Instructor Manual For Statistics A Tool for Social Research and Data Analysis 11e Joseph F. Healey Christopher Donoghue Chapter 1-15 Chapter 1 New to this Edition
Updated Learning Objectives for the chapter Updated “Using Statistics” box Updated “Statistics in Everyday Life” box on Push Polls Updated “The Goals of This Text” section Updated “Statistics in Everyday Life” box on Using Descriptive Statistics Updated “Statistics in Everyday Life” box on Using Inferential Statistics Updated “Statistics in Everyday Life” box on Changes in Socioeconomic Status in the U.S. New graph on Percent of Americans Identifying as Lower Class Some section titles have been changed for clarity Added one problem Updated “Reading Statistics” box
Learning Objectives: By the end of this chapter, students will be able to 1. 2. 3. 4.
Identify the key stages and terms in social scientific research Distinguish between descriptive and inferential statistics Provide examples of discrete and continuous variables Describe three levels of measurement and cite examples of each.
Chapter Summary The text begins by explaining the role of statistics in the research process. The discussion is guided by the "Wheel of Science" as conceptualized by Walter Wallace (Figure 1.1). The text always presents statistics in the context of the research enterprise. That is, statistics are presented as useful tools for answering sociological questions and testing social science theories, never as ends in themselves. The chapter also distinguishes between descriptive and inferential statistics and univariate, bivariate, and multivariate statistics. The distinction between discrete and continuous variables and the concept of level of measurement is presented in this chapter and the latter is stressed throughout the text as an organizational device and as a major criterion for selecting statistics appropriately. Exercises are provided at the end of the chapter for reviewing the characteristics of the three levels of measurement used in this text.
Chapter 2 BASIC DESCRIPTIVE STATISTICS: Tables, Percentages, Ratios and Rates, and Graphs
New to this Edition
Updated Learning Objectives for this chapter. Updated “Using Statistics” box at beginning of chapter Edited Tables 2.2 and 2.3 on Religious Affiliations Updated Table 2.4 on Religious Affiliations Updated Table 2.5 and Table 2.6 on Declared Major Fields of Two College Campuses Added Figure 2.1 on Percentages of People Living in Poverty by State Updated Table 2.7and 2.8 – Support for Birth Control on University Campuses Updated Table 2.9 and 2.10 – Ages of Students in a College Class Updated Table 2.11 – Finding Midpoints Updated “One Step at a Time: Finding Midpoints” box Updated Table 2.12 – Real Class Limits Updated Table 2.13 and 2.14 – Age of Students in a College Class Updated Table 2.15 – Distribution of Income by Household, United States, 2017 DELETED “Applying Statistics 2.3: Frequency Distributions” Added Social Research and Data Analysis 2.3: Frequency Distributions Updated Using SPSS: Frequency Distributions” box DELETED “Applying Statistics 2.4: Ratios” Added “Social Research and Data Analysis 2.4: Ratios” box DELETED “Applying Statistics 2.5: Rates” Added “Social Research and Data Analysis 2.5: Rates” box DELETED “Applying Statistics 2.6: Percentage Change” Updated Table 2.16 – Projected Population Growth for Six Nations, 2018-2050 Added new information to the Projected Population sections Added Table 2.17: Self-Described Religious Affiliation of Adult Americans, 2018 Added Figure 2.2: Self-Describe Religious Identification of Adult Americans, 2018 Added Figure 2.3: Self-Described Religious Identification of Adult Americans, 2018 DELETED Figures 2.4 to 2.3 – homicide rates Added Figure 2.4: Suicide Rates for Males and Females by Age Group, 2017 DELETED Figures 2.5 to 2.4 and changed example to age of US population Added Figure 2.5: Unemployment Rate and Earnings by Educational Attainment, 2018 Added Figure 2.6: Age Distribution of the United States, 2017 Added Figure 2.7: Age Distribution of the Population of the United States by Gender, 2017 Updated “Using SPSS: Graphs” box with 2 figures (pie chart and histogram) Added Table 2.18: U.S. Households by Type, 2018 Updated sections on Martial Status Updated Table 2.19 on Martial Status for Selected Years Added Figure 2.8: Rates of Marriage and Divorce, 1950-2017 Updated 4 Problems and 4 tables Updated “You Are the Researcher: Is There a “Culture War” section and 4 Step boxes Some sections were updated for clarity
Learning Objectives: By the end of this chapter, students will be able to 1. 2. 3. 4. 5.
