Test Bank For Statistics for Management and Economics 11th Edition by Gerald Keller Chapter 1_22

Page 1


Chapter 1 1.2 Descriptive techniques summarize data. Inferential techniques draw inferences about a population based on sample data.

1.3 a The population is the 25,000 registered voters. b The sample is the 200 registered voters. c The 48% figure is the statistic

1.4 a The population is the complete production run. b The sample is comprised of the 1,000 chips. c The parameter is the proportion of defective chips in the production run. d The statistic is the proportion of defective chips in the sample. e The 10% figure refers to the parameter. f The 7.5% figure refers to the statistic. g We can estimate the population proportion is 7.5%. Statistical inference methods will allow us to determine whether we have enough statistical evidence to reject the claim.as the sample proportion.

1.5 Draw a random sample from the population of graduates who have majored in your subject and a random sample of graduates of other majors and record their highest salary offers.

1.6 a Flip the coin (say 100 times) and record the number of heads (assuming that you are interested in the number of heads). b The population is composed of the theoretical result of flipping the coin an infinite number of times and recording either “heads” or “tails”. c The sample is comprised of the “heads” and “tails” in the sample. d The parameter is the proportion of heads (again assuming that your interest is the number of heads rather than tails) in the population. e The statistic is the proportion of heads (or tails depending on the choice made in part d). f The sample statistic can be used to judge whether the coin is actually fair.

1.7 a We would conclude that the coin is not fair. b We may conclude that there is some evidence that the coin is not fair.

1.8 a The population is made up of the propane mileage of all the cars in the fleet. b The parameter is the mean propane mileage of all the cars in the fleet. c The sample is composed of the propane mileage of the 50 cars.


d The statistic is the mean propane mileage of the 50 cars in the sample. e We can use the sample statistic to estimate the population parameter.


Chapter 2 2.1 Nominal: Occupation, undergraduate major. Ordinal: Rating of university professor, Taste test ratings. Interval: age, income 2.2 a Interval b Interval c Nominal d Ordinal

2.3 a Interval b Nominal c Ordinal d Interval e Interval

2.4 a Nominal b Interval c Nominal d Interval e Ordinal

2.5 a Interval b Interval c Nominal d Interval e Nominal

2.6 a Interval b Interval c Nominal d Ordinal e Interval

2.7 a Interval b Nominal c. Nominal

5


d Interval e Interval f Ordinal 2.8 a Interval b Ordinal c Nominal d Ordinal 2.9 a Interval b Nominal c Nominal 2.10 a Ordinal b Ordinal c Ordinal

2.11 a Nominal b Interval c Ordinal

2.12a Nominal b Interval c Interval d Interval

2.13

6


350 000 000 000 300 000 000 000 250 000 000 000 200 000 000 000 150 000 000 000 100 000 000 000 50 000 000 000 0

2.14

Percentage Brazil; 1,0%

United States; 2,3% United Arab Emirates; 6,3%

Venezuela; 19,1%

Canada; 11,0%

China; 1,6%

Iran; 10,1% Iraq; 9,2% Saudi Arabia; 17,2% Kazakhstan; 1,9% Libya; 3,1% Russia; 6,6%

Kuwait; 6,7%

Nigeria; 2,4%

Qatar; 1,6%

2.15

7


2.16 Residual fuel oil 3%

Marketable coke 5% Still gas 5%

Liquified Lubricants refinery 1% gas 3%

Asphalt and road oil 2%

Jet fuel 13%

Gasoline 51%

Distillate fuel oil 15%

2.17

8

Other 2%

United States

United Kingdom

Spain

Thailand

South Korea

Singapore

Saudi Arabia

Russia

Japan

Mexico

Italy

Iran

Indonesia

India

France

Germany

China

Brazil

Canada

Australia

20 000 000 18 000 000 16 000 000 14 000 000 12 000 000 10 000 000 8 000 000 6 000 000 4 000 000 2 000 000 0


6 000 000 000 000 5 000 000 000 000 4 000 000 000 000 3 000 000 000 000 2 000 000 000 000 1 000 000 000 000

Australia Brazil Canada China European Union France Germany India Italy Japan Korea, South Mexico Russia Saudi Arabia South Africa Spain Taiwan Turkey United Kingdom United States

0

2.18 9000,0 7706,8

8000,0 7000,0

5424,5

6000,0 5000,0 4000,0 3000,0 1591,1

2000,0

1000,0

417,7 541,0

765,6

1556,7

1098,0 528,6 407,9

528,1 443,6

0,0

2.19

9

438,2 451,2 519,9


Steel production 900,0 800,0 700,0 600,0 500,0 400,0 300,0 200,0 100,0 0,0

2.20 1 200 000 000 000 1 000 000 000 000 800 000 000 000

600 000 000 000 400 000 000 000 200 000 000 000 0

2.21

10


Other; 17,5% Metal; 4,1% Glass; 5,1%

Organic; 45,8%

Plastic; 10,3%

Paper; 17,2%

2.22 3500 3000 2500 2000 1500 1000 500 0

2.23

11


2.25

12 Tangerines

Strawberries

Plums and Sloes

Plantains

Pineapples

Persimmons

Pears

Peaches & Nectarines

Papayas

Oranges

Mangoes

Lemons & Limes

Kiwi Fruit

Grapes

Grapefruits

Dates

Cherries

Bananas

Avocados

Apricots

Apples

60 000

50 000

40 000

30 000

20 000

10 000

0

2.24

120

100

80

60

40

20

0


Minimum wage 11,40 11,20 11,00 10,80 10,60 10,40 10,20 10,00 9,80 9,60

11,25 10,70 10,20

10,60

10,55

10,45

10,50

10,50

10,30

10,20

Percent Earning Minimum Wage 14,00% 12,00% 10,00% 8,00% 6,00% 4,00% 2,00% 0,00%

11,70%

5,60% 2,20%

4,90%

5,90%

7,00%

3,30%

2.26

13

5,60% 6,00% 5,90%


Number of students 8%

Community 5%

Career focus 16%

Location 39%

Academic reputation 10%

Majors 22%

2.27 Internet 8%

Word of mouth 12%

Consumer guide 52% Dealership 28%

2.28

14


Living/dining room 9%

Basement 32%

Kitchen 27%

Bathroom 23%

Bedroom 9%

The basement is the top choice followed by kitchen, bathroom, bedroom, and living/dining room.

2.29 a

Newspaper

Frequency

Relative Frequency

Daily News

141

39.2%

Post

128

35.6%

Times

32

8.9%

WSJ

59

16.4%

b

New York Times 9%

Wall Street Journal 16% New York Daily News 39%

New York Post 36%

The Daily News and the Post dominate the market

15


2.30a

Degree

Frequency

BA

88

BBA

37

B Eng

51

B Sc

24

Other

30

b. 100

90

88

80 70 60

51

50 37

40

30 30

24

20 10 0 BA

BBA

Beng

BSc

Other

c Other 13%

B.Sc. 11%

B.A. 38%

B.Eng 22% B.B.A. 16%

d. About 4 applicants in 10 have the BA degree, about one-fifth have a BEng. and one-sixth have a BBA.

16


2.31a 45 39

40 35

30 25 25

21

20 13

15 10

5 0 HP

Lenovo

Dell

Other

b

HP; 21 Other; 25

Lenovo; 13

Dell; 39

c Dell is most popular with 40% proportion, followed by other, 26%, HP, 21% and Lenovo, 13%.

2.32 a

Software

Frequency

Excel

34

Minitab

17

SAS

3

SPSS

4

Other

12

b

17


Other 17%

SPSS 6% SAS 4%

Excel 49%

Minitab 24%

c Excel is the choice of about half the sample, one-quarter have opted for Minitab, and a small fraction chose SAS and SPSS.

2.33

Natural Light 9%

Other 6%

Bud Light 31% Miller Lite 21%

Michelob Light 4%

Busch Light 7% Coors Light 22%

2.34

18


Do not know 20% Many share 41%

Some share 39%

2.35 No opinion 3%

Fair share 20%

Too much 15% Too little 62%

2.36 a

19


Republicans Favor Poor 2%

Middle clas 29%

Rich 69%

b

Democrats Favor

Rich 29%

Poor 35%

Middle clas 36%

According to the survey Republicans favor the rich and Democrats are split among the middle class, poor, and rich.

2.37 a Category

`

Frequency

Relative Frequency

Mom: Full time, Dad: Full time

403

46.0%

Mom: Part time, Dad: Full time

149

17.0%

Mom: Not employed, Dad: Full time

228

26.0%

Mom: Full time, Dad: Part time or not employed

53

6.0%

Mom: Not employed, Dad: Not employed

18

2.1%

Other

26

3.0% 20


b Mom FT, Dad PT/Not 6%

Mom Not, Dad Not 2%

Mom Not, Dad FT 26%

Other 3%

Mom FT, Dad FT 46%

Mom PT, Dad FT 17%

c 450 Mom FT, Dad FT; 403 400 350 300

Mom Not, Dad FT; 228

250

Mom PT, Dad FT; 149

200 150

Mom FT, Dad PT/Not; 53 Mom Not, Dad Other; 26 Not; 18

100 50 0 Mom FT, Dad Mom PT, Dad FT FT

Mom Not, Dad FT

Mom FT, Dad PT/Not

Mom Not, Dad Not

Other

d In most households Dad is working full time. There are very few households where neither Mom nor Dad are working.

2.38

21


No opinion 2%

Favor 45% Oppose 53%

A small majority oppose the Affordable Care Act.

2.39a Views on social issues

Frequency

Relative Frequency

Liberal

322

31.4%

Moderate

328

32.0%

Conservative

375

36.6%

b

The country is split among the three views on social issues with a small plurality of conservatives.

22


2.40 a Views on economic issues

Frequency

Relative Frequency

Liberal

208

20.3%

Moderate

354

34.5%

Conservative

463

45.2%

b

Liberal 20%

Conservative 45%

Moderate 35%

Economically the country is conservative.

2.41 80000 70000 60000 50000

40000 30000 20000 10000 0 Education

Less than high school

Datenreihen1

High school

Datenreihen2

Some college College graduate

Datenreihen3

23

Datenreihen4


There is decreasing numbers of Americans who did not finish high school and increasing numbers of those that go to college.

2.42 180 000 160 000 140 000 120 000 100 000 80 000 60 000 40 000 20 000 0

Year 1995

Year 2000

Year 2005

Year 2008

Spending is increasing in all seven areas.

2.43 350 300 250 200 150 100 50 0

In general crime was decreasing until 2014 when it started increasing.

2.44

24


60 50

40 B.A. 30

B.Eng B.B.A.

20

Other 10 0 University 1

University 2

University 3

University 4

Universities 1 and 2 are similar and quite dissimilar from universities 3 and 4, which also differ. The two nominal variables appear to be related.

2.45

The column proportions are similar; the two nominal variables appear to be unrelated. There does not appear to be any brand loyalty.

2.46

25


The two variables are related.

2.47 700 600 500 400 Men 300

Women

200 100 0 Lost job

Left job

Reentrants

New entrants

There are large differences between men and women in terms of the reason for unemployment.

2.48

26


200 180 160 140 120 100 80 60 40 20 0

Year 1995 Year 2000

Year 2005 Year 2010

The number of prescriptions filled by all stores except independent drug stores has increased substantially.

2.49 40% 35%

30% 25%

20% 15%

Male

10%

Female

5% 0%

There appears to be differences between female and male students in their choice of light beer.

2.50

27


120 100

98 83

80

70 68

64

59 60 46

40

51

50

39

40 25

23

13

20 6 0 C. conservative M conservative Many share

Mixed

M liberal

Some share Don't know

C liberal

3

There are differences among the five groups.

2.51 300 259 236

250 187

200

Fair share 150 100

Too much 122

toolittle 81

No opnion

70 34 41

39

50 18

7

7

0 Conservative

Moderate

Liberal

All three groups say that upper-income people pay too little. However Conservatives are more likely to say fair share than Moderates or Liberals

2.52

28


600 481

500 401 400 300 200 96

94

100 10

12

Democrat

Republican

0

Favor

Oppose

No opinion

Democrats support and Republicans oppose the Affordable Care Act.

2.53 250

200

226

173 139

150 114 100

50

108

99

90

41

35

0 Liberal

Moderate Democrat

Independent

Conservative Republican

No surprise-on social issues Democrats are liberal and Republicans are conservative.

2.54

29


300 264 250 200 159 150

125

133 113

100 66

82

69

50 14 0 Democrat

Independent Liberal

Moderate

Republican Conservative

On economic issues Republicans are very conservative whereas Democrats and Moderates are mixed.

2.55 7 000,0 6 000,0

6 542,6 5 699,4

5 000,0 4 000,0 3 000,0

2 897,7

2 648,2

2 000,0 1 000,0

863,6

556,5

0,0 U.S. U.S. Social U.S. Federal U.S. Civil U.S. Military Individuals Security Trust Reserve Service Retirement and Fund Retirement Fund Institutions Fund

2.56

30

Foreign Nations


1 400,0

1 254,8

1 200,0

1 149,2

1 000,0 800,0 600,0 322,0

400,0

291,4

255,0

232,9

4 214

3 895

225,6

210,6

197,0

188,2

200,0 0,0

2.57 16 000

14 732

14 000 12 000

10 043

10 000 8 000 6 000

7 013

4 000

2 397 1 225

2 000 0

2.58

31

1 648


40 35 30 25 20 15 10 5 0 Married 0 children

Food

Married Couple w children

Housing

One Parent, At Least 1 < 18

Transportation

Healthcare

Insurance & pensions

Other

The pattern is about the same for the three households.

