/Less Than 1 Hour

1–2 Hours

More Than 2 Hours

1578

1278

1442

2561

1593

1536

Page 21

Name:

Class:

Date:

Chap_17_5e 115. (Table: Closeted or Out by Gender) In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays. One of the questions the researchers were interested in was whether a person's "outness" depended on his or her gender. All participants were classified as closeted or as out. The table depicts the number of students who were closeted or out as a function of their gender. Based on these data, does a person's "outness" depend on gender? Table: Closeted or Out by Gender

Male Female

Closeted 21 23

Out 20 16

116. A prospective college student was interested in whether the school was considering attending was as racially diverse as the three-county region surrounding it. He found census data and school data regarding the racial diversity of the counties and the university. The census data indicated that the region was 15% Asian, 48% black, and 37% white. Assume that the accepted freshman class at the target university totals 1042, and that 729 students self-identify as white, 202 as black, and 111 as Asian. Perform steps 3 through 6 of hypothesis testing to determine whether the distribution of students at this university reflects the racial diversity of the region.

117. In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband. Of the 80 participants, 27 assigned a verdict of guilty, 49 a verdict of not guilty by reason of self-defense, and 4 a verdict of not guilty by reason of insanity. (a) Perform the six steps of hypothesis testing using a chi-square test for goodness of fit to determine whether the distribution of verdicts differs from what would be expected by chance. (b) Calculate effect size for this test and interpret its meaning.

Page 22

Name:

Class:

Date:

Chap_17_5e 118. HIV testing is of particular importance during pregnancy because antiretroviral drugs can be administered in an effort to prevent transmission of the virus to the fetus. Anderson and Sansom (2007) used data from the Behavioral Risk Factor Surveillance System (an ongoing national telephone survey) to determine whether pregnant women were getting tested for HIV at a higher rate than nonpregnant women. Results of the survey indicated that in the general population of women aged 18 to 44, 15.4% had been tested for HIV within the past year. Of the 4855 pregnant women surveyed, 2627 had been tested for HIV within the past year. Perform steps 3 through 6 of hypothesis testing to answer the question whether pregnant women are being tested for HIV at a greater rate than nonpregnant women.

119. In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays. The researchers asked each participant to rate his or her "outness" on a scale from 1 to 8. Based on their rating, all participants were classified as closeted or as out. One question that the researchers wished to address is whether "outness" depends on gender. (a) The researchers analyzed their data using the categorization out versus closeted as their measure of "outness." Given this measure, what kind of analysis did the authors use to determine whether outness varies with gender? (b) What kind of analysis would the authors have conducted if they had used the original rating scale of "outness" as the dependent variable? Why? (c) Explain why using the original rating scale as the dependent measure would be preferable to using the categorization of closeted or out.

120. Is there a preferred age at which women get pregnant? A national large-sample survey conducted in the United States found that 1391 of pregnant women participants were between the ages of 18 and 24, 2691 were between the ages of 25 and 34, and 773 were between the ages of 35 and 44 (Anderson & Sansom, 2007). Perform steps 3 through 6 of hypothesis testing to determine whether there is a preferred age range within which women tend to get pregnant.

Page 23

Name:

Class:

Date:

Chap_17_5e 121. Sit and colleagues (2018) investigated the efficacy of adjunctive bright-light therapy in patients with bipolar disorder. Patients were randomly assigned to receive either bright-light therapy or dim-light therapy along with their regular medication. After six weeks, patients' clinical status was coded as improved or not improved. (a) Identify the variables and the measurement scale for each variable. (b) Based on your responses, indicate what kinds of analyses would be appropriate to use in this situation.

122. (Table: Abuse) In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband. In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical). The dependent variable was the number of guilty verdicts. (a) Perform the six steps of hypothesis testing using a chi-square test for independence to determine the effects of the type of abuse and the race of the defendant on the number of guilty verdicts. (b) Calculate the effect size for this test and explain its meaning. (c) Calculate the relative likelihood of being found guilty of emotional abuse if African American rather than Caucasian. The data appear in the table. Table: Abuse

African American Caucasian

Emotional Abuse 19 16

Physical Abuse 6 7

123. Corsi et al. (2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program. In the sample of 491 drug users, 38.1% were single; 27.5% were married or had a partner; and 34.4% were divorced, separated, or widowed. Of the drug users who entered a treatment program, 67 were single, 66 were married or had a partner, and 61 were divorced, widowed, or separated. Based on these data, would you say that marital status is associated with entry into a treatment program?

Page 24

Name:

Class:

Date:

Chap_17_5e 124. (Table: Light Therapy) Sit and colleagues (2018) investigated the efficacy of adjunctive bright-light therapy in patients with bipolar disorder. Patients were randomly assigned to receive either bright-light therapy or dimlight therapy along with their regular medication. After six weeks, patients' clinical status was coded as improved or not improved. (a) Perform the six steps of hypothesis testing using a chi-square test for independence to determine the effects of the type of light therapy on the patients' clinical outcomes. (b) Calculate the effect size for this test and explain its meaning. (c) Calculate the relative likelihood of clinical improvement if a patient receives bright-light therapy rather than dim-light therapy. The data appear in the table. Table: Light Therapy

Bright Light Dim Light

Improved 15 4

Not Improved 8 19

125. (Table: Education Levels of Pregnant Women) Is the amount of education completed by women who are currently pregnant different from the amount of education completed by nonpregnant women? The data from a national large-sample survey of U.S. women found that 10.3% have less than a high school education, 28.7% have a high school diploma, 31.3% have at least some college, and 29.7% have graduated from college (Anderson & Sansom, 2007). Use these proportions to determine the expected frequencies for the population, and test the idea that the education of pregnant women differs from that of women in the population, using the data for pregnant women in the table. Table: Education Levels of Pregnant Women Education < High school High school graduate Some college College graduate

Frequency 453 1356 1332 1713

Page 25

Name:

Class:

Date:

Chap_17_5e Answer Key 1. False 2. True 3. True 4. False 5. False 6. True 7. False 8. False 9. False 10. False 11. True 12. False 13. False 14. True 15. False 16. True 17. True 18. False 19. True 20. False 21. False 22. True 23. False 24. True 25. True 26. False Copyright Macmillan Learning. Powered by Cognero.