Explain how descriptive statistics can be used to make your data understandable Construct frequency distributions for variables at each of the three levels of measurement Compute percentages, proportions, rations, rates, and percentage change for numerical data Analyze pie and bar charts, histograms, and line graphs Create frequency distributions in SPSS and analyze the output
Chapter Summary This chapter covers relatively simple descriptive devices: frequency distributions, percentages and proportions, ratios, rates, percent change, pie and bar charts, histograms, and line charts. The emphasis is on frequency distributions and the construction and interpretation of these tables for variables measured at each of the three different levels. Instructors may want to supplement this material with additional examples of each technique and/or graphs and charts, especially those created by software such as Microsoft Excel. The underlying 'theme' of this chapter is the need to present results clearly; to communicate results accurately and concisely but without losing too much detail. Social Research and Data Analysis 2.3, for example, is intended to contrast the anarchy of raw, unorganized data with the clarity and simplicity of the frequency distributions. Chapter 3 MEASURES OF CENTRAL TENDENCY
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box DELTED Table 3.1: Overseas Travelers Added Table 3.1: Household Living Arrangements in the U.S., 2018 Updated Table 3.2: Distributions of Scores of Two Tests Updated Table 3.3: Distributions of Test Scores DELETED “Statistics in Everyday Life: The Changing American Religious Profile” Added “Statistics in Everyday Life: Communication between Dating Teenagers” Updated Table 3.4 and 3.5 on Finding the Median Added section on Using Percentiles to Find the Median and the Mode Updated Table 3.6 to Frequency Distribution for the Number of Children in the 2018 General Social Survey Updated “Social Research and Data Analysis 3.1: Finding Measures of Position” using U.S. divorce rates for states and Table 3.7 Added “One Step at a Time: Finding the Mean” box Updated Table 3.8: A Demonstration Showing that all Scores Cancel Out Around the Mean Updated Table 3.9: A Demonstration Showing That the Mean is the Point of Minimized Variation Updated Table 3.10: A Demonstration Showing That the Mean is Affected by Every Score Added section discussing “skewness” Updated “Statistics in Everyday Life” box on Baseball Salaries Updated “Using SPSS: Measures of Central Tendency” box Updated Table 3.11: The Relationship Between Level of Measure and Measures of Central Tendency
Added Figure 3.4 Updated Table 3.12: Choosing a Measure of Central Tendency Updated “Social Research and Data Analysis 3.2: The Mean, Mode, and Median” Added section, “Interpreting Statistics: Choosing a Measure of Central Tendency to Describe Income and added U.S. Census graph Updated data for Problems 3.4, 3.6, 3.11, 3.16, Added Problem 3.12 DELETED previous Problem 3.14 and created new
Learning Objectives: By the end of this chapter, students will be able to 1. Explain to others how measures of central tendency can make data understandable 2. Calculate the appropriate measure of central tendency for variables at each of the three levels of measurement 3. Explain the different in the types of information provided by the mode, median, and mean 4. Explain the mathematical characteristics of the mean, especially how it is affected by skew 5. Produce measures of central tendency using SPSS and explain their meaning
Chapter Summary This chapter presents three measures of central tendency: the mode (or the most common score), the median (or the middle score), and the mean (or the average score). The chapter stresses the importance of level of measurement in selecting a measure of central tendency and emphasizes the mean as the most important and commonly used measure. I discuss the algebraic and mathematical characteristics of the mean and stress the point that the mean is affected by every score in a distribution and, therefore, is pulled in the direction of extreme scores relative to the median. This theme of the "sensitivity" of the mean is picked up in several of the end-of-chapter problems. The chapter includes a section on other measures of location besides the median. The chapter closes with a section on selecting an appropriate measure of central tendency for a given purpose and a given level of measurement. Chapter 4 MEASURES OF VARIATION