2.59

Don't know/refused 4%

Other reasons 12%

Unemployed/wor k doesn't offer/not eligible at work 11%

Too expensive 47%

Told they were ineligible 7% Immigration status Don't know 7% Don't need it how to get it 6% 3%

Opposed to the ECA/prefer to pay penalty 3%

2.60

32


0,250

0,200 0,150 0,100 0,050 0,000

2013 Uninsured Rate

2014 Uninsured Rate

There are decreases in almost every state. However, there are many Americans without health insurance.

2.61 Strongly agree 4%

Strongly disagree 15%

Agree 23%

Disagree 20%

Neither agree nor disagree 38%

More students disagree than agree.

2.62

33


Very good 10%

Excellent 3%

Poor 15%

Fair 27%

Good 45%

More than 40% rate the food as less than good.

2.63

Manual 18%

Computer 44%

Computer and manual 38%

34


2.64 45% 40%

35% 30% 25% Children

20%

No children 15% 10% 5% 0% Poor

Fair

Good

Very good Excellent

Customers with children rated the restaurant more highly than did customers with no children.

2.65 45 40 35 30 25 20 15 10 5 0

Female Male

Males and females differ in their areas of employment. Females tend to choose accounting marketing/sales and males opt for finance.

b

35


40 35 30 25 20 15 10 5 0

Very satisfied Quite satisfied Little satisfied

Not satisfied

Area and job satisfaction are related. Graduates who work in finance and general management appear to be more satisfied than those in accounting, marketing/sales, and others.

2.66

Males 45% Females 55%

The survey oversampled women slightly.

2.67

36


Others 10% Blacks 15%

Whites 75%

2.68a

Married Widowed Divorced Separated Never married

1158 209 411 81 675

b. Pie chart c.

Never married 27%

Married 46%

Separated 3% Divorced 16%

Widowed 8%

37


2.69 Cpmpleted graduate degree 11% Left high school 13% Completed Bachelor's degree 19%

Completed junior college 7%

Graduated high school 50%

2.70 800 700 600 Left high school

500

High schoo; 400

Junior college

300

Bachelor's degree Graduate

200 100 0 Male

Female

The patterns are similar.

2.71

38


Government 19%

Private sector 81%

2.72 1600

1467

1400 1200 1000 800 600 400

White; 340

273 199

200

Black; 94

Other; 34

0 White

Black

Other

The patterns are similar.

2.73

39


1400 1196 1200 1000

949

800 Self-employed 600

Work for someone else

400 200

0 Male

Female

Males are slightly more likely to be self-employed than females.

2.74 Category 1 10% Category 5 30%

Category 2 13%

Category 3 14%

Category 4 33%

The ”married” categories (4 and 5) make up more than 60% of the households. 2.75

40


2500

2000

No high school

1500

High school Some college

1000

College degree 500

0 Male

Female

There are large differences between male and female heads of households.

2.76 Other 5%

Hispanic 9% Black 12%

White 74%

Whites make up three quarters of the survey.

2.77

41


1800 1600 1400 1200 1000 800 600 400 200 0 White

Black

Hispanic 1

2

Other 3

4

5

There are large differences between the four races in terms of family structure.

2.78 College degree; 2227

2500

2000

1500 High school; 953

1000

646

Some college; 567 463

High school

Some college

No high school; 500 252 294

613

0 No high school

Own

Otherwise

College degree holders are much more likely to own their homes.

42

College degree


Chapter 3 3.1 7 to 9

3.2 11 to 13

3.3 a 9 to 10 (or 11) b Upper limits: 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250

3.4 a 5 (or 6) b Upper limits: 5.2, 5.4, 5.6, 5.8, 6.0, 6.2

Frequency

3.5 14 12 10 8 6 4 2 0 5

8

11 Calls

14

17

3.6

Frequency

20 15 10 5 0 24

28

32

36

40

Ages

3.7

5

44


12 Frequency

10 8 6 4 2 0 70

80

90

100

110

20

25

Balls

3.8

Frequency

15 10 5 0 5

10

15 Calls

The number of calls is bimodal.

3.9

Frequency

30 20 10 0 31

32

33

34

35

36

37

38

39 More

Times

3.10

6

120


Frequency

25 20 15 10 5 0

1

2.5

4

5.5

7

8.5

10

More

Stores

b. The number of stores is bimodal and positively skewed.

3.11

Frequency

20 15 10 5

0 8

12

16

20

24

28

Games

The histogram is positively skewed.

3.12

Frequency

20 15 10

5 0 5

10

15

20

25

30

35

40

Rounds

The histogram is symmetric (approximately) and bimodal.

3.13

7

45


Frequency

80 60 40 20 0

The histogram is symmetric, unimodal, and bell shaped.

Frequency

3.14 70 60 50 40 30 20 10 0 2

4

6

8

10 12 Days

14

16

18

50

60

65

70

Most orders arrive within 12 days.

3.15a.

Frequency

80 60 40

20 0 25

30

35

40

45

55

Customers

b. The histogram is unimodal and positively skewed.

3.16 a.

8

More


25 Frequency

20 15 10 5 0 50

100

150

200

250 300 350 Downloads

400

450 More

b. The histogram is somewhat bell shaped.

Frequency

3.17 100 80 60 40 20 0 37

44

51

58

65

72

79

86

93 100

Marks

The histogram is negatively skewed, bimodal, and not bell shaped.

3.18

Frequency

60

40 20 0 18

19

20

21

22

23

24

Lengths

The histogram is unimodal, bell-shaped and roughly symmetric. Most of the lengths lie between 18 and 23 inches.

3.19

9


Frequency

30

20 10 0 200

400

600

800

1000

1200

1400

Copies

The histogram is unimodal and positively skewed. On most days the number of copies made is between 200 and 1000. On a small percentage of days more than 1000 copies are made.

Frequency

3.20 100 80 60 40 20 0

1

2

3

4

5

6

7

8

Weights

The histogram is unimodal, symmetric and bell-shaped. Most tomatoes weigh between 2 and 7 ounces with a small fraction weighing less than 2 ounces or more than 7 ounces.

3.21

Frequency

150 100 50 0 15

20

25

30

35

40

45

50

55

Gallons

The histogram is positively skewed and unimodal. Most households use between 20 and 45 gallons per day. The center of the distributions appears to be around 25 to 30 gallons. 3.22

10


Frequency

30

20 10 0 12000 15000 18000 21000 24000 27000 30000 Books

The histogram of the number of books shipped daily is negatively skewed. It appears that there is a maximum number that the company can ship.

3.23 600 Frequency

500 400 300 200 100 0 20

30

40

50

60

70

80

90

AGE The histogram is slightly positively skewed with few respondents under 20 or over 80.

3.24 1000 Frequency

800 600 400 200 0 2

4

6

8

10 12 EDUC

14

16

18

20

The histogram is bimodal. The modes represent high school completion (12 years) and university completion (16 years).

11


3.25

Frequency

500 400

300 200 100 0

RINCOME As expected the histogram is positively skewed.

3.26 1000 Frequency

800 600 400 200 0 2

4

6

8

10

12 14 16 TVHOURS

18

20

22

24

Many respondents watched less than 2 hours per day and almost all watched for less than 6 hours.

3.27 600 Frequency

500 400 300 200 100 0 16

20

24

28

32 36 40 44 AGEKDBRN

48

52

Most respondents had their first child born before the age of 24.

12

56

60


3.28

Frequency

1500 1000 500 0 20

30

40

50

60 AGE

70

80

90

100

The histogram is bell shaped with the 50-60 interval was the modal class.

3.29 5000 Frequency

4000 3000 2000 1000 0 1

2

3

4

5

6

7

8

KIDS The maximum number in the sample is 8.

3.30 a. Almost all the values are in the first interval. b. Most of the value are greater than 200,000.

2500 2000 1500 1000 500 0 100000 200000 300000 400000 500000 600000 700000 800000 900000 1000000 1100000 1200000 1300000 1400000 1500000 1600000 1700000 1800000 1900000 2000000 More

Frequency

c.

ASSET

13


This last histogram is best because it allows us to see the assets below $2,000,000.

3.31

Frequency

250 200 150 100 50 0

CDS We ignored the responses with 0 (most respondents reported no CDs.)

3.32 40 35 30 25 20 15 10 5 0

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 Canada

United States

Both countries are winning an increasing number of medals.

3.33a.

14


0,920 0,910 0,900 0,890 0,880 0,870 0,860 0,850 1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829303132

b. 4,50 4,00 3,50

3,00 2,50 2,00 1,50 1,00 0,50 0,00 1 2 3 4 5 6 7 8 9 10 11 1213 1415 1617 18 1920 2122 2324 25 2627 2829 3031 32

The save percentage is increasing and goals against are decreasing.

3.34a.

15


2 500 000

2 000 000

1 500 000

1 000 000

500 000

0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

b. 2 500 000

2 000 000

1 500 000

1 000 000

500 000

0

Both types of crime have been decreasing.

3.35a.

16


Health Care Total ($billions) 1600,000 1400,000 1200,000 1000,000 800,000 600,000 400,000 200,000 0,000 1 3 5 7 9 11131517192123252729313335373941434547495153555759616365

b.

Per capita 5 000 4 500 4 000 3 500 3 000 2 500 2 000 1 500 1 000 500 1 3 5 7 9 1113151719212325272931333537394143454749515355575961636567

c.

17


Per Capita Constant $ 2 000 1 800 1 600 1 400 1 200 1 000 800 600 400 200 1 3 5 7 9 1113151719212325272931333537394143454749515355575961636567

3.36a.

Part A (Hospital Insurance) 250

200

150

100

50

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

b.

18


Per Capita 700 600 500 400 300 200 100 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

c.

Per Capita Constant $ 300 250 200 150 100 50 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

3.37a.

19


Part B (Suppl. Med. Ins.) -$billions 160 140 120 100 80 60 40 20 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

b.

Per Capita 450 400 350 300 250 200 150 100 50 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

c.

20


Per Capita Constant $ 200 180 160 140 120 100 80 60 40 20 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

3.38a.

Medicaid ($billions) 500,00 450,00 400,00 350,00 300,00 250,00 200,00 150,00 100,00 50,00 0,00 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53

b.

21


Per Capita 1600 1400 1200 1000 800 600 400 200 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53

c.

Per Capita Constant $ 700 600 500 400 300 200 100 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53

3.39a.

22


Disability Ins (DI) -fed 160,00 140,00 120,00 100,00 80,00 60,00 40,00 20,00 0,00 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

b.

Per Capita 500 450 400 350 300 250 200 150 100 50 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

c.

23


Per Capita Constant $ 250

200

150

100

50

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

3.40a.

Old Age Survivor Ins (OASI)-fed $ billion 800 700 600 500 400 300 200 100 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

b.

24


Per Capita 2500

2000

1500

1000

500

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

c.

Per Capita Constant $ 1200 1000 800 600 400 200 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57

3.41a.

25


Defense ($billions) 1000 900 800 700 600 500 400 300 200 100 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

b.

Per Capita 3000 2500 2000 1500 1000 500 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

c.

26


Per Capita Constant $ 4000 3500 3000 2500 2000 1500 1000 500 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

3.42a.

Total spending ($billions) 7000 6000 5000 4000 3000 2000

1000 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

b.

27


Per Capita 25000

20000

15000

10000

5000

0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

c.

Per Capita Constant $ 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

3.43a.

28


Welfare ($Billions) 600 500 400 300 200 100 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

b.

Per Capita 1800 1600 1400 1200 1000 800 600 400 200 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

c.

29


Per Capita Constant $ 800 700 600 500 400 300 200 100 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49

3.44a.

Education ($billions) 1200 1000 800 600 400 200 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

b.

30


Per Capita 3500 3000 2500 2000 1500 1000 500 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

c.

Per Capita Constant $ 1600 1400 1200 1000 800 600 400 200 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

3.45a.

31


Interest Payments ($billions) 400 350 300 250 200 150 100 50 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

b.

Per Capita 1400 1200 1000 800 600 400 200 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

c.

32


Per Capita Constant $ 800 700 600 500 400 300 200 100 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

3.46b.

Violent Crime Per 100,000 800,0 700,0 600,0 500,0 400,0 300,0 200,0 100,0 0,0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

c.

33


Property Crime per 100,000 5 000 4 500 4 000 3 500 3 000 2 500 2 000 1 500 1 000 500 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20

d. In the period 1993 to 2012 crime was decreasing no matter how it is measured.

3.47 25 000

20 000

15 000

10 000

5 000

0 1

3

5

7

9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 Consumption

Production

The gap is closing.

3.48a.

34


GDP ($billions) 20 000,0 18 000,0 16 000,0 14 000,0 12 000,0 10 000,0 8 000,0 6 000,0 4 000,0 2 000,0 0,0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

b.