Page 26

Name:

Class:

Date:

Chap_17_5e 27. False 28. True 29. True 30. False 31. a 32. d 33. d 34. d 35. a 36. c 37. c 38. a 39. b 40. c 41. a 42. c 43. d 44. b 45. b 46. a 47. c 48. d 49. d 50. b 51. a 52. b 53. b 54. d Copyright Macmillan Learning. Powered by Cognero.

Page 27

Name:

Class:

Date:

Chap_17_5e 55. b 56. b 57. a 58. b 59. a 60. b 61. c 62. a 63. a 64. a 65. c 66. b 67. b 68. d 69. b 70. d 71. a 72. d 73. b 74. b 75. a 76. c 77. d 78. a 79. b 80. c 81. c 82. a Copyright Macmillan Learning. Powered by Cognero.

Page 28

Name:

Class:

Date:

Chap_17_5e 83. b 84. d 85. c 86. c 87. d 88. d 89. d 90. c 91. b 92. a 93. 7.333, 7.3 94. regression 95. conditional proportions 96. chi-square test for goodness of fit 97. 0.3667, 0.37 98. Type II 99. chi-square test for independence 100. adjusted standardized residuals 101. adjusted standardized residuals 102. proportions; percentages 103. 7.815 104. 2.0 105. 106. relative risk 107. Relative risk 108. categories, cells 109. nominal; ordinal 110. scale Copyright Macmillan Learning. Powered by Cognero.

Page 29

Name:

Class:

Date:

Chap_17_5e 111. two or 2 112. one, 1 113. 0.05 114. (a) The critical value for the chi-square test for independence based on a p level of 0.05 and df = 2 is 5.992. The calculated statistic is χ2 = 78.51. Because the calculated statistic exceeds the critical value, we reject the null hypothesis. Daily social media usage varies with age. (b) Cramér's ϕ = 0.09. This is a small effect size, according to Cohen's conventions for interpreting Cramér's ϕ. (c) First, we must calculate conditional probabilities:

Younger Older

Never/Less Than 1 Hour 1578/4298 = 0.367 2561/5690 = 0.450 4139

1–2 Hours 1278/4298 = 0.297 1593/5690 = 0.280 2871

More Than 2 Hours 1442/4298 = 0.336 1536/5690 = 0.270 2978

4298 5690 9988

Now we can graph these probabilities:

115. The critical value for the chi-square test for independence based on a p level of 0.05 and df = 1 is 3.841. The calculated statistic is χ2 = 0.49. Because the calculated statistic does not exceed the critical value, we fail to reject the null hypothesis. We have no evidence that a person's "outness" depends on gender. 116. Step 3: The comparison distribution is a chi-square distribution with df = 2. Step 4: The critical value based on a p level of 0.05 and df = 2 is 5.992. Step 5: χ2 = 496.84. Step 6: The calculated statistic exceeds the critical value. Therefore, we reject the null hypothesis. The profile of racial diversity at the university is not the same as the profile of racial diversity in the three-county region surrounding the school.

Page 30

Name:

Class:

Date:

Chap_17_5e 117. (a) Step 1: The comparison distribution will be a chi-square distribution. Three of the four assumptions of the chi square are met: The dependent variable is nominal, and each observation is independent of the other observations, but it is unlikely that the participants were randomly selected. There are three cells in the design; at least five times as many participants as cells, or 15 participants, are recommended, so this recommendation has been met with 80 people. Step 2: Null hypothesis: Defendants are equally likely to be assigned a verdict of guilty, not guilty by reason of self-defense, and not guilty by reason of insanity. Research hypothesis: There will be a difference in proportion of verdicts among the three categories. Step 3: The comparison distribution is a chi-square distribution with df = 2. Step 4: The critical value based on a p level of 0.05 and df = 2 is 5.992. Step 5: χ2 = 37.97. Category Guilty Self-defense Insanity

Observed 27 49 4

Expected 26.667 26.667 26.667

O–E 0.333 22.333 –22.667

(O – E)2 0.111 498.763 513.793

(O – E)2/E 0.0042 18.703 19.267

Step 6: The calculated statistic exceeds our critical value. Therefore, we reject the null hypothesis. The verdicts assigned to the defendants are not equally distributed among the three verdict categories. (b) The effect size is the square root of 37.97/(80)(2) = 0.49. According to Cohen's guidelines, this is a strong effect size. 118. Step 3: The comparison distribution is a chi-square distribution with df = 1. Step 4: The critical value based on a p level of 0.05 and df = 1 is 3.841. Step 5: χ2 = 5583.75. Step 6: The calculated statistic exceeds the critical value. Therefore, we reject the null hypothesis. Women who are pregnant are getting tested for HIV at a greater rate than women in the general population. 119. (a) Given the measure employed by the researchers, the analysis they would have used is a chi-square test for independence. (b) If the researchers had used their original scale, treating the data obtained as scale, then they could have used a parametric test such as the independent-samples t test. (c) Scale data convey more precise differences than do nominal data. Therefore, hypothesis tests using scale data typically have greater power to detect significant differences than do hypothesis tests using nominal data. 120. Step 3: The comparison distribution is a chi-square distribution with df = 2. Step 4: The critical value based on a p level of 0.05 and df = 2 is 5.992. Step 5: χ2 = 1184.48. Step 6: Our calculated statistic exceeds our critical value. Therefore, we reject the null hypothesis. Pregnant women are not equally likely to be in any one of the three age categories. There appears to be a preferred age for pregnancies.