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box at beginning of chapter Changed all references to “dispersion” to “variation” Updated Table 4.1: Birth Rates Added section “Using a Frequency Distribution to Find the Range and the Interquartile Range” Updated Table 4.2: What do you Think is the Ideal Number of Children…” Updated “One Step at a Time: Finding the Interquartile Range” box Updated “Statistics in Everyday Life: Increasing Racial Diversity” box Updated Table 4.3: A Demonstration that the Sum of the Deviations of the Scores…” Updated Table 4.4: Computing the Standard Deviation” Updated “One Step at a Time: Finding the Standard Deviation (s)…” box Updated “Social Research and Data Analysis 4.1: The Standard Deviation” box
Updated Table 4.5: Computing the Standard Deviation for Two Campuses Added Figure 4.2: Boxplot for Birth Rates for 99 Nations, 2018 Added Figured 4.3: Boxplots for Birth Rates by Income Level…, 2018 Updated Table 4.6: Summary Statistics for Birth Rate by Income Levels for 99 Nations, 2018 Updated “Statistics in Everyday Life: Life Expectancy around the Globe” box Updated Table 4.7: Age on a College Campus in Two Different Years Updated “Social Research and Data Analysis 4.2: Describing Variation Updated “Using SPSS: Measures of Variation” Added Figure 4.4 Added Figure 4.5 Updated “Reading Statistics: Pornography and Religiosity” Updated “Statistics in Professional Literature” Added new Problems 4.3, 4.6, 4.11, 4.16, and 4.17 Updated Problems 4.4, 4.5, 4.9, 4.10 and 4.12, Updated “You are the Researcher” section
Learning Objectives: By the end of this chapter, you will be able to 1. Explain the purpose of measures of variation and the information they convey. 2. Compute and explain the range (R), the interquartile range (Q), the standard deviation (s), and the variance (s2). 3. Select an appropriate measure of variation for a given level of measurement. 4. Interpret the components of a boxplot. 5. Explain the meaning of the standard deviation and the variance. 6. Use SPSS to calculate the range and the standard deviation. Chapter Summary This chapter covers the range and quartile deviation for ordinal and interval-ratio variables, and the standard deviation and variance for interval-ratio variables, and boxplots. The emphasis is on the standard deviation as the most important and commonly used measure of dispersion. The material on the standard deviation begins by stating some characteristics that a reasonable measure of dispersion should have. Our purpose here is to demonstrate that statistics don't just 'happen' -- they are invented to fulfill specific purposes and designed according to an explicit logic. Beginning students often have trouble with the interpretation of the standard deviation. In several sections, we demonstrate several ways in which this statistic can be interpreted. Also, almost all of the end-of-chapter problems require interpretation as well as computation and some "model" interpretations are supplied in the Answers to Odd Numbered Problems at the back of the text. Chapter 5 THE NORMAL CURVE
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box at beginning of chapter Updated data to include 2018 GSS Reformatted table “Between” and “Lies” on standard deviation
Updated “Statistics in Everyday Life: Tests of Intelligence” Updated “One Step at a Time: Finding Z Scores” Updated Table 5.1: “An Illustration of How to Find Areas Under the Normal Curve…” Updated Table 5.2: “Finding Positive Z Scores with….” Updated Table 5.3: “Finding Negative Z Scores with….” Updated Table 5.4: “Finding Areas Above and Below Positive and Negative Scores” Updated “One Step at a Time: Finding Areas Above and Below…” Updated Table 5.5: “Finding Areas Between Scores” Updated “One Step at a Time: Finding Areas Between Z Scores” Updated “One Step at a Time: Finding Probabilities” Updated “Social Research and Data Analysis 5.3: Finding Probabilities” Updated “Statistics in Everyday Life: Probability and Theory in Action” Updated data in Problems 5.3, 5.4, 5.5, 5.6, and 5.9 Updated “You are the Researcher” section and accompanying bar graphs