GDP Constant $billions 8000 7000 6000 5000 4000 3000 2000 1000 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

c. GDP has been steadily increasing.

3.49a.

35


1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

U.S. Exports to Canada

30000

25000

20000

15000

10000

5000

0

b.

U.S. Imports from Canada

35000

30000

25000

20000

15000

10000

5000

0

c.

36


Trade Balamce 2000

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0 -2000 -4000 -6000 -8000 -10000 -12000

d. In recent years the balance has become equal.

3.50a.

U.S. Exports to Japan 7000 6000 5000 4000 3000 2000 1000

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0

b.

37


U.S. Imports from Japan 16000 14000 12000 10000

8000 6000 4000 2000 1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0

c.

Trade Balance -1000

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0

-2000 -3000 -4000 -5000 -6000 -7000 -8000 -9000

d. The U.S. imports more from Japan than Japan imports from the U.S.

3.51a.

38


1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

U.S. Exports to China

14000

12000

10000

8000

6000

4000

2000

0

b.

U.S. Imports from China

50000

45000

40000

35000

30000

25000

20000

15000

10000

5000

0

c.

39


Trade Balamce 5000

-5000

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0

-10000 -15000 -20000

-25000 -30000 -35000 -40000

d. The U.S. imports far more from China than China imports from the U.S.

3.52

Canadian Dollars to One U.S. Dollar 1,8000 1,6000 1,4000 1,2000 1,0000 0,8000 0,6000 0,4000 0,2000 1 19 37 55 73 91 109 127 145 163 181 199 217 235 253 271 289 307 325 343 361 379 397 415 433 451 469 487 505 523 541

0,0000

The recent trend reveals that the Canadian dollar has been losing value relative to the U.S. dollar.

3.53

40


Japanese Yen to One U.S. Dollar 400,00 350,00 300,00 250,00

200,00 150,00 100,00 50,00 1 19 37 55 73 91 109 127 145 163 181 199 217 235 253 271 289 307 325 343 361 379 397 415 433 451 469 487 505 523 541

0,00

The value of the Japanese yen has been increasing relative to the U.S. dollar.

3.54

DJ Industrial Average Close 20 000,00

18 000,00 16 000,00 14 000,00 12 000,00 10 000,00 8 000,00 6 000,00 4 000,00 2 000,00 1 29 57 85 113 141 169 197 225 253 281 309 337 365 393 421 449 477 505 533 561 589 617 645 673 701 729 757 785

0,00

3.55

41


DJA Constant $ 9000 8000 7000 6000 5000 4000 3000 2000 1000 1 27 53 79 105 131 157 183 209 235 261 287 313 339 365 391 417 443 469 495 521 547 573 599 625 651 677 703 729 755 781

0

The stock index has increased whether measured in actual or constant dollars.

3.56 450 400 350 Savings

300 250 200 150 100 50 0 0,0

1,0

2,0

3,0

4,0 Time

5,0

There is a weak positive linear relationship.

3.57a

42

6,0

7,0

8,0


40 30

Returns

20 10 0

0.0

1.0

2.0

3.0

4.0

5.0

6.0

80

90

100

-10 -20

Inflation

b. There is very weak linear relationship.

3.58a 90 85

Statistics

80 75 70 65 60 40

50

60

70 Calculus

b. There is a positive linear relationship between calculus and statistics marks.

3.59

43


2500

2000

Cost

1500 1000 500 0 0

5

10

15

20

25

Speed

There is a strong positive linear relationship.

3.60 25

Publications

20 15

10 5 0 0

10

20

30

40

50

Age

There is a negative linear relationship.

3.61

44

60

70

80


100 95

90

Mark

85 80 75 70 65 60 55 50 60

65

70

75

80

85

90

95

Time

There is no linear relationship

3.62a 1600

1400

Electricity

1200 1000 800 600

400 200 0 0

1

2

3

4

5

6

Occupants

b. There is a moderately strong positive linear relationship.

3.63

45

7

8

9


700 600 500

Tickets

400 300 200 100 0

-40

-30

-20

-10

0

10

20

30

40

Temperature

There is a moderately strong linear relationship. More lift tickets are sold during warmer days.

3.64a 90 80

70 Income

60 50

40 30 20

10 0 60

65

70

75

Height

b. There is a very weak positive linear relationship.

3.65

46

80


2000000

Compensation

1500000

1000000

500000

0 0

5000

10000

-500000

15000

20000

25000

30000

20000

25000

30000

Profit

500 450

Compensation

400 350 300 250 200 150 100 50 0 0

5000

10000

15000

3-Yr share return

There does not appear to be a linear relationship between compensation of profit and between compensation and 3year share return.

3.66

47


600 500

Tenure

400 300 200 100 0 0

20

40

60

80

100

Age

There is a moderately strong positive linear relationship.

3.67 3000 2500

Sales

2000 1500 1000 500 0 0

5

10

15

20 Time

There is no linear relationship.

3.68

48

25

30

35


400

Livestock sub-index

350 300 250 200 150

100 50 0 0

100

200

300

400

500

600

8

10

12

Grains sub-index

There is moderately strong positive linear relationship.

3.69

Bank prime loan rate

25

20 15 10 5 0 0

2

4

6 Unemployment rate

There is a weak positive linear relationship.

3.70

49


6000 5000

Price

4000 3000

2000 1000 0 0

10

20

30

40

50

60

70

Grade

There is a strong nonlinear positive relationship.

3.71

222 220 218 216 214 212 210 208

0

20

40

60

80

There is a positive linear relationship.

3.72

50

100

120


25

SPEDUC

20

15

10

5

0 0

5

10

15

20

25

15

20

25

EDUC There is a positive linear relationship.

3.73 25

EDUC

20

15

10

5

0 0

5

10

PAEDUC

There is a weak positive linear relationship.

3.74

51


25

EDUC

20

15

10

5

0 0

5

10

15

20

25

MAEDUC

There is a weak positive linear relationship.

3.75 100 90 80 70 60 50 40 30 20 10 0 0

10

20

30

40

50

60

70

There is no relationship.

3.76

52

80

90

100


30

25 20 15

10 5 0 0

10

20

30

40

50

60

70

80

90

100

There is no relationship.

3.77 30 25 20 15 10 5 0 0

5

10

15

20

There is a weak positive linear relationship.

3.78

53

25


60

50 40 30

20 10 0 0

5

10

15

20

25

There is a weak positive linear relationship.

3.79 1 200 000,00 1 000 000,00 800 000,00 600 000,00 400 000,00 200 000,00 0,00 0

10

20

30

40

50

60

There is a negative linear relationship.

3.80

54

70

80

90

100


18 16 14 12 10 8 6 4 2 0 0

10

20

30

40

50

60

70

80

90

100

80

90

100

-2

There is a negative linear relationship.

3.81 900 000,00 800 000,00 700 000,00 600 000,00 500 000,00 400 000,00 300 000,00 200 000,00 100 000,00 0,00 0

10

20

30

40

50

60

70

There is a negative linear relationship.

3.82 a We convert the numbers to accident rate and fatal accident rate.

55


1 2 3 4 5 6 7 8 9

A Age group Under 20 20-24 25-34 35-44 45-54 55-64 65-74 0ver 74

B Accident rate per driver 0.373 0.173 0.209 0.162 0.133 0.108 0.095 0.093

C Fatal accidient rate (per 1,000 drivers) 0.643 0.352 0.305 0.251 0.214 0.208 0.177 0.304

b 0.7

0.6 0.5

0.4

Accidents/driver

0.3 Fatal accidents/1000 drivers

0.2

0.1 0

Under 20-24 25-34 35-44 45-54 55-64 65-74 0ver 20 74

c. The accident rate generally decreases as the ages increase. The fatal accident rate decreases until the over 64 age category where there is an increase.

3.83

56


2.500 2.000 1.500 1.000

Injury rate Fatal injury rate per 100

0.500 0.000

c Older drivers who are in accidents are more likely to be killed or injured. d Exercise 3.9 addressed the issue of accident rates, whereas in this exercise we consider the severity of the accidents.

3.84a 600 550 500 450 400

Verbal All

350

Math All

300 250 200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

57


b 550 540

530 520

Verbal All Math All

510

500 490 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

c 560 540 520 500 480

Verbal Male

460

Verbal Female

440 420 400 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

58


600 500

400 300

Math Male Math Female

200

100 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

d 550

540

530

520

Verbal Male Verbal Female

510

500

490 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

59


550 540 530 520 510 Math Male 500

Math Female

490 480 470 460 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

3.85a 7.8 7.6

7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 1

2

3

4

5

6

7

8

b

60

9

10

11

12


8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1

2

3

4

5

6

7

8

9

10

11

12

c Caption a: Unemployment rate falling rapidly. Caption b: Unemployment rate virtually unchanged. d The chart in (a) is more honest.

3.86 40%

Return on sub-index

30% 20% 10% 0% -10%

-20% -30% -40% -1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

Inflation rate

There is no linear relationship between the inflation rate and the return on the precious metals sub-index.

61


3.87 1,6 1,4 1,2

1 0,8 0,6 0,4 0,2

1 19 37 55 73 91 109 127 145 163 181 199 217 235 253 271 289 307 325 343 361 379 397 415 433 451 469 487 505 523 541

0

There has been a long-term decline in the value of the Australian dollar.

3.88 160

140

IQ Twin 2

120 100 80 60

40 20 0 0

20

40

60

80

100

IQ Twin 1

There is a strong positive linear relationship.

62

120

140


3.89 180 160

Currencies index

140 120 100

80 60 40

20 0 60

70

80

90

100

110

120

130

Interest rate index

There is a very strong positive linear relationship.

3.90

Frequency

80 60 40

20 0 2

4

6

8 10 12 Spam Emails

14

The histogram is symmetric and bell shaped.

3.91a

63

16

18

140


74 72

Son's height

70 68 66 64 62 60 62

64

66

68

70

72

74

76

Father's height

b. The slope is positive c. There is a moderately strong linear relationship.

3.92 12 10 8 6 4 2 0 0,00 2 000,004 000,006 000,008 000,0010 000,00 12 000,00 14 000,00 16 000,00 18 000,00 20 000,00

There is no linear relationship between the Dow Jones Industrial average and the unemployment rate.

3.93

64


3 2,5 2

1,5 1 0,5

1 19 37 55 73 91 109 127 145 163 181 199 217 235 253 271 289 307 325 343 361 379 397 415 433 451 469 487 505 523 541

0

The value of the British pound has fluctuated quite a bit has stayed at current exchange rate for many months.

3.94 350 330

310 290 Time

270 250 230 210 190 170 150 300

320

340

360

380

400

420

Score

There is a strong positive linear relationship. Poorer players take longer to complete their rounds.

3.95a.

65


1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

U.S. Exports to Mexico

25 000,0

20 000,0

15 000,0

10 000,0

5 000,0

0,0

b.

U.S. Imports from Mexico

30 000,0

25 000,0

20 000,0

15 000,0

10 000,0

5 000,0

0,0

c.

66


Trade Balance 2 000,0 1 000,0

-1 000,0

1 14 27 40 53 66 79 92 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 365

0,0

-2 000,0 -3 000,0 -4 000,0 -5 000,0 -6 000,0 -7 000,0 -8 000,0

Mexico exports more to the U.S. than the U.S. exports to Mexico.

3.96 1400 1200

Credit

1000 800 600 400 200 0 0

100

200

300 Debit

There is a moderately strong negative linear relationship.

3.97

67

400

500


Frequency

40 30 20 10

0 25000

40000

55000

70000

85000 100000

Pay

Frequency

3.98 50 40 30 20 10 0

7

10

13

16

19

22

25

Meetings

The histogram of the number of meetings is positively skewed.

3.99 45 40 35 30 25 20 15 10 5 0 1 3 5 7 9 1113 1517 1921 2325 2729 3133 3537 3941 4345 4749 51 535557 59 6163 65

The number of fatal accidents has been steadily decreasing.

3.100 68


Frequency

200 150 100

50 0 10

20

30

40

50

60

70

80

90 100 110

Miles

The histogram is positively skewed and bimodal.

3.101

Frequency

200 150 100

50 0 30

45

60

75

90 105 120 135 150 165 Times

The histogram tells us that about 70% of gallery visitors stay for 60 minutes or less and most of the remainder leave within 120 minutes. Although there are other plans, the gallery director proposed the following plan. Admit 200 visitors every hour. We expect that about 140 will leave within 1 hour and about 60 will stay for an additional hour. During the next 1-hour period, 200 new visitors will be admitted. If 60 of the previous hour’s admittances remain, there will be a total of 260 people in the gallery. If this pattern persists during the day, there will be a maximum of 260 visitors at any time. This plan should permit as many people as possible to see the exhibit and yet maintain comfort and safety.

3.102 Business Statistics course (Example 3.3)

69


100 90

Statistics

80 70 60 50 40 25

50

75

100

75

100

Calculus

Mathematical Statistics course (Example 3.4) 100 90

Statistics

80 70 60 50 40 30 25

50 Calculus

There appears to be a stronger linear relationship between marks in the mathematical statistics course and calculus than the relationship between the marks in the business statistics course and the marks in calculus.