Page 31

Name:

Class:

Date:

Chap_17_5e 121. (a) There are two variables: (1) the type of light therapy, with levels bright light or dim light; and (2) is the patient's clinical status, with levels improved or not improved. Both are nominal variables. (b) It would be appropriate to use a chi-square test for independence for this study. 122. (a) Step 1: The comparison distribution will be a chi-square distribution. Three of the four assumptions for the chisquare test of independence are met: The dependent variable is nominal, each participant is classified into only one cell, and the number of participants is greater than five times the number of cells, but the participants were not randomly selected. Step 2: Null hypothesis: The difference in the number of guilty verdicts for emotional and physical abuse cases does not depend on the defendant's race. Research hypothesis: The difference in the number of guilty verdicts for emotional and physical abuse cases depends on the defendant's race. Step 3: The comparison distribution is a chi-square distribution with df = 1. Step 4: The critical value based on a p level of 0.05 and df = 1 is 3.841. Step 5: χ2 = 0.25. Category African American, Emotional African American, Physical Caucasian, Emotional Caucasian, Physical

Observed 19 6 16 7

Expected 18.23 6.77 16.77 6.23

O–E 0.77 –0.77 –0.77 0.77

(O – E)2 (O – E)2/E 0.593 0.033 0.593 0.088 0.593 0.035 0.593 0.095

Step 6: The calculated statistic does not exceed the critical value. Therefore, we fail to reject the null hypothesis. We do not have evidence that the proportion of guilty verdicts for emotional and physical abuse cases depends on the defendant's race. (b) The effect size is the square root of 0.251/(48)(1) = 0.072. According to Cohen's guidelines, this is even smaller than a small effect size. (c) The relative likelihood of being found guilty of emotional abuse if African American rather than Caucasian is calculated as follows: Chance of being found guilty of emotional abuse if African American = 19/25 = 0.76 Chance of being found guilty of emotional abuse if Caucasian = 16/23 = 0.70 Relative likelihood = 0.76/0.70 = 1.086 Thus, the chance of being found guilty of emotional abuse if the defendant is African American is 1.086 times the chance if the defendant is Caucasian.

123. Of the 194 drug users who entered a treatment program, the expected frequency of those who were single is 73.91; of those who were married or partnered, it is 53.35; and of those who were divorced, widowed, or separated, it is 66.74. The critical value for the chi-square test for goodness of fit based on a p level of 0.05 and df = 2 is 5.992. The calculated statistic is χ2 = 4.14. Because the calculated statistic does not exceed the critical value, we fail to reject the null hypothesis. We have no evidence that marital status is associated with entry into a treatment program. Copyright Macmillan Learning. Powered by Cognero.

Page 32

Name:

Class:

Date:

Chap_17_5e 124. (a) Step 1: The comparison distribution will be a chi-square distribution. Three of the four assumptions for the chisquare test of independence are met: The dependent variable is nominal, each participant is classified into only one cell, and the number of participants is greater than five times the number of cells, but the participants were not randomly selected. Step 2: Null hypothesis: The likelihood of clinical improvement does not depend on the type of light therapy the patient receives. Research hypothesis: The likelihood of clinical improvement depends on the type of light therapy the patient receives. Step 3: The comparison distribution is a chi-square distribution with df = 1. Step 4: The critical value based on a p level of 0.05 and df = 1 is 3.841. Step 5: χ2 = 10.85. Category Bright Light, Improved Bright Light, Not Improved Dim Light, Improved Dim Light, Not Improved

Observed 15 8 4 19

Expected 9.5 13.5 9.5 13.5

O–E 5.5 -5.5 -5.5 5.5

(O – E)2 (O – E)2/E 30.25 3.184 30.25 2.241 30.25 3.184 30.25 2.241

Step 6: The calculated statistic exceeds the critical value. Therefore, we reject the null hypothesis. The data suggest that the type of light therapy a patient with bipolar disorder receives impacts the likelihood of clinical improvement. (b) The effect size is the square root of 10.85/(46)(1) = 0.49. According to Cohen's guidelines, this is a large effect size. (c) The relative likelihood of clinical improvement if a patient receives bright-light therapy rather than dim-light therapy is calculated as follows: Chance of improvement if receiving bright-light therapy = 15/23 = 0.65 Chance of improvement if receiving dim-light therapy = 4/23 = 0.17 Relative likelihood = 0.65/0.17 = 3.82 Thus, the likelihood of clinical improvement if the patient receives bright-light therapy is 3.82 times the likelihood if the patient was treated with dim-light therapy. 125. The expected frequency for women who have less than a high school education is 499.96; for high school graduates, it is 1393.10; for those with some college education, it is 1519.30; and for college graduates, it is 1441.64. The critical value for the chi-square test for goodness of fit based on a p level of 0.05 and df = 3 is 7.815. The calculated statistic is χ2 = 79.57. Because our calculated statistic exceeds our critical value, we reject the null hypothesis. The distribution of pregnant women among levels of education is not the same as the distribution of women in the general population.

Page 33

Name:

Class:

Date:

Chap_18_5e Indicate whether the statement is true or false. 1. When a researcher is investigating a question that involves two scale variables, the regression equation can be an appropriate means of testing a hypothesis. a. True b. False 2. A researcher is investigating whether celebrities, such as movie stars and musicians, die at an earlier age than do members of the general public. The researcher collects age of death data for a group of recently deceased celebrities and compares their age of death to mortality data from the National Center for Health Statistics. In answering this question, the appropriate statistical test is an independent-samples t test. a. True b. False 3. A researcher compares depressed and nondepressed individuals to see if they differ in the number of Instagram followers they have. For this comparison, an independent-samples t test would be appropriate. a. True b. False 4. The Methods section of a research paper is where the researchers provide justification for the sample size that they used. a. True b. False 5. When investigating a question involving a scale dependent variable and two or more nominal independent variables, a one-way between-groups ANOVA is the appropriate test. a. True b. False 6. A researcher would use a factorial ANOVA to test a hypothesis about levels of depression among men and women who are either single or never married; separated or divorced; and married or partnered. a. True b. False 7. A researcher would use a one-way between-groups ANOVA to test a hypothesis about levels of depression among three groups of men: single or never-married men; separated or divorced men; and married or partnered men. a. True b. False