Learning Objectives: By the end of this chapter, you will be able to 1. Explain the normal curve and its characteristics 2. Convert numbers around the mean into Z scores 3. Use Z scores and the normal curve table (Appendix A) to find areas above, below, and between points on the curve 4. Express areas under the curve in terms of possibility
Summary This chapter introduces the normal curve. In combination with the mean and standard deviation, the normal curve is used to describe the area above, below, and between scores. Numerous problems are provided at the end of the chapter for homework and/or in-class demonstrations of computing Z scores and finding areas. This chapter is intended to bridge the course from descriptive to inferential statistics. The section on probability is particularly important because it is the major treatment of this concept in this text. Students should acquire two main ideas from this section: (1) probabilities can be estimated using the properties of the normal curve (even though the application is not particularly realistic or commonly used), and (2) it is extremely likely that scores selected randomly from a normally distributed variable will be close in value to the mean of the distribution. In Part II, these ideas will be linked to the concept of the sampling distribution and used to estimate the probability of a particular sample outcome from the theoretical distribution of all possible sample outcomes (that is, the sampling distribution). Chapter 6 INTRODUCTION TO INFERENTIAL STATISTICS Sampling & the Sampling Distribution
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box at beginning of chapter Updated data to include 2018 GSS
Updated new “Statistics in Everyday Life: Advertising and Sampling” box Updated section on Simple Random Sampling Updated “One Step at a Time: a Simple Random Sample” box Updated “One Step at a Time: Drawing a Systematic Random Sample” Updated section on Stratified Random Sampling Updated “One Step at a Time: Drawing a Stratified Random Sample” box Updated section on Cluster Sampling Updated “One Step at a Time: Drawing a Cluster Sample” box Updated “Social Research and Data Analysis 6.1: The American Community Survey” Updated “Statistics in Everyday Life: Sampling and Telephone Soliciting” Added new examples and data to Two Theorems section Added new “Summary: Two Theorems” section Updated “Statistics in Everyday Life: Using the GSS to Measure Changing American Attitudes” Updated “Statistics in Everyday Life: Using Surveys to Compare American Attitudes with Other Nations” Updated Table 6.1: Symbols for Means and Standard Deviations of Three Distributions” Added definition for Sampling frame Updated Problem 6.2 Updated Reading Statistics: Who Offers Extra Credit?
Learning Objectives: By the end of this chapter, students will be able to 1. Define these key terms: population, sample, parameter, statistic, representative, EPSEM 2. Describe four different kinds of EPSEM (Equal Probability of Selection Method) sampling strategies 3. Differentiate between the sampling distribution, the sample, and the population 4. Explain how inferential statistics can be used to generalize from a sample to a population 5. Explain the two theorems presented in the chapter
Summary This chapter begins with a brief overview of probability sampling techniques, but the primary goal is to familiarize students with the concept of the sampling distribution and its relationship to samples and populations. This concept is, of course, both difficult for students and crucial for an understanding of inferential statistics. We have tried to provide a more concrete view of this abstract concept by discussing, in detail, the construction and final shape of a sampling distribution. Also, we have applied the basic concepts of sample, population, and sampling distribution to the General Social Survey, the data set used for end-of-chapter SPSS exercises. Chapter 7 ESTIMATION PROCEDURES
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box at beginning of chapter Updated data to include 2018 GSS