3.103 Business Statistics course (Example 3.3)

70


100 90

Final

80 70 60 50 40 50

60

70

80

90

100

80

90

100

Midterm

Mathematical Statistics course (Example 3.4) 100 90

Final

80 70 60 50 40 30 50

60

70 Midterm

The relationship between midterm marks and final marks appear to be similar for both statistics courses. That is, there is weak positive linear relationship.

3.104a.

71


Frequency

80 60 40

20 0 30

34

38

42

46

50

54

58

62

66

Salary

The histogram is approximately bell shaped and symmetric.

b 70 60

Salary

50 40 30 20 10 0 0

1

2

3

4

5

6

7

Search

There is no linear relationship between the amount of time needed to land a job and salary.

3.105

72


Federal Government Debt 20 000 000 000 000 18 000 000 000 000 16 000 000 000 000 14 000 000 000 000 12 000 000 000 000

10 000 000 000 000 8 000 000 000 000 6 000 000 000 000 4 000 000 000 000 2 000 000 000 000 1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172 181 190 199 208 217 226

0

The debt was quite manageable until 100 years ago. It is quite unmanageable now.

3.106

Per Capita Debt 60000 50000 40000 30000 20000 10000 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

Per capita debt is still unmanageable.

3.107

73


Per Capita Debt Constant $ 30000 25000 20000 15000 10000 5000 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79

This doesn’t look as bad.

3.108

National Debt 700 000 000 000 600 000 000 000 500 000 000 000 400 000 000 000 300 000 000 000 200 000 000 000 100 000 000 000 0 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89

3.109

74


Per Capita Debt 20000 18000 16000 14000 12000

10000 8000 6000 4000 2000 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136 141 146

0

3.110

Per Capita Debt Constant $ 25000 20000 15000 10000 5000

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101

0

75


Case 3.1 Line Chart of Temperature Anomalies 2.5 2 1.5 1 0.5 0 -0.5 -1 -1.5 -2 -2.5 1

121

241

361

481

601

721

841

961

1081

1201

1321

1441

There is a clear upward trend of about 1 degree Celsius over the 130 years.

Scatter Diagram of National Climate Data Center Land and Sea Temperature Anomalies and CO2 Levels (19582009) 2

Temperature anomalies

1.5 1 0.5 0 -0.5 -1 -1.5 300

320

340

360

380

400

CO2

There is a moderately strong positive linear relationship between carbon dioxide levels and temperature anomalies.

76


Case 3.4 90,000 80,000

70,000

GDP

60,000 50,000

40,000 30,000 20,000

10,000 0 0

20

40

60

80

100

Freedom scores

There is a moderately strong positive linear relationship between freedom scores and gross domestic product.

77


Chapter 4 4.1 a x 

 x  52  25  15  0  104  44  60  30  33  81  40  5 = 489 = 40.75 i

n

12

12

Ordered data: 0, 5, 15, 25, 30, 33, 40, 44, 52, 60, 81, 104; Median = (33 + 40)/2 = 36.5 Mode = all

4.2 x 

 x  5  7  0  3  15  6  5  9  3  8  10  5  2  0  12 = 90 = 6.0 i

n

15

15

Ordered data: 0, 0, 2, 3, 3, 5, 5, 5, 6, 7, 8, 9, 10, 12, 15; Median = 5 Mode = 5

4.3 a x 

 x  5.5  7.2  1.6  22. 0  8.7  2.8  5.3  3.4  12.5  18.6  8.3  6.6 = 102 .5 = 8.54 i

n

12

12

Ordered data: 1.6, 2.8, 3.4, 5.3, 5.5, 6.6, 7.2, 8.3, 8.7, 12.5, 18.6, 22.0; Median = 6.9 Mode = all b The mean number of miles jogged is 8.54. Half the sample jogged more than 6.9 miles and half jogged less.

4.4 a x 

 x  33  29  45  60  42  19  52  38  36 = 354 = 39.33 i

n

9

9

Ordered data: 19, 29, 33, 36, 38, 42, 45, 52, 60; Median = 38 Mode: all b The mean amount of time is 39.33 minutes. Half the group took less than 38 minutes.

4.5 a x 

 x  14  8  3  2  6  4  9  13  10  12  7  4  9  13  15  8  11  12  4  0

=

i

n

20

164 = 8.2 20

Ordered data: 0, 2, 3, 4, 4, 4, 6, 7, 8, 8, 9, 9, 10, 11, 12, 12, 13, 13, 14, 15; Median = 8.5 Mode = 4 b The mean number of days to submit grades is 8.2, the median is 8.5, and the mode is 4.

4.6 R g  3 (1  R1)(1  R 2 )(1  R3 )  1 = 3 (1  .25 )(1  .10 )(1  .50 )  1 = .19 4.7 R g  4 (1  R1)(1  R 2 )(1  R3 )(1  R 4 )  1 = 4 (1  .50 )(1  .30 )(1  .50 )(1  .25)  1 = –.075

97


4.8 a x 

 x  .10  .22  .06  .05  .20 = .53 = .106 i

n

5

5

Ordered data: –.05, .06, .10, .20, .22; Median = .10 b R g  5 (1  R1)(1  R 2 )(1  R3 )(1  R 4 )(1  R5 )  1 = 5 (1  .10 )(1  .22 )(1  .06 )(1  .05)(1  .20 )  1 = .102 c The geometric mean is best.

4.9 a x 

 x  - .15 - .20  .15  .08  .50 = .22 = .044 i

n

5

5

Ordered data: –.20, –.15, –.08, .15, .50; Median = –.08 b R g  5 (1  R1)(1  R 2 )(1  R3 )(1  R 4 )(1  R5 )  1 = 5 (1  .15)(1  .20 )(1  .15)(1  .08 )(1  .50 )  1 = .015 c The geometric mean is best.

4.10 a Year 1 rate of return =

1200  1000 = .20 1000

Year 2 rate of return =

1200  1200 =0 1200

Year 3 rate of return =

1500  1200 = .25 1200

Year 4 rate of return =

2000  1500 = .33 1500

b x

 x  .20  0  .25  .33 = .78 = .195 i

n

4

4

Ordered data: 0, .20, .25, .33; Median = .225 c R g  4 (1  R1)(1  R 2 )(1  R3 )(1  R 4 )  1 = 4 (1  .20 )(1  0)(1  .25 )(1  .33)  1 = .188 d The geometric mean is best because 1000(1.188) 4 = 2000.

4.11 a Year 1 rate of return =

10  12 = –.167 12

Year 2 rate of return =

14  10 = .40 10

Year 3 rate of return =

15  14 = .071 14

Year 4 rate of return =

22  15 = .467 15

98


Year 5 rate of return =

30  22 = .364 22

Year 6 rate of return =

25  30 = –.167 30

b x

 x  - .167  .40  .071  .467  .364  .167 = .968 = .161 i

n

6

6

Ordered data: –.167, –.167, .071, .364, .40, .467; Median = .218 c R g  6 (1  R 1 )(1  R 2 )(1  R 3 )(1  R 4 )(1  R 5 )(1  R 6 )  1 = 6 (1  .167 )(1  .40 )(1  .071)(1  .467 )(1  .364 )(1  .167 )  1 = .130 d The geometric mean is best because 12(1.130)6 = 25.

4.12 x = 54.91; median = 55. The mean number of bidders 54.91. Half of the auctions had 55 or fewer bidders. 4.13 x = 65,784; median = 65,805. The mean starting salary is $65,684 and half the sample’s starting salaries were less than or equal to $65,805.

4.14a x = 45.60; median = 45 b The mean commuting time is 45.60 minutes and half the sample took 45 minutes or less.

4.15 x = 26.56; median = 27. The mean commute is 26.56 minutes and half the sample commutes for 27 or less minutes.

4.16a x = .81; median = .84 b The mean percentage is .81. Half the sample paid less than .84. 4.17a x = 32.91; median = 32; mode = 32 b The mean speed is 32.91 mph. Half the sample traveled slower than 32 mph and half traveled faster. The mode is 32.

4.18a x = 122.76; median = 124 b The mean age is 122.76 months. Half the sample was 124 months or less. 4.19 x = 49.01; median = 49 4.20 x = 13.7; median = 14

99


4.21 x = 2.98; median = 2 4.22 x = 45,247, median = 32,500 4.23 x = 51.75; median = 52 4.24 x = 853,135; median = 60,872 4.25 x = 9,282,946; median = 249,000 4.26 x = 264,572; median = 23,690

4.27 x 

 x  9  3  7  4  1  7  5  4 = 40 = 5 i

n

8

 (x  x)  [(9  5)  (3  5)  ...  (4  5) = 46 = 6.57 2

s2 

i

2

2

2

8 1

n 1

4.28 x 

7

 x  4  5  3  6  5  6  5  6 = 40 = 5 i

n

8

8

 (x  x)  [(4  5)  (5  5)  ...  (6  5) = 8 = 1.14 2

s2 

8

4.29 x 

i

2

2

2

8 1

n 1

7

 x  12  6  22  31  23  13  15  17  21 = 160 = 17.78 i

n

9

9

 (x  x)  [(12  17.78)  (6  17.78)  ...  (21  17.78) = 433 .56 = 54.19 s  2

2

i

n 1

2

2

2

9 1

8

s  s 2 = 54.19 = 7.36

100


4.30 x 

 x  0  (5)  (3)  6  4  (4)  1  (5)  0  3 = 3 = –.30 i

n

 (x  x)  [(0  (.3))  ((5)  (.3))  ...  (3  (.3)) = 136 .1 = 15.12 2

s2 

10

10

i

2

n 1

2

2

10  1

9

s  s 2 = 15.12 = 3.89

4.31 The data in (b) appear to be most similar to one another.

4.32 a: s 2 = 51.5 b: s 2 = 6.5 c: s 2 = 174.5 4.33 Variance cannot be negative because it is the sum of squared differences.

4.34 6, 6, 6, 6, 6

4.35 a. About 68% b. About 95% c. About 99.7%

4.36 a From the empirical rule we know that approximately 68% of the observations fall between 46 and 54. Thus 16% are less than 46 (the other 16% are above 54). b. Approximately 95% of the observations are between 42 and 58. Thus, only 2.5% are above 58 and all the rest, 97.5% are below 58. c. See (a) above; 16% are above 54.

4.37 a at least 75% b at least 88.9%

4.38 a Nothing b At least 75% lie between 60 and 180. c At least 88.9% lie between 30 and 210.

101


4.39 Range = 25.85, s 2  29.46, and s = 5.43; there is considerable variation between prices; at least 75% of the prices lie within 10.86 of the mean; at least 88.9% of the prices lie within 16.29 of the mean.

4.40 s 2  40.73 mph 2 and s = 6.38 mph; at least 75% of the speeds lie within 12.76 mph of the mean; at least 88.9% of the speeds lie within 19.14 mph of the mean

4.41 a Punter

Variance

Standard deviation

1

40.22

6.34

2

14.81

3.85

3

3.63

1.91

b Punter 3 is the most consistent.

4.42 s 2  .0858 cm 2 , and s = .2929cm; at least 75% of the lengths lie within .5858 of the mean; at least 88.9% of the rods will lie within .8787 cm of the mean. 4.43 x  175.73 and s = 62.1; At least 75% of the withdrawals lie within $124.20 of the mean; at least 88.9% of the withdrawals lie within $186.30 of the mean..

4.44a s = 15.01 b In approximately 68% of the days the number of arrivals falls within 15.01 of the mean; in approximately 95% of the hours the number of arrivals falls within 30.02 of the mean; in approximately 99.7% of the hours the number of arrivals falls within 45.03 of the mean 4.45 x  26.02 and s = 11.81 4.46 x  95.09 and s = 7.51 4.47 x  1937 and s = 950.0 4.48 x  749.7 and s = 215.9 4.49 x  49.01, s2 = 303.2, and s = 17.41 4.50 x  13.70 and s = 3.07 4.51 x  2.98 and s = 2.59

102


4.52 x  45,247 and s = 39,885 4.53 x  51.75 and s = 16.17 4.54 x  853,135 and s = 5,952,380 4.55 x  9,282,946 and s = 54,480,433 4.56 x  264,572 and s = 3,793,046

4.57 First quartile: L25  (10  1) Second quartile: L50  (10  1) Third quartile: L75  (10  1)

25 = (11)(.25) = 2.75; the first quartile is 5.5. 100

50 = (11)(.5) = 5.5; the second quartile is 11. 100

75 = (11)(.75) = 8.25; the third quartile is 16.5. 100

4.58 First quartile: L25  (15  1) Second quartile: L50  (15  1) Third quartile: L75  (15  1)

25 = (16)(.25) = 4; the fourth number is 3. 100

50 = (16)(.5) = 8; the eighth number is 5. 100

75 = (16)(.75) = 12; the twelfth number is 7. 100

4.59 30th percentile: L30  (10  1) 80th percentile: L80  (10  1)

80 = (11)(.80) = 8.8; the 80th percentile 30.8. 100

4.60 20th percentile: L 20  (10  1) 40th percentile: L 40  (10  1)

30 = (11)(.30) = 3.3; the 30th percentile is 22.3. 100

20 = (11)(.20) = 2.2; the 20th percentile is 43 + .2(51–43) = 44.6. 100

40 = (11)(.40) = 4.4; the 40th percentile is 52 +.4(60–52) = 55.2. 100

4.61 First quartile: L25  (13  1) Second quartile: L50  (13  1)

25 = (14)(.25) = 3.5; the first quartile is 13.05. 100

50 = (14)(.5) = 7; the second quartile is 14.7. 100 103


Third quartile: L75  (13  1)

75 = (14)(.75) = 10.5; the third quartile is 15.6. 100

4.62 Third decile: L 30  (15  1) Sixth decile: L60  (15  1)

30 = (16)(.30) = 4.8; the third decile is 5 + .8(7 – 5) = 6.6. 100

60 = (16)(.60) = 9.6; the sixth decile is 17 + .6(18 – 17) = 17.6. 100

4.63 Interquartile range = 15.6 –13.05 = 2.55

4.64 First quartile: L25  (15  1) Third quartile: L75  (15  1)

25 = (16)(.25) = 4; the fourth number is 13. 100

75 = (16)(.75) = 12; the twelfth number is 24. 100

Interquartile range = 24 – 13 = 11 4.65 First quartile = 5.75, third quartile = 15; interquartile range = 15 – 5.75 = 9.25

4.66 First quartile: L25  (15  1) Third quartile: L75  (15  1)

25 = (16)(.25) = 4; the fourth number is 3. 100

75 = (16)(.75) = 12; the twelfth number is 21. 100

Interquartile range = 21 –3 = 18 4.67 L85 = 75; The speed limit should be set at 75 mph. 4.68 a First quartile = 2, second quartile = 4, and third quartile = 8. b Most executives spend little time reading resumes. Keep it short.