Page 1

Name:

Class:

Date:

Chap_18_5e 8. A researcher would use a one-way within-groups ANOVA to test a hypothesis about levels of depression among three groups of men: single or never-married men; separated or divorced men; and married or partnered men. a. True b. False 9. A researcher compares depressed and nondepressed individuals to see if they differ in the number of Instagram followers they have. For this comparison, a paired-samples t test would be appropriate. a. True b. False 10. The Results section of a research paper is where the researchers describe the statistical analyses that they employed. a. True b. False 11. A researcher is investigating whether celebrities, such as movie stars and musicians, die at an earlier age than do members of the general public. The researcher collects age of death data for a group of recently deceased celebrities and compares their age of death to mortality data from the National Center for Health Statistics. In answering this question, the appropriate statistical test is a z test or a single-sample t test. a. True b. False 12. When investigating a question involving a scale dependent variable and a nominal independent variable (IV) with three or more levels in which participants receive all levels of the IV, a one-way between-groups ANOVA is the appropriate test. a. True b. False 13. When investigating a question involving two nominal variables, the appropriate test is a chi-square test of independence. a. True b. False 14. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by the population, a z test or a single-sample t test is appropriate. a. True b. False

Page 2

Name:

Class:

Date:

Chap_18_5e 15. When investigating a question involving a single nominal variable, the appropriate test is a chi-square test of independence. a. True b. False 16. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by the population, a paired-samples t test or an independent-samples t test is appropriate. a. True b. False 17. The Methods section of a research paper is where the researchers describe the statistical analyses that they employed. a. True b. False 18. When investigating a question involving a scale dependent variable and two or more nominal independent variables, a one-way within-groups ANOVA is the appropriate test. a. True b. False 19. When investigating a question involving a scale dependent variable and a nominal independent variable (IV) with three or more levels in which different participants receive each level of the IV, a one-way betweengroups ANOVA is the appropriate test. a. True b. False 20. When investigating a question involving a single nominal variable, the appropriate test is a chi-square test for goodness of fit. a. True b. False 21. When investigating a question involving only nominal variables, the appropriate test is a chi-square test. a. True b. False 22. When investigating a question involving a scale dependent variable and a nominal independent variable (IV) with three or more levels in which different participants receive each level of the IV, a one-way within-groups ANOVA is the appropriate test. a. True b. False

Page 3

Name:

Class:

Date:

Chap_18_5e 23. When investigating a question involving a scale dependent variable and a nominal independent variable (IV) with three or more levels in which participants receive all levels of the IV, a one-way within-groups ANOVA is the appropriate test. a. True b. False 24. When reporting the results of your research, you need to report only the statistically significant results of your hypothesis tests. a. True b. False 25. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where each level is represented by a sample, a paired-samples t test or an independent-samples t test is appropriate. a. True b. False 26. When reporting the results of your research, you should include new statistics such as confidence intervals and effect sizes in addition to traditional findings. a. True b. False 27. When investigating a question involving a scale dependent variable and two or more nominal independent variables, a factorial ANOVA is the appropriate test. a. True b. False 28. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by the population, a paired-samples t test or an independent-samples t test is appropriate. a. True b. False 29. When investigating a question involving a scale dependent variable and a nominal independent variable with three or more levels, an independent-samples t test is the appropriate test. a. True b. False 30. The Pearson correlation coefficient is appropriate to use when a researcher wishes to investigate the extent to which family income predicts a student's score on a college entrance exam. a. True b. False Copyright Macmillan Learning. Powered by Cognero.

Page 4

Name:

Class:

Date:

Chap_18_5e 31. When reporting the results of your research, you should include the results of all statistical tests conducted, whether they were statistically significant or not. a. True b. False 32. When investigating a question involving two nominal variables, the appropriate test is a chi-square test for goodness of fit. a. True b. False 33. When a researcher is investigating a question that involves two scale variables, the chi-square test of independence is an appropriate hypothesis test. a. True b. False 34. The Results section of a research paper is where the researchers provide justification for the sample size that they used. a. True b. False 35. When a researcher is investigating a question that involves two scale variables, the Pearson correlation coefficient is an appropriate hypothesis test. a. True b. False 36. A regression equation allows an education researcher to investigate the extent to which family incomes predicts a student's score on a college entrance exam. a. True b. False Indicate the answer choice that best completes the statement or answers the question. 37. Dr. Blake is interested in the effects of intoxication on driving ability. Participants in her study are assigned to consume one to four mixed drinks during a one-hour period. Dr. Blake records the number of driving errors made while the participants navigate a 20-mile course in a driving simulator. What statistical test should Dr. Blake use to analyze these data? a. Pearson correlation coefficient b. one-way between-groups ANOVA c. one-way within-groups ANOVA d. chi-square test of independence

Page 5

Name:

Class:

Date:

Chap_18_5e 38. When investigating a question involving a scale dependent variable and a single nominal independent variable (IV) with three or more levels in which participants receive all levels of the IV, the appropriate test is a(n): a. independent-samples t test. b. paired-samples t test. c. one-way within-groups ANOVA. d. one-way between-groups ANOVA. 39. Does spending time online lead to greater unhappiness? To investigate this question, a researcher administers a depression inventory to a group of 40 individuals. The researcher divides these participants into two groups of low and high screen time based on data collected from their smartphones. The appropriate statistical test is a(n): a. Pearson correlation. b. paired-samples t test. c. independent-samples t test. d. one-way between-groups ANOVA. 40. When investigating a question involving a single nominal variable, the appropriate test is a: a. z test. b. single-sample t test. c. chi-square test for independence. d. chi-square test for goodness of fit. 41. Which of the following is NOT considered one of the traditional results of hypothesis testing reported in a research paper? a. the symbol for the statistic used b. the p value associated with the test statistic c. effect size d. degrees of freedom 42. To investigate whether one-pound bags of potato chips actually contain a full pound of chips, a skeptical statistics student goes to the supermarket and weighs every bag on the shelf. To analyze these data, the student should use a: a. chi-square test for goodness of fit. b. z test. c. single-sample t test. d. one-way between-groups ANOVA.