Update Table 7.1, “Standard Deviation of the Sampling Distributions of Two Samples, Assuming…” Updated Table 7.2, “Scores for Various Levels of Alpha” Update “One Step at a Time” boxes (Constructing Confidence…& Interpreting the Confidence Level) Updated “Social Research and Data Analysis 7.1: Estimating a Population Mean” Updated “One Step at a Time” boxes (Constructing Confidence Intervals…& Interpreting the Confidence Level) Updated “Social Research and Data Analysis 7.2: Estimating Population Proportions” Updated “Statistics in Everyday Life: The National Mood” Updated Table 7.3, “Choosing Formulas for Confidence Intervals” Updated Table 7.4, “Confidence Intervals Grow Wider as Confidence Levels Increase” Updated Table 7.5, “Confidence Intervals Grow Narrower as Sample Size Increases” Updated “Using SPSS: Constructing Confidence Intervals” Updated Table 7.6, “Polls and Actual Results, Presidential Race, 2016” Added Figure 7.6 Updated section on Presidential results for currency Updated “Statistics in Everyday Life: The Classic Election Prediction Error” Updated “Statistics in Everyday Life: The Big Miss by Pollsters in the 2016 Presidential Election” Updated “Reading Statistics: Public-Opinion Polls” Added new Problems 7.4 and 7.6, Updated Problem 7.16,
Learning Objectives: By the end of this chapter, students will be able to 1. Explain the logic of estimation and the role of the sample, sampling distribution, and the population 2. Define the concepts of bias and efficiency 3. Calculate confidence intervals using sample means and sample proportions 4. Explain the relationships between the confidence level, sample size, and the width of the confidence interval 5. Use SPSS to construct confidence intervals with sample means and proportions
Summary We find that estimation procedures, as presented in this chapter, provide a more straightforward entree into inferential statistics than hypothesis testing. Students are more familiar with estimation, at least as consumers, and the logic involved is somewhat simpler. Once this material is covered, students will understand the concept of the sampling distribution, alpha levels, and several other pieces of the logic that will be introduced and developed in the hypothesis testing chapters. Thus, the overall goal of the chapter is partly to teach students how to estimate population values and partly to build more groundwork for hypothesis testing. The chapter emphasizes the construction of interval estimates for large sample means but also includes the techniques for estimating proportions. There is a full discussion of the relationships between interval width, sample size, and confidence level. Chapter 8 HYPOTHESIS TESTING I The One-Sample Case
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box at beginning of chapter Updated data to include 2018 GSS Updated “Statistics in Everyday Life: Testing Drugs” Updated “One Step at a Time: Completing Step 4 of the Five-Step Model…” Updated Table 8.1, “Making a Decision in Step Five…” Updated “One Step at a Time: Completing Step 5 of the Five-Step Model…” Updated “Statistics in Everyday Life: Hypothesis Testing and Gambling” Updated Table 8.2, “One- vs. Two-Tailed Tests…” Updated Table 8.4, “The Relationship Between Alpha and Z…” Updated Table 8.5, “Decision Making and the Five-Step Model” Updated “Statistics in Everyday Life: Alpha Levels” Updated Table 8.6, “Choosing a Sampling Distribution When Testing…” Updated “One Step at a Time: Completing Step 4 of the Five-Step Model” Updated “One Step at a Time: Completing Step 5 of the Five-Step Model” Updated “Research and Data Analysis 8.1: a Sample Mean for Significance” Updated “One Step at a Time: Completing Step 4 of the Five-Step Model…” Updated “One Step at a Time: Completing Step 5 of the Five-Step Model…” Updated “Social Research and Data Analysis 8.2: Testing a Sample Proportion for Significance” Added new Problems 8.2, 8.3, and 8.6
Learning Objectives: By the end of this chapter, students will be able to 1. Explain the logic of hypothesis testing, including the concepts of the null hypothesis, the sampling distribution, the alpha level, and the test statistic. 2. Explain what it means to “reject the null hypothesis” or “fail to reject the null hypothesis”. 3. Identify and cite examples of situations in which one-sample tests and hypotheses are appropriate. 4. Use the five-step model to test the significance of single-sample means and proportions and correctly interpret the results. 5. Explain the difference between one- and two-tailed tests, and specify when each is appropriate. 6. Use the Student’s t distribution to conduct a hypothesis test on a small sample.