4.69 Dogs: First quartile = 1097.5, second quartile = 1204, and third quartile = 1337. Cats: First quartile = 743, second quartile = 856, and third quartile = 988. Dogs cost more money than cats. Both sets of expenses are positively skewed.

4.70 First quartile = 50, second quartile = 125, and third quartile = 260. The amounts are positively skewed.

4.71 BA First quartile = 25,730, second quartile = 27,765, and third quartile = 29836 BSc First quartile = 29,927, second quartile = 33,397, and third quartile = 36,745

104


BBA First quartile = 31,316, second quartile = 34,284, and third quartile = 39,551 Other First quartile = 28,254, second quartile = 29,951, and third quartile = 32,905 The starting salaries of BA and other are the lowest and least variable. Starting salaries for BBA and BSc are higher.

4.72 The quartiles are 145.51, 164.17, and 174.64. The data are positively skewed. One-quarter of the times are below 145.51 and one-quarter are above 174.64.

4.73 Private course: The quartiles are 228, 237, and 245 Public course: The quartiles are 279, 296, and 307 The amount of time taken to complete rounds on the public course are larger and more variable than those played on private courses.

4.74a. The quartiles are 26, 28.5, and 32 b. The times are positively skewed.

4.75 The quartiles are 8081.81, 9890.48, and 11,692.92. One-quarter of mortgage payments are less than $607.19 and one quarter exceed $909.38. 4.76 The quartiles are 2377, 2765, and 3214. One-quarter of the amounts are less than $2377 and one quarter exceed $3214. 4.77 The quartiles are 16,250, 32,500, and 55,000. One-quarter of the amounts are less than $16,500 and one quarter exceed $55,000.

4.78 The quartiles are 12, 14, and 16. One-quarter of the amounts are less than or equal to 12 and one quarter are greater than or equal to 16. 4.79 The quartiles are 1, 2, and 4. One-quarter of the amounts are less than or equal to 1 and one quarter are greater than or equal to 4.

4.80 The quartiles are 28,407, 60,872, and 144,063. One-quarter of the amounts are less than $28,407 and one quarter are greater $144,063.

4.81 The quartiles are 32,100, 246,920, and 1,139,010. One-quarter of the amounts are less than $32,100 and one quarter are greater than $1,139,010.

4.82 The quartiles are 0, 23,690, and 156,100. One-quarter of the amounts are less than or equal to 0 and one quarter are greater than $156,100.

105


4.83 There is a negative linear relationship. The strength is unknown.

4.84 a. r 

s xy sxsy

 150  .7813 (16 )(12 )

There is a moderately strong negative linear relationship. b. R2 = r2 = (− .7813)2 = .6104 61.04% of the variation in y is explained by the variation in x.

4.85a.

Total

xi

yi

xi2

yi2

x i yi

20

14

400

196

280

40

16

1600

256

640

60

18

3600

324

1080

50

17

2500

289

850

50

18

2500

324

900

55

18

3025

324

990

60

18

3600

324

1080

70

20

4900

400

1400

405

139

22,125

2,437

7,220

n

n

x i = 405

i 1

 i 1

n

n

n

y i = 139

x i2 = 22,125

i 1

y i2 = 2,437

i 1

 x y = 7,220 i

i 1

n n   xi yi   n (405 )(139 )  1  1   i 1 7,220  s xy  x i y i  i 1 =     26 .16  8 1  8 n  1  i 1 n      

 

2   n      xi      n 2  1  x i2   i 1   = 1 22,125  (405 )   231 .7 s 2x  n  1  i 1 n  8  1  8       

106

i


s x  s 2x  231 .7  15 .22 2   n      yi  n     2  1  y i2   i 1   = 1 2,437  (139 )   3.13 s 2y  n n  1  i 1  8  1  8       

s y  s 2y  3.13  1.77 r

s xy sxsy

26 .16

 ..9711

(15 .22 )(1.77 )

R2 = r2 = .97112 = .9430 The covariance is 26.16, the coefficient of correlation is .9711 and the coefficient of determination is .9430. 94.30% of the variation in expenses is explained by the variation in total sales. b.

b1 

s xy

=

s 2x

26 .16  .113 231 .7

x

 x  405  50.63

y

 y  139  17.38

i

n

8

i

n

8

b 0  y  b1x = 17.38 – (.113)(50.63) = 11.66 The least squares line is

ŷ = 11.66 + .113x The estimated variable cost is .113 and the estimated fixed cost is 11.66.

4.86

xi

yi

xi2

yi2

x i yi

40 42 37 47 25 44 41 48 35 28

77 63 79 86 51 78 83 90 65 47

1,600 1,764 1,369 2,209 625 1,936 1,681 2,304 1,225 784

5,929 3,969 6,241 7,396 2,601 6,084 6,889 8,100 4,225 2,209

3,080 2,646 2,923 4,041 1,276 3,432 3,403 4,320 2,275 1,316

107


Total

387

719

n

n

n

i 1

28,712

n

n

2 i

2 i

i

i

53,643

 x = 15,497  y = 53,643  x y = 28,712

 y = 719

 x = 387 a

15,497

i 1

i 1

i 1

i

i

i 1

n n   xi yi   n (387 )( 719 )  1  1   i 1 28,712  s xy  x i y i  i 1 =     98 .52  10  1  10 n n  1  i 1      

 

2   n     xi   n    1  (387 ) 2  1  i 1    2 2  xi  sx  = 15,497    57 .79 n  1  i 1 n  10  1  10       

2   n     yi   n    2  1   y i2   i 1   = 1 53,643  (719 )   216 .32 s 2y  n  1  i 1 n  10  1  10       

b

r

s xy sxsy

98 .52

 .8811

(57 .79 )( 216 .32 )

c

R2 = r2 = .88112 = .7763

d

b1 

s xy

=

s 2x

98 .52  1.705 57 .79

x

 x  387  38.7

y

 y  719  71.9

i

n

10

i

n

10

b 0  y  b1x = 71.9 – (1.705)(38.7) = 5.917 The least squares line is

ŷ = 5.917 + 1.705x e. There is a strong positive linear relationship between marks and study time. For each additional hour of study time marks increased on average by 1.705.

4.87

xi

yi

xi2

yi2

x i yi

599 689

9.6 8.8

358,801 474,721

92.16 77.44

5750.4 6063.2

108


Total

584 631 594 643 656 594 710 611 593 683 7,587

7.4 10.0 7.8 9.2 9.6 8.4 11.2 7.6 8.8 8.0 106.4

n

n

x i =7,587

341,056 398,161 352,836 413,449 430,336 352,836 504,100 373,321 351,649 466,489 4,817,755

x i2 = 4,817,755

i 1

4321.6 6310.0 4632.2 5915.6 6297.6 4989.6 7952.0 4643.6 5218.4 5464.0 67,559.2 n

n

n

y i = 106.4

i 1

i 1

54.76 100.00 60.84 84.64 92.16 70.56 125.44 57.76 77.44 64.00 957.2 y i2 = 957.2

i 1

 x y = 67,559.2 i

i

i 1

n n   xi yi   n (7,587 )(106 .4)  1  1   i 1 67 ,559 .2  s xy  x i y i  i 1 =     26 .16  12  1  12 n n  1  i 1      

 

2   n      xi  (7,587 ) 2  1      n 4 , 817 , 755  1    1,897 .7  x i2   i 1   = 12  1  s 2x  12   n  1  i 1 n       

s x  s 2x  1,897 .7  43 .56 2   n     yi   (106 .4) 2  1      n 957 . 2  1    1.25  i 1    y i2   s 2y  = 12  1  12  n  1  i 1 n       

s Y  s 2Y  1.25  1.12 r

s xy sxsy

26 .16

 .5362

(43 .56 )(1.12 )

R2 = r2 = .53622 = .2875 The covariance is 26.16, the coefficient of correlation is .5362, and the coefficient of determination is .2875. The coefficient of determination tells us that 28.75% of the variation in MBA GPAs is explained by the variation in GMAT scores.

4.88

109


R2 = r2 = (-.1913)2 = .0366

4.89 a

R2 = r2 = (.2543)2 = .0647. b There is a weak linear relationship between age and medical expenses. Only 6.47% of the variation in average medical bills is explained by the variation in age. c 80.00 70.00

60.00

Expense

50.00

y = 0.2257x - 5.9662 R² = 0.0647

40.00 30.00 20.00 10.00

0.00 -10.00 0

20

40

60

80

100

Age

The least squares line is ŷ  5.966  .2257 x d For each additional year of age mean medical expenses increase on average by $.2257 or 23 cents. e Charge 25 cents per day per year of age.

4.90

R2= (−.0886)2 = .0078 110


Only 0.78% of the variation in the number of houses sold is explained by the variation in interest rates.

4.91 300 250 y = 0,049x + 74,204 R² = 0,0004

200 150 100 50 0 0

20

40

60

80

100

120

140

160

Only 0.04% of the variation in the number of wells drilled is explained by the variation in the price of oil. The relationship is too weak to interpret the value of the slope coefficient.

4.92

R2 = (.0830)2 = .0069. There is a very weak positive relationship between the two variables.

111


4.93 480 460

440 Labor cost

420 400

y = 3.3x + 315.5 R² = 0.5925

380 360 340

320 300 0

10

20

30

40

50

Batch size

ŷ = 315.5 + 3.3x; Fixed costs = $315.50, variable costs = $3.30

4.94 1800 1600 y = 71.654x + 263.4 R² = 0.5437

1400

Cost

1200 1000

800 600 400

200 0 0

2

4

6

8

10

12

14

16

Time

ŷ = 263.4 + 71.65x; Estimated fixed costs = $263.40, estimated variable costs = $71.65

112


4.95

4.96

4.97

4.98

4.99

4.100

113


All five commodities are negatively linearly related to the exchange rate. The relationships are moderately strong.

4.101 y = 366,89x + 771,69 R² = 0,2766

50000 45000 40000 35000 30000

25000 20000 15000 10000 5000 0 0

20

40

60

80

100

a. The slope coefficient is 366.89 and the coefficient of determination is .2766 b. For each additional win home attendance increases on average by 366.89.

4.102

114

120


y = 20,901x + 28811 R² = 0,0124

40000 35000 30000 25000 20000 15000 10000 5000

0 0

20

40

60

80

100

120

a. The coefficient of determination is .0124, which means that there is very little correlation between wins and away attendance. b. For each additional win away attendance increases on average by 20.90.

4.103a. y = 14771x + 1E+06 R² = 0,0753

4 000 000 3 500 000 3 000 000 2 500 000 2 000 000 1 500 000 1 000 000 500 000 0 0

20

40

60

80

100

b. For each additional win the home attendance increases on average by 14,771.

4.104a.

115

120


y = 2278,1x + 2E+06 R² = 0,0371

3 000 000 2 500 000 2 000 000 1 500 000 1 000 000 500 000

0 0

20

40

60

80

100

120

b. For each additional win the home attendance increases on average by 2278.1.

4.105 80

y = 0,641x - 9,6418 R² = 0,271

70 60 50 40 30 20 10 0 0,000

20,000

40,000

60,000

80,000

100,000

a. The marginal cost of one more win is 1 million/.641 = $1,560,000. b.

116

120,000


y = 71,69x + 14910 R² = 0,2625

25 000

20 000

15 000

10 000

5 000

0 0

10

20

30

40

50

60

70

80

b. For each additional win the home attendance increases on average by 71.69. c. y = 8,7705x + 17489 R² = 0,0533

19 500 19 000 18 500 18 000 17 500 17 000 16 500 0

10

20

30

40

50

60

70

b. For each additional win the away attendance increases on average by 8.77.

4.106a.