Page 6

Name:

Class:

Date:

Chap_18_5e 43. To investigate whether men and women differ in regard to their favorite movie genre, a social scientist surveys 100 college students and notes whether they prefer horror, comedy, drama, action/thriller, or scifi/fantasy. To analyze these data, the scientist should use a(n): a. independent-samples t test. b. one-way between-groups ANOVA. c. chi-square test for independence. d. chi-square test for goodness of fit. 44. Recent polling finds that 61% of the American public supports same-sex marriage, 31% is opposed to samesex marriage, and 8% is undecided. To see if students at your university have attitudes that mirror the larger U.S. population, you survey 120 students on your campus on their attitudes toward same-sex marriage. To analyze these data, the most appropriate test would be a: a. chi-square test for goodness of fit. b. one-way between-groups ANOVA. c. single-sample t test. d. z test. 45. To investigate serotonin's role in suicidal behavior, the amounts of serotonin metabolites in the cerebral spinal fluid of 15 violent-suicide attempters are compared to the levels found in a group of 20 nonviolent-suicide attempters. The levels of serotonin metabolites are coded as low or normal/high, and the two groups are compared. These data would most appropriately be analyzed with a(n): a. paired-samples t test. b. independent-samples t test. c. factorial ANOVA. d. chi-square test for independence. 46. The APA now requires researchers to include "new statistics" in the results sections. Which of the following is an example of one of these new statistics? a. justification for the sample size used in the study b. effect size c. whether the study was preregistered d. the actual value of the statistic 47. To investigate whether time spent online is related to levels of depression, a psychologist collects depression scores from 100 college freshmen, along with data from their phones about their past week's screen time. To address this question, the psychologist should use a: a. Pearson correlation. b. regression. c. one-way between-groups ANOVA. d. chi-square test for goodness of fit. Copyright Macmillan Learning. Powered by Cognero.

Page 7

Name:

Class:

Date:

Chap_18_5e 48. To answer the question of whether there is an association between the number of adverse events an individual experiences in childhood and the person's level of depression at age 20, the researcher should use a: a. Pearson correlation. b. regression. c. one-way between-groups ANOVA. d. chi-square test for goodness of fit. 49. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by a population that has a known mean and standard deviation, a(n) _____ test is appropriate. a. z b. paired-samples t test c. single-sample t test d. independent-samples t test 50. Which section of a research report includes information about the psychometric properties of the measures used in the study? a. Abstract b. Data c. Methods d. Results 51. Does spending time online lead to greater unhappiness and if so, are these effects different for men and women? To investigate this question, a researcher administers a depression inventory to a group of 60 individuals. The researcher divides these participants into three groups of low, medium, and high screen time based on data collected from their smartphones and compares the data from the male and female participants. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. factorial ANOVA. 52. A clinical researcher is examining the relations between two scale variables. What kind of statistical test should the researcher use to determine if the scores on one variable predict the scores on the other variable? a. regression b. Pearson correlation c. chi-square test of independence d. independent-samples t test

Page 8

Name:

Class:

Date:

Chap_18_5e 53. Which of the following is NOT part of the Results section of a research report? a. summary statistics, such as means, standard deviations, and sample sizes b. results of hypothesis testing c. tables and figures d. psychometrics for the measures used 54. Does spending time online lead to greater unhappiness? To investigate this question, a researcher administers a depression inventory to a group of 45 individuals. The researcher divides these participants into three groups of low, medium, and high screen time based on data collected from their smartphones. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. factorial ANOVA. 55. To investigate serotonin's role in suicidal behavior, the amounts of serotonin metabolites in the cerebral spinal fluid of 15 violent-suicide attempters are compared to the levels found in a group of 20 nonviolent-suicide attempters. These data would most appropriately be analyzed with a(n): a. paired-samples t test. b. independent-samples t test. c. factorial ANOVA. d. chi-square test for independence. 56. Which badge developed by the Open Science Framework indicates that the researchers have made their experimental materials available to others wishing to conduct their own replication of the study? a. Preregistration b. Open Data c. Open Materials d. Open Access 57. Which of the following is NOT something that would be included in the Methods section of a research report? a. details of the statistical analyses b. information regarding preregistration of the study c. data regarding the reliability and validity of any measures used d. justification for the sample size used in the study

Page 9

Name:

Class:

Date:

Chap_18_5e 58. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by a population for which the standard deviation is unknown, a(n) _____ test is appropriate. a. z b. paired-samples t test c. single-sample t test d. independent-samples t test 59. In a research report, confidence intervals and effect sizes are part of: a. psychometrics. b. traditional findings. c. new statistics. d. summary statistics. 60. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where different participants are exposed to the different levels of the independent variable, a(n) _____ test is appropriate. a. z b. paired-samples t test c. single-sample t test d. independent-samples t test 61. To answer the question of whether the number of adverse events an individual experiences in childhood predicts the person's level of depression at age 20, the researcher should use a: a. Pearson correlation. b. regression. c. one-way between-groups ANOVA. d. chi-square test for goodness of fit. 62. Does cell phone use interfere with a person's ability to drive? To investigate this question, a researcher recruits a group of 24 participants. He randomly assigns half to use their phone while driving in a test simulator; the other half are instructed to put their phones away. The researcher records the number of driving errors during a 20-minute driving simulation. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. paired-samples t test.