Summary This chapter introduces the logic of hypothesis testing in the context of testing for the significance of single sample means and proportions against a population value. Also, a five-step model for organizing all decisions and computations is introduced. This model is used throughout the hypothesis testing chapters to maximize continuity for the students. The chapter begins by considering the kinds of situations in which one-sample tests of hypothesis would be appropriate and then presents an example problem, in informal language, before presenting the formal concepts and processes. Figure 8.1 provides a visual overview of the process. Our purpose here is to attempt to focus the student's attention on the overall logic of hypothesis testing before confronting them with the process in detail or raising too many computational issues. This material is difficult for most
students and this section is intended to counteract their tendency to get bogged down in the details of the process by emphasizing the “big picture.” In the section following the example problem, the five-step model and the formal terminology of hypothesis testing are presented. Tests for means computed on large samples, means computed from small samples (the t distribution is introduced here), and proportions computed from large samples are covered. We also discuss the use of one- versus two-tailed tests and Type I and Type II errors. Chapter 9 HYPOTHESIS TESTING II The Two-Sample Case
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Updated “One Step at a Time: Completing Step 4 of the Five-Step Model” box Updated “Social Research and Data Analysis 9.1: A Test of Significance for Sample Means” and accompanying data Updated “Statistics in Everyday Life: Reading and Math” box Updated “Social Research and Data Analysis 9.2: Testing the Significance of the Difference Between Sample Proportions” to have data sets about Protestants and Catholics Updated “Reading Statistics: Comparing Ethnic and Racial Social Distance by Gender” Updated Problems 9.15 and 9.17 for currency Added new Problems 9.6 and 9.8
Learning Objectives By the end of this chapter, students will be able to 7. Identify and cite examples of situations in which the two-sample test of hypothesis is appropriate. 8. Explain the log of hypothesis testing as applied to the two-sample case. 9. Explain the concept of independent random sampling. 10. Perform a hypothesis test for two samples means and two sample proportions following the five-step model and correctly interpret the results. 11. Explain how each of the factors (especially sample size) affect the probability of rejecting the null hypothesis. 12. Use the Student’s t distribution to conduct a hypothesis test on small samples. 13. Explain the differences between statistical significance and importance. 14. Use SPSS to conduct a test of significance for the difference between sample means.
Summary
After considering the kinds of research situations in which two sample tests of significance are appropriate, this chapter presents tests of significance for sample means (both large and small samples) and sample proportions. Towards the end of the chapter, we consider the factors that affect the probability of rejecting the null. One of our purposes here is simply to increase awareness and understanding of the hypothesis testing process. More importantly, we want to make the distinction between statistical significance and theoretical and/or practical importance. We hope that by the time students encounter this section, they will be comfortable enough with the overall process to begin to appreciate the strengths and limitations of hypothesis testing. The theme of statistical significance versus importance is picked up again at the end of Chapter 11 and the beginning of Chapter 12 as a way of making the transition from inferential statistics to measures of association. Chapter 10 HYPOTHESIS TESTING III: The Analysis of Variance
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Updated the computations to find SST using Formula 10.1 Updated the data in “Social Research and Data Analysis 10.1: The Analysis of Variance” Updated “Using SPSS: The ANOVA Test” Updated the section on Post Hoc Analysis Corrected the spelling in Formula 10.9 Added new Problems 10.8 and 10.10 Updated the “You are the Researcher” section
Learning Objectives: By the end of this chapter, students will be able to 1. 2. 3. 4.
Cite examples of situations in which analysis of variance (ANOVA) is appropriate. Explain the logic of hypothesis testing as applied to ANOVA. Perform the ANOVA test, using the five-step model as a guide and interpret the results. Explain the concepts of population variance, total sum of squares, sum of squares between, sum of squares within, mean square estimates, and post hoc tests. 5. Use SPSS to conduct analysis of variance test.
Summary This chapter presents an introduction to the one-way analysis of variance test of significance. We begin with an example and stress the association between ANOVA and the t test for sample means presented in Chapter 9. This connection with previous material and the continued use of the five-step model should maximize continuity for students and the chapter includes an introduction to post-hoc analysis. Chapter 11 HYPOTHESIS TESTING IV Chi Square
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Added footnote to the sample of Formula 11.3 Updated the data in “Social Research and Data Analysis 11.1: The Chi Square Test” Updated “Using SPSS: Crosstabs and Chi Square” Updated “Social Research and Data Analysis 11.2” section Update “Reading Statistics: Chi Square in Action” Update Problem 11.7 for currency Update “You are the Researcher”
Learning Objectives: By the end of this chapter, students will be able to 1. 2. 3. 4. 5. 6. 7. 8.