117

80


y = 0,6812x - 4,4414 R² = 0,3335

70 60 50 40 30 20 10

0 0,000

20,000

40,000

60,000

80,000

100,000

120,000

a. The marginal cost of one more win is 1 million/.6812 = $1,467,998. b.

y = 3328,2x + 574331 R² = 0,2605

1 000 000 900 000 800 000 700 000 600 000 500 000 400 000 300 000 200 000 100 000 0 0

10

20

30

40

50

60

For each additional win the home attendance increases on average by 3328.2. c.

118

70


800 000 y = 1432,3x + 652547 R² = 0,4136

780 000 760 000 740 000 720 000 700 000 680 000 660 000

640 000 0

10

20

30

40

50

60

70

c. For each additional win the away attendance increases on average by 1432.3.

4.107a. y = 0,025x + 4,9339 R² = 0,0206

16 14 12 10 8 6 4 2 0 0,000

20,000

40,000

60,000

80,000 100,000 120,000 140,000 160,000 180,000

The marginal cost of one more win is 1 million/.025 = $40,000,000.

119


b. 100 000 y = 109,98x + 67394 R² = 0,0016

90 000 80 000 70 000 60 000 50 000 40 000 30 000 20 000 10 000 0

0

2

4

6

8

10

12

14

16

For each additional win the home attendance increases on average by 109.98. c. y = 85,184x + 67718 R² = 0,007

76 000 74 000 72 000 70 000 68 000

66 000 64 000 62 000 0

2

4

6

8

10

12

14

c. For each additional win the away attendance increases on average by 85.184.

4.108a.

120

16


y = -0,0329x + 10,511 R² = 0,0135

14 12 10 8 6 4 2 0 0,000

20,000

40,000

60,000

80,000

100,000

120,000

This is a strange result-a negative slope means that higher payrolls lead to fewer wins. b. y = 5796,6x + 490457 R² = 0,0697

800 000 700 000 600 000 500 000 400 000 300 000 200 000 100 000 0 0

2

4

6

8

10

12

For each additional win the home attendance increases on average by 5796.6. c.

121

14


y = 855,4x + 529987 R² = 0,0068

700 000 600 000 500 000 400 000 300 000 200 000 100 000

0 0

2

4

6

8

10

12

14

For each additional win the away attendance increases on average by 855.4.

4.109a. y = 0,394x + 13,304 R² = 0,2955

60 50 40 30 20 10 0 0,000

20,000

40,000

60,000

80,000

a. The marginal cost of one more win is 1 million/.394 = $2,538,071. b.

122

100,000


y = 42,729x + 16886 R² = 0,027

25 000

20 000

15 000

10 000

5 000

0 0

10

20

30

40

50

60

For each additional win the home attendance increases on average by 42.7294. c. 19 500 y = -1,5092x + 17571 R² = 0,0006

19 000 18 500 18 000 17 500 17 000 16 500 0

10

20

30

40

There is a negative relationship-more wins smaller attendance.

4.110a.

123

50

60


y = 0,3745x + 2,8509 R² = 0,1609

40 35 30 25 20 15 10 5

0 0,000

10,000

20,000

30,000

40,000

50,000

60,000

70,000

The marginal cost of one more win is 1 million/.3745 = $2,670,227. b. y = 3131,3x + 350143 R² = 0,1154

600 000 500 000 400 000 300 000 200 000 100 000 0 0

5

10

15

20

25

30

35

For each additional win the home attendance increases on average by 3131.3. c.

124

40


y = 458,6x + 414278 R² = 0,0533

450 000 445 000 440 000 435 000 430 000 425 000 420 000

415 000 410 000 405 000 400 000

395 000 0

5

10

15

20

25

30

35

For each additional win the away attendance increases on average by 458.6.

4.111 .4646; .1946 4.112 1.1193; .4355 4.113 1.1907; .5485 4.114 .8844; .4169 4.115 .6197; .1247 4.116 .8266; .3427 4.117 x 

1.1057  1.6057  .8823 3.5937   1.1979 3 3

4.118 x 

1.1665  1.1186  .46463 2.7497   .9166 3 3

4.119 x 

.6558  .9031  .5841 2.143   .7143 3 3

4.120 1.5876; .1236 4.121 .2312; .0185 4.122 .4031; .0829 4.123 x 

.7588  .8089  .8572 2.4249   .8083 3 3

125

40


4.124 x 

1.7742  .2312  1.6824 3.6878   1.2293 3 3

4.125 x 

.3805  .4031 .7836   .3918 2 2

4.126 1.0815; .2846 4.127 .6446; .0629 4.128 1.7154; .1069 4.129 x 

.9556  1.173  .9528 3.0814   1.0271 3 3

4.130 x 

1.173  .2818  1.1056 2.5604   .8535 3 3

4.131 x 

1.1397  .8228  .8076 2.7701   .9234 3 3

4.132

R2 = .67842 = .4603; 46.03% of the variation in statistics marks is explained by the variation in calculus marks. The coefficient of determination provides a more precise indication of the strength of the linear relationship.

4.133 2500 y = 116.53x - 369.93 R² = 0.9334

2000

Cost

1500 1000 500 0 0

5

10

15 Speed

126

20

25


The least squares line is ŷ = 369.93 + 116.53x. On average for each addition mph the cost of repair increases by $116.53.

4.134 y = 0.6041x + 17.933 R² = 0.0505

90 80

70 Income

60 50

40 30 20

10 0 60

65

70

75

80

Height

a ŷ = 17.933 + .6041x b The coefficient of determination is .0505, which indicates that only 5.05% of the variation in incomes is explained by the variation in heights.

4.135 3000.00

y = 19.059x + 1087.7 R² = 0.0779

2500.00

Sales

2000.00 1500.00 1000.00 500.00 0.00 0

5

10

15

20 Time

127

25

30

35


The coefficient of determination is .0779, which indicates that only 7.79% of the variation in sales is explained by the time between movies.

4.136 400

350

y = 0.07x + 103.44 R² = 0.5201

300

Price

250 200 150

100 50 0 0

500

1000

1500

2000

2500

3000

3500

4000

Size

a ŷ = 103.44 + .07x b. For each additional square foot the price increases on average by $.07 thousand. More simply for each additional square foot the price increases on average by$70. c. From the least squares line we can more precisely measure the relationship between the two variables.

Frequency

4.137 Private course 50 40 30 20 10 0

220

230

240

250

260

Time

Public course

128


Frequency

50 40 30 20 10 0

250

275

300

325

350

375

Time

The information obtained here is more detailed than the information provided by the box plots.

4.138

Frequency

80 60 40 20 0 24

28

32

36

40

44

48

52

Times The times are positively skewed.

4.139

a x  35.01, median = 36

129

56


b s = 7.68 c Half of the bone density losses lie below 36. At least 75% of the numbers lie between 19.64 and 50.38, at least 88.9% of the numbers lie between 11.96 and 58.06.

4.140

a x = 29,913, median = 30,660 b s 2 = 148,213,791; s = 12,174 c The number of coffees sold varies considerably.

4.141

R2 = r2 = .57422 = .3297; 32.97% of the variation in bone loss is explained by the variation in age.

4.142 a & b

130


70000 y = -553,7x + 49337 R² = 0,5489

60000 50000 40000 30000 20000 10000

0 -20

-10

0

10

20

30

40

50

60

70

R2 = .5489 and the least squares line is ŷ = 49,337 – 553.7x c 54.8% of the variation in the number of coffees sold is explained by the variation in temperature. For each additional degree of temperature the number of coffees sold decreases on average by 554 cups. Alternatively for each 1-degree drop in temperature the number of coffees increases on average, by 553.7 cups. d We can measure the strength of the linear relationship accurately and the slope coefficient gives information about how temperature and the number of coffees sold are related.

4.143a mean, median, and standard deviation b

x = 93.90, s = 7.72

131


c We hope Chris is better at statistics than he is golf.

4.144

Mean, median, mode: 34,656, 34,636, 35,149

4.145

There is a moderately strong positive linear relationship.

4.146

132


25 y = 0,3295x + 10,153 R² = 0,2036 20

15

10

5

0 0

5

10

15

20

25

a. The coefficient of determination is .2036. b. The slope coefficient is .3295; for each additional year of education of the father the son’s years of education increase on average by .3295.

4.147 25

y = 0,3452x + 9,8579 R² = 0,1945

20

15

10

5

0 0

5

10

15

20

25

a. The coefficient of determination is .1945. b. The slope coefficient is .3452; for each additional year of education of the father the son’s years of education increase on average by .3452.

4.148

133


30 y = 0,0205x + 1,9801 R² = 0,019

25 20 15 10 5

0 0

20

40

60

80

100

The relationship is very weak. For each additional year of age the number of hours of TV increases on average by .0205.

4.149 y = 0,7744x + 13,903 R² = 0,1514

60 50 40 30 20 10 0 0

5

10

15

20

a. The coefficient of determination is .1514. b. For each additional year of education the age increases by .7744.

4.150

134

25


30 y = -0,1757x + 5,3998 R² = 0,0424

25 20 15 10 5

0 0

5

10

15

20

25

a. The coefficient of determination is .0424. b. For each additional year of education the number of hours of TV decreases by .1757.

4.151

The coefficient of correlation is almost 1.

4.152 y = 6,0186x + 4E+06 R² = 0,4324

1 400 000 000,00 1 200 000 000,00 1 000 000 000,00 800 000 000,00

600 000 000,00 400 000 000,00 200 000 000,00 0,00 0,00

50 000 000,00 100 000 000,00 150 000 000,00 200 000 000,00

-200 000 000,00

The coefficient of determination is .4324. 135


For each additional dollar of income assets increase on average by 6.0186

4.153 y = 17,799x + 7E+06 R² = 0,047

1 400 000 000,00 1 200 000 000,00 1 000 000 000,00 800 000 000,00 600 000 000,00

400 000 000,00 200 000 000,00 0,00 0,00

5 000 000,0010 000 000,0015 000 000,0020 000 000,0025 000 000,00

-200 000 000,00

The coefficient of determination is .047 indicating a weak linear relationship. For each additional dollar of wage income assets increase on average by 17.799.

4.154 9 y = -0,0249x + 2,1289 R² = 0,1176

8 7 6 5 4 3 2 1 0 -1

0

20

40

60

80

100

For each additional year of age the number of children decreases on average by .0249.

136


Case 4.1 a Scatter diagrams with time as the independent variable and temperature anomalies as the dependent variable

3 y = 0,0008x - 0,6381 R² = 0,5404

2,5

Temperature anomalies

2 1,5 1 0,5 0 -0,5

0

200

400

600

800

1000

1200

1400

1600

1800

-1 -1,5 -2 -2,5

Month

Monthly average increase is .0008. For the 1637 month period the increase was 1637(.0008) = 1.31o Celsius.

Scatter diagram with carbon dioxide levels as the independent variable and temperature anomalies as the dependent variable

3 y = 0,0156x - 5,0641 R² = 0,6123

Temperature anomalies

2,5 2 1,5 1 0,5 0 -0,5

300

320

340

360

380

-1 -1,5

Month

137

400

420


The coefficient of determination is .6123, which means that 61.23% of the variation in temperature anomalies is explained by the variation in CO2levels. There is a moderately strong linear relationship.

Case 4.2 1880 to 1940 1,5

y = 0,0007x - 0,5429 R² = 0,1793

Temperature anomalies

1 0,5 0 0

100

200

300

400

500

600

700

800

-0,5 -1 -1,5 -2 -2,5

Month

From 1880 to 1940 the earth warned at an average monthly rate of .0007 o Celsius.

1941 to 1975 y = -0,0001x - 0,0064 R² = 0,0023

1

Temperature anomalies

0,5

0 0

50

100

150

200

250

300

350

400

-0,5

-1

-1,5

Month

From 1941 to 1975 the earth cooled at an average monthly rate of .0001o Celsius

138

450


1976 to 1997 y2 = 0,0021x + 0,0388 R² = 0,2026

Temperature anomalies

1,5 1 0,5 0 0

50

100

150

200

250

300

-0,5 -1

Month

From 1976 to 1997 the earth warmed at an average monthly rate of .0021 o Celsius.

1998 to 2016 y = 0,0018x + 0,7418 R² = 0,1015

Temperature anomalies

2,5

2

1,5

1

0,5

0 0

50

100

150

200

Month

From 1998 to 2012 the earth warmed at an average monthly rate of .001 o Celsius

Over different periods of time the earth has warmed and cooled.

139

250


Case 4.3 2003-04 Season 60 y = 0.1526x + 28.559 R² = 0.0876

50

Wins

40 30 20 10 0 $0

$20

$40

$60

$80

$100

Payroll ($millions)

The cost of winning one additional game is 1million/.1526 = $6.553 million. However, the coefficient of determination is only .0876, which tells us that there are many other variables that determine how well a team will do.

2005-06 Season 70 y = 0.7795x + 14.256 R² = 0.3072

60

Wins

50 40 30 20 10 0 $0

$10

$20

$30

$40

$50

Payroll ($millions)

The cost of winning one additional game is 1million/.7795 = $1.283 million. The coefficient of determination is .3072.