Page 10

Name:

Class:

Date:

Chap_18_5e 63. A professor wants to know if student test grades in her class are normally distributed. The professor examines the number of grades that fall one, two, or more than two standard deviations above and below the class mean. To analyze these data, the professor should use a(n): a. independent-samples t test. b. one-way between-groups ANOVA. c. chi-square test for independence. d. chi-square test for goodness of fit. 64. Which of the following is one of the Open Science Framework badges that recognize ethical and transparent practices? a. Transparency b. Preregistered c. Open Access d. Replication 65. The IMRAD format used for reporting scientific results stands for: a. Introduction, Methods, Results, and Discussion. b. Introduction, Materials, Results, and Discussion. c. Introduction, Methods, Results, Analysis, and Data. d. Introduction, Methods, Results, Abstract, and Discussion. 66. Which of the following is NOT one of the parts of a traditional research report known as the IMRAD format? a. Methods b. Results c. Abstract d. Discussion 67. The Open Science Framework has developed different badges to recognize ethical and transparent practices. Which badge indicates that the researchers have made their raw data available to others wishing to conduct their own analyses of the data? a. Open Statistics b. Open Data c. Open Materials d. Open Access

Page 11

Name:

Class:

Date:

Chap_18_5e 68. Do people's moods change with the season of the year? To investigate this question, a researcher recruits a group of 24 participants and administers a depression inventory to them at four different time points—once during each season of the year. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. factorial ANOVA. 69. A clinical researcher is interested in studying cerebral asymmetry and depression. The researcher recruits 36 participants. Each participant completes a depression inventory to assess his or her level of depression and is coded as either depressed or nondepressed. Each participant then has his or her resting brain electrical activity recorded, and based on the brain activity is coded as right- or left-activated. What kind of statistical test could be used to address this question? a. one-way between-groups ANOVA b. Pearson correlation c. chi-square test of independence d. independent-samples t test 70. When investigating a question involving a scale dependent variable and two or more independent variables, the appropriate statistical test is a: a. regression. b. one-way between-groups ANOVA. c. factorial ANOVA. d. chi-square test for independence. 71. Does cell phone use interfere with a person's ability to drive? To investigate this question, a researcher recruits a group of 24 participants and has them drive a simulated test course on two occasions. On one occasion, the participants are instructed to use their phone while driving; on the other occasion, they are instructed to put their phones away. The researcher records the number of driving errors during a two 15minute driving simulations. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. paired-samples t test.

Page 12

Name:

Class:

Date:

Chap_18_5e 72. Which of the following is NOT one of the Open Science Framework badges that recognize ethical and transparent practices? a. Open Access b. Open Data c. Open Materials d. Preregistered 73. Does breastfeeding lead to increases in IQ scores? To investigate this question, researchers compared IQ scores from 125 young adults who had either not been breastfed, breastfed for less than three months, breastfed for three to six months, or breastfed for more than six months. The appropriate statistical test to use in this situation is a: a. chi-square test for goodness of fit. b. independent-samples t test. c. single-sample t test. d. one-way between-groups ANOVA. 74. When investigating a question involving two nominal variables, the appropriate test is a(n): a. Pearson correlation. b. independent-samples t test. c. chi-square test for independence. d. chi-square test for goodness of fit. 75. A clinical researcher is interested in studying cerebral asymmetry and depression. The researcher recruits 36 participants. Each participant completes a depression inventory to assess his or her level of depression. Each participant then has his or her resting brain electrical activity recorded, and a left/right asymmetry score is calculated. What kind of statistical test could be used to address this question? a. one-way between-groups ANOVA b. Pearson correlation c. chi-square test of independence d. independent-samples t test 76. In which section of a research paper does the researcher provide information regarding the justification for the sample size utilized? a. Introduction b. Methods c. Results d. Analysis

Page 13

Name:

Class:

Date:

Chap_18_5e 77. Does cell phone use interfere with a person's ability to drive? To investigate this question, a researcher recruits a group of 24 participants. He randomly assigns one-third to use their phone to make calls while driving in a test simulator, one-third to send text messages to a friend while driving on the course, and onethird to keep their phones turned off while they are driving. The researcher records the number of driving errors during a 20-minute driving simulation. The appropriate statistical test is a(n): a. one-way within-groups ANOVA. b. one-way between-groups ANOVA. c. independent-samples t test. d. paired-samples t test. 78. Does breastfeeding lead to increases in IQ scores? To investigate this question, researchers collected IQ scores from 275 young adults who had been breastfed for six months or more as infants and compared those scores to the known U.S. population data (µ = 100, σ = 15). The appropriate statistical test to use in this situation is a: a. chi-square test for goodness of fit. b. z test. c. single-sample t test. d. paired-samples t test. 79. A clinical researcher is examining the relations between two scale variables. What kind of statistical test should the researcher employ to determine if there is an association between the variables? a. chi-square test of association b. Pearson correlation c. chi-square test of independence d. independent-samples t test 80. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where participants are exposed to both levels of the independent variable, a(n) _____ test is appropriate a. z b. paired-samples t test c. single-sample t test d. independent-samples t test

Page 14

Name:

Class:

Date:

Chap_18_5e 81. Dr. Blake is interested in the effects of intoxication on driving ability. Participants in her study are encouraged to consume one to four mixed drinks during a one-hour period. Dr. Blake records the number of driving errors made while the participants navigate a 20-mile course in a driving simulator and measures the participants' blood alcohol content at the end of the simulation. What statistical test should Dr. Blake use to analyze these data? a. Pearson correlation coefficient b. one-way between-groups ANOVA c. one-way within-groups ANOVA d. chi-square test of independence 82. Dr. Blake is interested in the effects of intoxication on driving ability. Participants in her study are assigned to one of four usage conditions: no alcohol and no cannabis, high alcohol and no cannabis; no alcohol and high cannabis; and high alcohol and high cannabis. Dr. Blake then records the number of driving errors made while the participants navigate a 20-mile course in a driving simulator. What statistical test should Dr. Blake use to analyze this data? a. paired-samples t test b. one-way within-groups ANOVA c. one-way between-groups ANOVA d. two-way between-groups ANOVA 83. To investigate whether time spent online predicts subsequent levels of depression, a psychologist collects depression scores from 100 college freshmen, along with data from their phones about their past week's screen time. To address this question, the psychologist should use a: a. Pearson correlation. b. regression. c. one-way between-groups ANOVA. d. chi-square test for goodness of fit. 84. Does breastfeeding lead to increases in IQ scores? To investigate this question, researchers collected IQ scores from 1275 young adults as well as data on how long these individuals were breastfed as infants. The appropriate statistical test to use in this situation is a(n): a. one-way between-groups ANOVA. b. z test. c. Pearson correlation. d. independent-samples t test.