Cite examples of situations in which the chi square test is appropriate Explain the structure of a bivariate table Explain the concept of independence as it relates to bivariate tables Explain the logic of hypothesis testing in terms of chi square Perform the chi square test using the five-step model and correctly interpret the results Conduct a goodness-of-fit test and correctly interpret the results Explain the limitations of the chi square test Use SPSS to conduct the chi square test of significance
Summary This chapter centers on the logic of and the computational routines for the chi square test of independence, but also includes an explanation of the structure of bivariate tables. Yate's correction for small samples is presented towards the end of the chapter, along with a restatement of the distinction between statistical significance and importance, first introduced in Chapters 9. This distinction is stressed, in part, as a way of making the transition to measures of association, which are more direct indicators of the "importance" of relationships between two variables. Chapter 12 BIVARIATE ASSOCIATION FOR NOMINAL- AND ORDINAL-LEVEL VARIABLES
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Placed material on Maximum Difference in a separate sub-section DELETED footnote from Table 12.1 DELETED footnote from Table 12.4 Updated “Social Research and Data Analysis 12.2: Using Cramer’s V and Lambda” Updated “Social Research and Data Analysis 12.3: Interpreting Direction with Ordinal-Level Variables Updated “Reading Statistics: Student Anxiety Social Statistics Courses”
Updated Problem 12.5 for currency
Learning Objectives: By the end of this chapter, you will be able to 15. Analyze the importance (vs. statistical significance) of a bivariate relationship using measures of association. 16. Explain the concept of association in the context of bivariate tables. 17. List and explain the three characteristics of a bivariate relationship. 18. Calculate column percentages and measures of association for bivariate tables. 19. Analyze a bivariate relationship using column percentages and a measure of association. 20. Compute and interpret Phi, Cramer’s V, and lambda for nominal-level variables. 21. Explain the concept of proportional reduction in error. 22. Computer and interpret Gamma and Spearman’s rho, for ordinal-level variables. 23. Use SPSS to calculate column percentages for a bivariate table and produce nominal and ordinal measures of association.
Summary This chapter introduces the concept of association between variables in the context of bivariate tables. The chapter begins with a definition of bivariate association, applies the definition to an example problem, and then presents three characteristics of a bivariate association. These three characteristics are used throughout Part III to provide continuity across the various measures of association. Procedures for calculating percentages for bivariate tables and interpreting the results are presented and the distinction between association and causation is made early in the chapter and then repeated in the summary. Bivariate tables in the text are always organized with the column variable increasing in value from left to right and the row variable increasing in value from top to bottom. Although there is no "right way" to set up tables, I choose this convention so as to make the tables comparable to output from statistics packages like SPSS. This chapter also a variety of measures of association: phi, Cramer's V, and lambda for nominal-level variables and gamma and Spearman’s rho for ordinal-level variables. There is a full explanation of the logic of proportional reduction in error for both lambda and gamma.
Chapter 13 ASSOCIATION BETWEEN VARIABLES MEASURED AT THE INTERVAL-RATIO LEVEL
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Updated “Statistics in Everyday Life: Who Does the Housework?” Added “Using SPSS: Scatterplots” and accompanying figure
Updated “Social Research and Data Analysis 13.1: Computing the Regression” data Added “Using SPSS: Producing a Correlation Matrix” box Updated “Reading Statistics: Predicting Knowledge Sharing in Organizations, Part 1” and accompanying table Added Table 13.6, Homicide Rates and Family Poverty, Basic Statistics Added Figure 13.8, Homicide Rate by Family Poverty Added footnote to Summary section Added new Problems 13.4 and 13.5 Updated Problems 13.6, 13.7, 13.8, 13.9, 13.10, 13.11 and 13.12 Updated “You are the Researcher” section
Learning Objective: By the end of this chapter, students will be able to 1. 2. 3. 4. 5.
Interpret a scattergram. Calculate and interpret slope (b), Y intercept (a), and Pearson’s r and r2. Find and explain the least-squares regression line and use it to predict values of Y. Explain the concepts of total, explained, and unexplained variance. Use regression and correlation techniques to analyze and describe a bivariate relationship in terms of the three questions introduced in Chapter 12. 6. Test Pearson’s r for significance.