140


The small coefficient of determination in the year before the strike seems to indicate that team owners were spending large amounts of money and getting little in return. The results are markedly different in the year after the strike. There is a much stronger linear relationship between payroll and the number of wins and the cost of winning one additional game is considerably smaller.

Case 4.4

The coefficient of determination is (−.1787)2 = .0319. There is a weak negative linear relationship between percentage of rejected ballots and Percentage of “yes” votes.

The coefficient of determination is (.3600)2 = .1296. There is a moderate positive linear relationship between percentage of rejected ballots and Percentage of Allophones.

The coefficient of determination is (.0678)2 = .0046. There is a very weak positive linear relationship between percentage of rejected ballots and Percentage of Allophones.

The statistics provide some evidence that electoral fraud has taken place.

141


Chapter 5 5.1 In an observational study, there is no attempt to control factors that might influence the variable of interest. In an experimental study, a factor (such as regular use of a fitness center) is controlled by randomly selecting who is exposed to that factor, thereby reducing the influence of other factors on the variable of interest.

5.2a The study is observational. The statistics practitioner did not randomly assign stores to buy cans or bottles. b Randomly assign some stores to receive only cans and others to receive only bottles.

5.3 Randomly sample smokers and nonsmokers and compute the proportion of each group that has lung cancer. b The study is observational. Experimental data would require the statistics practitioner to randomly assign some people to smoke and others not to smoke.

5.4a A survey can be conducted by means of a personal interview, a telephone interview, or a selfadministered questionnaire. b A personal interview has a high response rate relative to other survey methods, but is expensive because of the need to hire well-trained interviewers and possibly pay travel-related costs if the survey is conducted over a large geographical area. A personal interview also will likely result in fewer incorrect responses that arise when respondents misunderstand some questions. A telephone interview is less expensive, but will likely result in a lower response rate. A self-administered questionnaire is least expensive, but suffers from lower response rates and accuracy than interviews.

5.5 Five important points to consider when designing a questionnaire are as follows: (1) The questionnaire should be short. (2) Questions should be short, clearly worded, and unambiguous. (3) Consider using dichotomous or multiple-choice questions, but take care that respondents needn’t make unspecified assumptions before answering the questions. (4) Avoid using leading questions. (5) When preparing the questions, think about how you intend to tabulate and analyze the responses.

5.6a The sampled population will exclude those who avoid large department stores in favor or smaller shops, as well as those who consider their time too valuable to spend participating in a 153


survey. The sampled population will therefore differ from the target population of all customers who regularly shop at the mall. b The sampled population will contain a disproportionate number of thick books, because of the manner in which the sample is selected. c The sampled population consists of those eligible voters who are at home in the afternoon, thereby excluding most of those with full-time jobs (or at school).

5.7a The Literary Digest was a popular magazine in the 1920s and 1930s which had correctly predicted the outcome of many presidential elections. To help predict the outcome of the 1936 presidential election, the Literary Digest mailed sample ballots to 10 million prospective voters. Based on the results of the ballots returned, the magazine predicted that the Republican candidate, Alfred Landon, would defeat the Democratic incumbent, Franklin D. Roosevelt, by a 3 to 2 margin. In fact, Roosevelt won a landslide victory, capturing 62% of the votes. b The main reason for the poll being so wrong was nonresponse bias resulting from a self-selected sample, causing the sample to be unrepresentative of the target population. (Only 2.3 million ballots were returned.) The second reason was selection bias, resulting from poor sampling design, causing the sampled population and the target population to differ. Most of those to whom a ballot was sent were selected from the Literary Digest’s subscription list and from telephone directories. These people tended to be wealthier than average and tended to vote Republican.

5.8a A self-selected sample is a sample formed primarily on the basis of voluntary inclusion, with little control by the designer of the survey. b Choose any recent radio or television poll based on responses of listeners who phone in on a volunteer basis. c Self-selected samples are usually biased, because those who participate are more interested in the issue than those who don’t, and therefore probably have a different opinion.

5.9 We should ignore the results because this is an example of a self-selected sample.

5.10 No, because the sampled population consists of the responses about the professor’s course. We cannot make draw inferences about all courses.

5.11 We used Excel to generate 40 three-digit random numbers. Because we will ignore all randomly generated numbers over 800, we can expect to ignore about 20% (or about 8 to 10) of the randomly generated numbers. We will also ignore any duplication. We therefore chose to generate 40 three-digit random numbers, and will use the first 25 unique random numbers less than 801 to select our sample. The 40 numbers generated are shown below, with a stroke through

154


those to be ignored.

6

357

456

449

862

154

55

412

475

430

999

912

60

207

717

651

10

294

327

165

576

871

990

354

390

540

893

181

496

870

738

820

32

963

160

32

231

86

970

46

5.12 We used Excel to generate 30 six-digit random numbers. Because we will ignore any duplicate numbers generated, we generated 30 six-digit random numbers and will use the first 20 unique random numbers to select our sample. The 30 numbers generated are shown below.

169,470

744,530

22,554

918,730

320,262

503,129

318,858

698,203

822,383

938,262

800,806

56,643

836,116

123,936

80,539

154,211

391,278

940,154

110,630

856,380

222,145

692,313

949,828

561,511

909,269

811,274

288,553

749,627

858,944

39,308

5.13 Stratified random sampling is recommended. The strata are the school of business, the faculty of arts, the graduate school and the all the other schools and faculties would be the fourth stratum. The data can be used to acquire information about the entire campus but also compare the four strata.

5.14 A stratified random sampling plan accomplishes the president’s goals. The strata are the four areas enabling the statistics practitioner to learn about the entire population but also compare the four areas.

5.15 The operations manager can select stratified random samples where the strata are the four departments. Simple random sampling can be conducted in each department.

5.16 Use cluster sampling, letting each city block represent a cluster.

5.17a Sampling error refers to an inaccuracy in a statement about a population that arises because the statement is based only on sample data. We expect this type of error to occur because we are making a statement based on incomplete information. Nonsampling error refers to mistakes made in the acquisition of data or due to the sample observations being selected improperly. b Nonsampling error is more serious because, unlike sampling error, it cannot be diminished by taking a larger sample.

155


5.18 Three types of nonsampling errors: (1) Error due to incorrect responses (2)Nonresponse error, which refers to error introduced when responses are not obtained from some members of the sample. This may result in the sample being unrepresentative of the target population. (3)Error due to selection bias, which arises when the sampling plan is such that some members of the target population cannot possibly be selected for inclusion in the sample.

5.19 Yes. A census will likely contain significantly more nonsampling errors than a carefully conducted sample survey.

156


Chapter 6 6.1 a Relative frequency approach b If the conditions today repeat themselves an infinite number of days rain will fall on 10% of the next days. 6.2 a Subjective approach b If all the teams in major league baseball have exactly the same players the New York Yankees will win 25% of all World Series. 6.3 a {a is correct, b is correct, c is correct, d is correct, e is correct} b P(a is correct) = P(b is correct) = P(c is correct) = P(d is correct) = P(e is correct) = .2 c Classical approach d In the long run all answers are equally likely to be correct. 6.4 a Subjective approach b The Dow Jones Industrial Index will increase on 60% of the days if economic conditions remain unchanged. 6.5 a P(even number) = P(2) + P(4) + P(6) = 1/6 + 1/6 + 1/6 = 3/6 = 1/2 b P(number less than or equal to 4) = P(1) + P(2) + P(3) + P(4) = 1/6 + 1/6 + 1/6 +1/6 = 4/6 = 2/3 c P(number greater than or equal to 5) = P(5) + P(6) = 1/6 + 1/6 = 2/6 = 1/3 6.6 {Adams wins. Brown wins, Collins wins, Dalton wins} 6.7a P(Adams loses) = P(Brown wins) + P(Collins wins) + P(Dalton wins) = .09 + .27 + .22 = .58 b P(either Brown or Dalton wins) = P(Brown wins) + P(Dalton wins) = .09 + .22 = .31 c P(either Adams, Brown, or Collins wins) = P(Adams wins) + P(Brown wins) + P(Collins wins) = .42 + .09 + .27 = .78 6.8 a {0, 1, 2, 3, 4, 5} b {4, 5} c P(5) = .10 d P(2, 3, or 4) = P(2) + P(3) + P(4) = .26 + .21 + .18 = .65 e P(6) = 0 6.9 {Contractor 1 wins, Contractor 2 wins, Contractor 3 wins} 6.10 P(Contractor 1 wins) = 2/6, P(Contractor 2 wins) = 3/6, P(Contractor 3 wins) = 1/6


6.11 a {Shopper pays cash, shopper pays by credit card, shopper pays by debit card} b P(Shopper pays cash) = .30, P(Shopper pays by credit card) = .60, P(Shopper pays by debit card) = .10 c Relative frequency approach 6.12 a P(shopper does not use credit card) = P(shopper pays cash) + P(shopper pays by debit card) = .30 + .10 = .40 b P(shopper pays cash or uses a credit card) = P(shopper pays cash) + P(shopper pays by credit card) = .30 + .60 = .90 6.13 {single, divorced, widowed} 6.14 a P(single) = .15, P(married) = .50, P(divorced) = .25, P(widowed) = .10 b Relative frequency approach 6.15 a P(single) = .15 b P(adult is not divorced) = P(single) + P(married) + P(widowed) = .15+ .50 + .10 = .75 c P(adult is either widowed or divorced) = P(divorced) + P(widowed) = .25 + .10 = .35 6.16 {Spanish, Chinese, Tagalog, Vietnamese, French, Korean, others} 6.17 a .619 b .381 c .0435 d .245 6.18 {Very safe, Somewhat safe, Somewhat unsafe, Very unsafe, Not sure} 6.19a .17 b .45 c .12

6.20 P( A1 ) = .1 + .2 = .3, P( A 2 ) = .3 + .1 = .4, P( A 3 ) = .2 + .1 = .3. P( B1 ) = .1 + .3 + .2 = .6, P( B 2 ) = .2 + .1 + .1 = .4.

6.21 P( A1 ) = .4 + .2 = .6, P( A 2 ) = .3 + .1 = .4. P( B1 ) = .4 + .3 = .7, P( B 2 ) = .2 + .1 = .3.


P(A1 and B1 )

6.22 a P(A1 | B1 ) 

P(B1 )

.4  .57 .7

P(A 2 and B1 ) .3   .43 P(B1 ) .7

b P(A 2 | B1 ) 

c Yes. It is not a coincidence. Given B1 the events A1 and A 2 constitute the entire sample space.

6.23 a P(A1 | B 2 )  b P( B 2 | A 1 ) 

P(A1 and B 2 ) P( B 2 )

.2  .67 .3

P(A1 and B2 ) .2   .33 P(A1 ) .6

c One of the conditional probabilities would be greater than 1, which is not possible. 6.24 The events are not independent because P(A1 | B 2 )  P(A1 ) . 6.25 a P( A 1 or B1 ) = P(A1 )  P(B1 )  P(A1 and B1 ) = .6 + .7 - .4 = .9 b P( A 1 or B 2 ) = P(A1 )  P(B 2 )  P(A1 and B 2 ) = .6 + .3 - .2 = .7 c P( A 1 or A 2 ) = P(A1 )  P(A 2 ) = .6 + .4 = 1

6.26 P(A 1 | B1 ) 

P(A 1 and B1 ) .20   .25 ; P(A1 )  .20  .05  .25 ; the events are independent. P(B1 ) .20  .60

6.27 P(A1 | B1 ) 

P(A1 and B1 ) .20   .571 ; P(A1 )  .20  .60  .80 ; the events are dependent. P(B1 ) .20  .15

6.28 P( A1 ) = .15 + .25 = .40, P( A 2 ) = .20 + .25 = .45, P( A 3 ) = .10 + .05 = .15. P( B1 ) = .15 + 20 + .10 = .45, P( B 2 ) = .25 + .25 + .05 = .55.