Page 15

Name:

Class:

Date:

Chap_18_5e 85. A clinical researcher is interested in studying cerebral asymmetry and depression. The researcher recruits 36 participants. Each participant completes a depression inventory to assess his or her level of depression and is coded as either depressed or nondepressed. Each participant then has his or her resting brain electrical activity recorded, and a left/right asymmetry score is calculated. What kind of statistical test could be used to address this question? a. one-way between-groups ANOVA b. Pearson correlation c. chi-square test of independence d. independent-samples t test Enter the appropriate word(s) to complete the statement. 86. When investigating a question involving a single _______ variable, the appropriate test is a chi-square test for goodness of fit.

87. When reporting the new statistics, researchers should include information about _______ and confidence intervals.

88. When investigating a question involving a single nominal variable, the appropriate test is a chi-square test for _______.

89. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by a population that has a known mean and standard deviation, a(n) _______ test is appropriate.

Page 16

Name:

Class:

Date:

Chap_18_5e 90. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where different participants are exposed to the different levels of the independent variable, a(n) _______ t test is appropriate.

91. The format used for reporting scientific results,_______, stands for the four sections of a standard research report.

92. When investigating a question involving a scale dependent variable and two or more independent variables, a(n) _______ ANOVA is the appropriate test.

93. When investigating a question involving two nominal variables, the appropriate test is a chi-square test for _______.

94. When reporting the new statistics, researchers should include information about effect size and _______.

95. When investigating a question involving a scale dependent variable and a single _______ independent variable (IV) with three or more levels in which participants receive all levels of the IV, a one-way withingroups ANOVA is the appropriate test.

Page 17

Name:

Class:

Date:

Chap_18_5e 96. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where one level is represented by the sample and the other level is represented by the population, for which the standard deviation is unknown, a(n) _______ test is appropriate.

97. When investigating a question involving a scale dependent variable and _______ independent variable(s), a factorial ANOVA is the appropriate test.

98. When investigating a hypothesis involving two scale variables, you can use either _______ or regression to analyze the data.

99. When investigating a question involving a scale dependent variable and a(n) _______ independent variable (IV) with three or more levels in which different participants receive each level of the IV, a one-way between-groups ANOVA is the appropriate test.

100. When investigating a hypothesis involving two scale variables, you can use either correlation or _______ to analyze the data.

Page 18

Name:

Class:

Date:

Chap_18_5e 101. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where participants are exposed to both levels of the independent variable, a(n) _______ t test is appropriate.

102. The _______ badge developed by the Open Science Framework indicates that the researchers have made their data available for others who wish to replicate their analyses.

103. The _______ section of a research paper is intended to inform the reader about how the data were collected and to provide justification for the sample size used in the study. It is also the section where the authors would provide information regarding preregistration of the study.

104. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where participants are exposed to _______ levels of the independent variable, a paired-samples t test is appropriate.

105. When investigating a question involving a scale dependent variable and a nominal independent variable with two levels where participants are exposed to _______ levels of the independent variable, an independentsamples t test is appropriate.

Page 19

Name:

Class:

Date:

Chap_18_5e 106. The _______ section of a research paper is where the authors report their analyses and findings.

107. When investigating a question involving a scale dependent variable and a single nominal independent variable (IV) with three or more levels in which participants receive all levels of the IV, a one-way _______ ANOVA is the appropriate test.

108. When investigating a question involving two _______ variables, the appropriate test is a chi-square test for independence.

109. When investigating a question involving a scale dependent variable and a nominal independent variable (IV) with three or more levels in which different participants receive each level of the IV, a one-way _______ ANOVA is the appropriate test.

110. The _______ badge developed by the Open Science Framework indicates that the researchers have made their experimental materials available for others who wish to replicate their analyses.

111. What is the purpose of the badges developed by the Open Science Framework, and what do these different badges represent?

Page 20

Name:

Class:

Date:

Chap_18_5e 112. To investigate whether there are differences in student performance between the fall and spring semesters, a statistics professor examines final test grades for students enrolled in the fall and spring semesters of the same academic year. The 58 students enrolled in the fall semester had an average test grade of 73.22 points, with a standard deviation of 16.24. The 57 students enrolled in the spring had an average test grade of 72.02, with a standard deviation of 11.86. The 95% confidence interval for the difference between the two means ranged from –4.05 to 6.47, and the effect size was 0.08. How would you report these statistics in APA format?

113. A graduate student in psychology is interested in determining whether you actually get what you pay for when you purchase beer. Does the price paid actually reflect differences in quality, or is it simply an issue of marketing and hype? The graduate student recruits 12 fellow graduate students and has them blind taste three brands of beer that differ in their selling price: a low-priced bargain beer, a medium-priced national beer, and a high-priced craft beer. The beers are presented in random order, and each beer is rated on a one to ten scale. What statistical test would the student use to analyze these data and why?

114. Are there gender differences in pet ownership? To investigate this question, a psychology student surveys shoppers at a local shopping mall and records whether they own cats, dogs, some other type of pet, or no pets at all. What sort of statistical test would the student use to analyze their data, and why?

115. Identify and briefly describe the sections of a standard research report.

Page 21

Name:

Class:

Date:

Chap_18_5e 116. An economist is interested in whether the price of an object impacts an individual's perception of the object's quality. To study this question, the economist has seven individuals taste samples of two red wines that they are told cost either $10 or $90. In reality, both wines are exactly the same. The order of presentation is randomized, and all subjects taste both wines and rate the wines on a 100-point scale. The mean for the $10 wine was 56.71, with a standard deviation of 14.61; the mean for the $90 wine was 61.71, with a standard deviation of 15.65. The calculated test statistic was 3.01 and the p value was 0.024. The 95% confidence interval for the difference between the two means ranged from 0.93 to 9.07, and the effect size was 1.14. How would you report these statistics in APA format?