Summary This chapter is a full treatment of regression and correlation. The chapter begins with the construction and interpretation of scatterplots and then proceeds to the least-squares regression line and the regression equation. Pearson's r is presented in a separate section, followed by a section on the logic and interpretation of the coefficient of determination. The chapter also includes sections on the correlation matrix, dummy variables, and testing the significance of r. The chapter ends with a section on “Interpreting Statistics” using crime data. This chapter presents many new concepts and a single example problem is used throughout to provide continuity and to help students integrate this material.
Chapter 14 ELABORATING BIVARIATE TABLES New to this Edition
Updated the Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Added new information to the “Controlling for a Third Variable” section Added footnote to the computation of gamma Added footnote to the original frequencies of Table 14.1 Updated “Statistics in Everyday Life: Spuriousness” Updated “Social Research and Data Analysis 14.1: Elaborating Bivariate Tables” DELETED section on Analyzing Tattoos Updated “Interpreting Statistics: Another Look at Concern for the Environment” Updated Table 14.6 and 14.7 Updated “Reading Statistics: Predicting Knowledge Sharing Attitudes, Part 2”
DELETED section on the Values of Gamma for Both Genders Updated Problems 14.7, 14.8, 14.9 and 14.10,
Learning Objectives: By the end of this chapter, you will be able to 24. Explain the purpose of multivariate analysis and the purpose of a control variable. 25. Construct and interpret partial tables. 26. Compute and interpret measures of association for partial tables. 27. Recognize and interpret direct, spurious, or intervening, and interactive relationships. 28. Compute and interpret partial gamma. 29. Explain the limitations of elaborating bivariable tables. 30. Use SPSS to conduct multivariate analysis with partial tables.
Summary This chapter presents elaboration as a method for multivariate analysis. We discuss the purpose of multivariate analysis, the techniques by which third variables are controlled, and three basic patterns of relationships between the partial tables and the original table. The overall logic of elaboration is summarized in Table 14.5. We have also included partial gamma, a consideration of where control variables come from, and a discussion of the limitations of elaboration as a technique. To enhance continuity, a single example problem has been used throughout the chapter. This example problem involves variables measured at the ordinal level to permit the introduction of partial gamma. Instructors may want to illustrate the process of elaboration using nominally measured variables and lambda as the measure of association.
Chapter 15 PARTIAL CORRELATION AND MULTIPLE REGRESSION AND CORRELATION
New to this Edition
Updated Learning Objectives for this chapter Updated “Using Statistics” box Updated data to include 2018 GSS Updated “Social Research and Data Analysis 15.2: Using SPSS to Determine…” Updated data in “Social Research and Data Analysis 15.3: Multiple Regression and Correlation” Updated “Statistics in Everyday Life: Statistics and Baseball” Updated data in “Reading Statistics: Do Sociology Courses Increase Empathy in Students” Updated data in “Multiple Regression in the Professional Literature” Update Problems 15.4, 15.5, 15.6, 15.7, 15.8, 15.9 and 15.10 Some sections were updated for clarity Updated “You are the Researcher” data sets
Learning Objectives: By the end of this chapter, students will be able to
1. 2. 3. 4. 5. 6.
Compute and interpret partial correlation coefficients. Find and interpret the least-squares multiple regression equation with partial slopes. Find and interpret standardized partial slopes or beta-weights (b*). Calculate and interpret the coefficient of multiple determination (R2). Explain the limits of partial and multiple regression analysis. Use SPSS to generate partial correlations and conduct multiple regression analysis
Summary This chapter begins with partial correlation. Students are occasionally referred back to Chapter 14 to provide continuity and underline the similarities in the multivariate techniques. I discuss partial correlation, multiple regression using partial slopes to predict scores on Y, and standardized partial slopes (beta-weights) are introduced as a way of assessing the relative importance of the independents. The chapter ends with the coefficient of multiple determination and an example of how to interpret the results of a regression analysis.