117. What statistics-related information is presented in the Methods section of a research report?

118. What are traditional and new statistics, and where are they reported in the standard research paper?

119. Dr. Daniels is interested in studying the impact of texting on students' test grades in his psychology class. He collects data from the students on the number of text messages sent and received during the past week, and classifies the students as either low- or high-texting users and then compares these two groups on the number of points they earned on their next test. (a) What sort of statistical test would be used in this situation? (b) If Dr. Daniels had not split the students into two groups, what sort of statistical procedure could he have used, and why, to investigate the question of texting and test performance?

Page 22

Name:

Class:

Date:

Chap_18_5e Answer Key 1. True 2. False 3. True 4. True 5. False 6. True 7. True 8. False 9. False 10. True 11. True 12. False 13. True 14. True 15. False 16. False 17. False 18. False 19. True 20. True 21. True 22. False 23. True 24. False 25. True 26. True Copyright Macmillan Learning. Powered by Cognero.

Page 23

Name:

Class:

Date:

Chap_18_5e 27. True 28. False 29. False 30. False 31. True 32. False 33. False 34. False 35. True 36. True 37. b 38. d 39. c 40. d 41. c 42. c 43. c 44. a 45. d 46. b 47. a 48. a 49. a 50. c 51. d 52. a 53. d 54. b Copyright Macmillan Learning. Powered by Cognero.

Page 24

Name:

Class:

Date:

Chap_18_5e 55. b 56. c 57. a 58. c 59. c 60. d 61. b 62. c 63. d 64. b 65. a 66. c 67. b 68. a 69. c 70. c 71. d 72. a 73. d 74. c 75. b 76. b 77. b 78. b 79. b 80. b 81. a 82. d Copyright Macmillan Learning. Powered by Cognero.

Page 25

Name:

Class:

Date:

Chap_18_5e 83. b 84. c 85. d 86. nominal 87. effect size 88. goodness of fit 89. z 90. independent-samples 91. IMRAD 92. factorial 93. independence 94. confidence intervals 95. nominal 96. single-sample t 97. two or more 98. correlation 99. nominal 100. regression 101. paired-samples 102. Open Data 103. Methods 104. both 105. different 106. Results 107. within-groups 108. nominal 109. between-groups 110. Open Materials Copyright Macmillan Learning. Powered by Cognero.

Page 26

Name:

Class:

Date:

Chap_18_5e 111. Suggested Answer: The badges developed by the Open Science Framework are intended to encourage researchers to engage in ethical and transparent practices in their research endeavors. These badges are often included in a Note section at the end of the paper. Currently, three badges can be earned: (1) The Open Data badge indicates that the researchers have made their data available, so other researchers can analyze the data themselves; (2) the Open Materials badge recognizes researchers who willing make their research materials (e.g., stimuli, questionnaires) available to anyone interested in replicating their study; and (3) the Preregistered badge indicates that the researchers preregistered their study. 112. Suggested Answer: The results failed to support the hypothesis that student performance differs in statistics as a function of the semester enrolled. Students in the fall semester scored only slightly higher on their final exam (M = 73.22; SD = 16.24) than students enrolled in the spring semester (M = 72.02; SD = 11.86), t(113) = 0.454, p = 0.650, d = 0.08, 95% confidence interval (CI) = [–4.05, 6.47]. 113. Suggested Answer: The student would use a one-way within-groups ANOVA to analyze the results of this study because the participants were exposed to all levels of the independent variable, beer price point, and the dependent variable, rating score, can be considered a scale variable. 114. Suggested Answer: The student would use a chi-square test for independence to answer this research question, as both variables, gender and pet preference, are nominal variables. 115. Suggested Answer: The four sections of a standard research report are known by the acronym IMRAD, which stands for Introduction, Methods, Results, and Discussion. In the Introduction section, the authors summarize the results of earlier research, explain the reasons for the current study, and describe the study and their research hypotheses. The Methods section is where the researchers describe who the study participants were and how the study was conducted. Information is also provided about any materials used, along with any psychometric data, and what the justification was for the study's sample size. The Results section is where the researchers report their analyses and findings—descriptive statistics for the study, along with both the results of traditional hypothesis tests and new statistics such as effect size and confidence intervals. The Discussion section summarizes the findings, puts them in context, and suggests directions for future research. 116. Suggested Answer: The results supported the hypothesis that product price impacts consumers' perception of quality. When participants were drinking what they thought was $90 wine, they rated the wine higher (M = 61.71; SD = 15.65) than wine that they thought cost $10 per bottle (M = 56.71; SD = 14.61, t(6) = 3.01, p = .024, d = 1.14, 95% confidence interval (CI) = [0.93, 9.07].

Page 27

Name:

Class:

Date:

Chap_18_5e 117. Suggested Answer: The statistics-related information presented in the Methods section includes information as to whether the study was preregistered and where that preregistration can be found, so that readers can review it, if desired. The authors should also provide justification for the study's sample size, which is typically determined on the basis of a power analysis. The Methods section also includes information about the plan for data analysis, including criteria for discarding data, and any psychometric data (i.e., reliability and validity) for each measurement scale used. 118. Suggested Answer: Both traditional and new statistics are reported in the Results section of a research paper. Traditional statistics include summary statistics such as means, standard deviations, and sample sizes. For each hypothesis test, traditional statistics include the symbol for the statistic used, the degrees of freedom, the actual value of the statistic, and the p value for that statistic. The new statistics recommended by the APA include information about effect size and any appropriate confidence intervals. 119. Suggested Answer: (a) Dr. Daniels would use an independent-samples t test to compare the test grades of high- and low-texting students. (b) Had Dr. Daniels not split the students into two groups, he could have used a Pearson correlation to examine the association between number of text messages and test grade, as both are scale variables.

Page 28