6.29 a P( A 2 | B 2 ) =

P(A 2 and B 2 ) .25   .455 P( B 2 ) .55

b P( B 2 | A 2 ) =

P(A 2 and B 2 ) .25   .556 P( A 2 ) .45

c P( B1 | A 2 ) =

P(A 2 and B1 ) .20   .444 P( A 2 ) .45

6.30 a P( A 1 or A 2 ) = P( A 1 ) + P( A 2 ) = .40 + .45 = .85 b P( A 2 or B 2 ) = P( A 2 ) + P( B 2 ) – P( A 2 and B 2 ) = .45 + .55 - .25 = .75


c P( A 3 or B1 ) =P( A 3 ) + P( B1 ) – P( A 3 and B1 ) = .15 + .45 - .10 = .50

P(promoted and female ) .03   .20 P(female ) .03  .12

6.31 a P(promoted | female) =

b P(promoted | male) =

P(promoted and male ) .17  .20  P(male ) .17  .68

c No, because promotion and gender are independent events. 6.32 a P(debit card) = .04 + .18 + .14 = .36 b P(over $100 | credit card) =

P(credit card and over $100 .23   .49 P(credit card) .03  .21  .23

c P(credit card or debit card) = P(credit card) + P(debit card) = .47 + .36 = .83 6.33 a P(Less than high school) = .057 + .104 = .161 b P(college/university | female) =

c P(high school | male) =

P(college / university and female ) .095   .226 P(female ) .057  .136  .132  .095

P( high school and male ) .224   .622 P( male ) .136  .224

6.34 a P(He is a smoker) = .12 + .19 = .31 b P(He does not have lung disease) = .19 + .66 = .85 c P(He has lung disease | he is a smoker) =

P(he has lung disease and he is a smo ker) .12   .387 P(he is a smo ker) .31

d P(He has lung disease | he does not smoke) =

P(he has lung disease and he does not smoke ) .03   .043 P(he does not smoke ) .69

6.35 The events are dependent because P(he has lung disease) = .15, P(he has lung disease | he is a smoker) = .387

P(manual and math  stats) .23   .390 P(math  stats) .23  .36

6.36a. P(manual | math-stats) = b. P(computer) = .36 + .30 = .66

c. No, because P(manual) = .23 + .11 = .34, which is not equal to P(manual | math-stats). 6.37 a P(customer will return and good rating) =.35 b P(good rating | will return) =

P(good rating and will return ) .35 .35    .538 P( will return ) .02  .08  .35  .20 .65


c P(will return| good rating)

P(good rating and will return ) .35 .35    .714 P(good rating ) .35  .14 .49

d (a) is the joint probability and (b) and (c) are conditional probabilities 6.38 a P(ulcer) = .01 + .03 + .03 + .04 = .11 b P(ulcer | none) =

P(ulcer and none) .01 .01   .043  P(none) .01  .22 .23

c P(none | ulcer) =

P(ulcer and none) .01 .01    .091 P(ulcer) .01  .03  .03  .04 .11

d P(One, two, or more than two | ulcer) = 1 

6.39 a P(Insufficient work | 25-54) =

P(ulcer and none)  1  .091  .909 P(ulcer)

P(Insufficie nt work and 25  54 ) .180  .252  P(25  54 ) .320  .180  .214

b P(65 and over) = .029 + .011 + .016 = .056 c P(65 and over |plant or company closed or moved) =

P(65 and over and plant or company closed or moved ) .029  .064  P(plant or company closed or moved ) .015  .320  ..089  .029

6.40 a P(remember) = .15 + .18 = .33 b P(remember | violent) =

P(remember and violent ) .15 .15    .30 P( violent ) .15  .35 .50

c Yes, the events are dependent.

6.41 a P(above average | murderer) =

P(above average and murderer ) .27 .27    .563 P(murderer ) .27  .21 .48

b No, because P(above average) = .27 + .24 = .51, which is not equal to P(above average testosterone | murderer). 6.42a P(Health insurance) = .167 +.209 +.225 +.177 = .778 b. P(Person 55-64 | No health insurance) =

P(Person 55  64 and No health insurance) .026 .026    .128 P(Person 55  64 ) .177  .026 .203

c. P(Person 25-34|No health insurance) =

P(Person 25  34 and No health insurance) .085 .085    .385 P( No health insurance) .085  .061  .049  .026 .221

6.43a

P(Violent crime and primary school) .393 .393    .673 P(Pr imary school) .393  .191 .584


b. P(No violent crime) = .191 + .010 + .007 + .015 = .223

6.44a

b.

P(Violent crime and enrollment less than 300 ) .159 .159    .636 P(Enrollment less than 300 ) .159  .091 .250

P(Violent crime and enrollment less than 300 ) .159 .159    .205 P(Violent crime ) .159  .221  .289  .108 .777

6.45 a P(new | overdue) =

b P(overdue | new) =

P(new and overdue) .06 .06    .103 P(overdue) .06  .52 .58

P(new and overdue) .06 .06    .316 P(new) .06  .13 .19

c Yes, because P(new) = .19  P(new | overdue) 6.46 a P(under 20) = .464 + .147 + .237 = .848 b P(retail) = .237 + .035 + .005 = .277 c P(20 to 99 | construction) =

P(20 to 99 and construction) .039 .039    .077 P(construction) .464  .039  .005 .508

6.47 a P(fully repaid) = .19 + .64 = .83 b P(fully repaid | under 400) =

P(fully repaid and under 400 ) .19 .19   .594  P(under 400 ) .19  .13 .32

c P(fully repaid | 400 or more) =

P(fully repaid and 400 or more ) .64 .64    .941 P(400 or more ) .64  .04 .68

d No, because P(fully repaid)  P(fully repaid | under 400)

6.48 P(purchase | see ad) =

P(purchase and see ad) .18 .18    .30; P(purchase) = .18 + .12 = .30. The P(see ad) .18  .42 .60

events are independent and thus, the ads are not effective. 6.49 a P(unemployed | high school graduate) =

P( unemployed and high school graduate) .035 .035    .120 P( high school graduate) .257  .035 .292 b P(employed) = .075 + .2572 + .155 + .096 + .211 + .118 = .912 c P(advanced degree | unemployed) = P(advanced deg ree and unemployed) .004 .004    .044 P( unemployed) .015  .035  .016  .008  .012  .004 .090

d P(not a high school graduate) = .075 + .015 = .090


6.50 a P(bachelor’s degree | west) =

P( bachelor' s deg ree and west) .050 .050    .216 P( west) .032  .058  .044  .022  .050  .025 .231

b P(northeast | high school graduate) =

P( northeast and high school graduate) .062 .062    .198 P( high school graduate) .062  .075  .118  .058 .313

c P(south) = .053 + .118 + .062 + .032 + .067 + .036 = .368 d P(not south) = 1 – P(south) = 1−.368 = .632 6.51a. P(Some | White) = .356/.851 = .418 b. P(Very little or no | Black) = .06/.149 = .403 c. P(White | Some) = .356/.404 = .881 6.52a. P(Arthritis | Over 80) = .105/.140 = .750 b. P(No arthritis | 50-60) = .360/.400 = .900 c. P(60-70 | Arthritis) = .075/.292) = .257 6.53a. P(New Democrats | Men) = .044/.490 = .0898 b. P(Women | Liberal) = .224/.415 = .540 c. P(Conservative) = .255 + .215 = .470 6.54a. P(Married | Millennial) = .089/.331) = .269 b. P(Single, never married | Baby boomer) = .030/.310 = .0968 c. P(Married) = .089 + .223 + .201 = .513 d. P(Generation X | Living with not married) = .025/.064 =.391 6.55a. P(Trust) = .0896 + .1386 + .1944 + .0629 + .0144 = .4999 b. P(Distrust | Consistent conservative) = .0558/.0900 = .6200 c. P(Neither | Consistent liberal) = .0576/.1600 = .3600 d. P(Consistent liberal) = .0896 + .0096 + .0576 + .0032 = .1600 6.56a. P(Distrust | Mostly conservative) = .0680/.1700 = .4000 b. P(Neither | Mixed) = .1120/.3600 = .3111 c. P(Trust) = .0832 + .1056 + .1400 + .0442 + .0063 = .3797 d. P(Mostly conservative) = .0442 + .0680 + .0442 + .0136 = .1700 6.57a. P(Distrust | Consistent liberal) = .1296/.1600 = .8100


b. P(Trust | Mostly conservative) = .1224/.1700 = .7200 c. P(Neither | Consistent conservative) = .0045/.0900 = .0500 d. P(Consistent conservative) = .0792 + .0027 + .0045 + .0036 = .0900 6.58a. P(Distrust) = .0192 + .0242 + .0504 + .0561 + .0549 = .2048 b. P(Trust | Consistent conservative) = .0126/.0900 = .1400 c. P(Neither | Mostly liberal) = .0396/.2200 = .1800 d. P(Mixed) = .2196 + .0504 + .0612 + .0288 = .3600 6.59

6.60


6.61

6.62

6.63


6.64

a P(R and R) = .81 b P(L and L) = .01 c P(R and L) + P(L and R) = .09 + .09 = .18 d P(Rand L) + P(L and R) + P(R and R) = .09 + .09 + .81 = .99 6.65 a & b

c

0 right-handers

1

1 right-hander

3

2 right-handers

3

3 right-handers

1

d P(0 right-handers) = .001


P(1 right-hander) = 3(.009) = .027 P(2 right-handers) = 3(.081) = .243 P(3 right-handers) = .729 6.66a

b P(RR) = .8091 c P(LL) = .0091 d P(RL) + P(LR) = .0909 + .0909 = .1818 e P(RL) + P(LR) + P(RR) = .0909 + .0909 + .8091 = .9909 6.67a

P(0 right-handers) = (10/100)(9/99)(8/98) = .0007


P(1 right-hander) = 3(90/100)(10/99)(9/98) = .0249 P)2 right-handers) = 3(90/100)(89/99)(10/98) = .2478 P(3 right-handers) = (90/100)(89/99)(88/98) = .7265

6.68

a P(win both) = .28 b P(lose both) = .30 c P(win only one) = .12 + .30 = .42 6.69

P(sale) = .04

6.70


P(D) = .02 + .018 = .038 6.71

P(Same party affiliation) = P(DD) + P(RR) + P(OO) = .1936 + .1369 + .0361.3666

6.72


Diversity index = .12 + .04 + .12 + .0075 + .04 + .0075 = .335 6.73

P(heart attack) = .0504 + .0792 = .1296

6.74


P(pass) = .228 + .243 + .227 = .698 6.75

P(good ) = .3132 + .0416 = .3548

6.76


P(myopic) = .1008 + .1512 = .2520

6.77

P(does not have to be discarded) = .1848 + .78 = .9648 6.78 Let A = mutual fund outperforms the market in the first year B = mutual outperforms the market in the second year P(A and B) = P(A)P(B | A) = (.15)(.22) = .033 6.79 Let A = DJIA increase and B = NASDAQ increase P(A) = .60 and P(B | A) = .77 P(A and B) = P(A)P(B | A) = (.60)(.77) = .462

6.80 Define the events: M: The main control will fail. B1: The first backup will fail. B2: The second backup will fail


The probability that the plane will crash is P(M and B1 and B2) = [P(M)][ P(B1)][ P(B2)] = (.0001) (.01) (.01) = .00000001 We have assumed that the 3 systems will fail independently of one another. 6.81 P( wireless Web user uses it primarily for e-mail) = .69 P(3 wireless Web users use it primarily for e-mail) = (.69)(.69)(.69) = .3285 6.82

P(Increase) = .05 + .5625 = .6125 6.83 Number saying leaving is a bad thing = 630(.62) + 590(.74) + 480(.57) = 1100.8 P(bad thing) = 1100.8/(630 + 590 + 480) = .6475 6.84 Number disapproving = 385(.94) + 420(.70) + 475(.66) = 820.5 P(Disapprove) = 820.5/(385 + 420 + 475) = .6408 6.85 P(Student debt) = (.32)(.45) + (.15)(.39) + (.53)(.27) = .3456 6.86 P(pass) = (.38)(.79) + (.41)(.74) + (.13)(.68) + (.08)(.57) = .7376 6.87 P(A and B) = .36, P(B) = .36 + .07 = .43 P(A | B) =

P(A and B) .36   .837 P(B) .43

6.88 P(A and B) = .32, P(AC and B) = .14, P(B) = .46, P(B C) = .54


a P(A | B) =

P(A and B) .32   .696 P(B) .46

b P(AC | B) =

P(A C and B) .14   .304 P(B) .46

c P(A and BC) = .48; P(A | BC ) =

P(A and B C )

d P(AC and BC) = .06; P(AC | BC) =

C

P( B )

.48  .889 .54

P(A C and B C ) P( B C )

.06  .111 .54

6.89

P(B) = .4940 + .0115 = .5055 P(A | B) =

P(A and B) .4940   .9773 P(B) .5055

6.90 P(F | D) =

P(F and D) .020   .526 P(D) .038

6.91 Define events: A = crash with fatality, B = BAC is greater than .09) P(A) = .01, P(B | A) = .084, P(B) = .12 P(A and B) = (.01)(.084) = .00084 P(A | B) =

P(A and B) .00084   .007 P(B) .12


6.92 P(CFA I | passed) =

P(CFA I and passed) .228   .327 P(passed) .698

6.93 Define events: A = heart attack, B = periodontal disease P(A) = .10, P(B | A) = .85, P(B | AC ) = .29

P(B ) = .085 + .261 = .346 P(A | B) =

P(A and B) .085   .246 P(B) .346

6.94 P(A) = .40, P(B | A) = .85, P(B | AC ) = .29

P(B ) = .34 + .174 = .514 P(A | B) =

P(A and B) .34   .661 P(B) .514

6.95 Define events: A = smoke, B1 = did not finish high school, B 2 = high school graduate, B 3 = some college, no degree, B 4 = completed a degree


P(A | B1 ) = .40, P(A | B 2 ) = .34, P(A | B 3 ) = .24, P(A | B 4 ) = .14 From Exercise 6.45: P( B1 ) = .1055, P( B 2 ) = .3236, P( B 3 ) = .1847, P( B 4 ) = .3862

P(A) = .0422 + .1100 + .0443 + .0541 = .2506 P( B 4 | A) = .0541/.2506 = .2159

6.96 Define events: A, B, C = airlines A, B, and C, D = on time P(A) = .50, P(B) = .30, P(C) = .20, P(D | A) = .80, P(D | B) = .65, P(D | C) = .40

P(D) = .40 + .195 + .08 = .675


P(A | D) =

P(A and D) .40   .593 P(D) .675

6.97 Define events: A = win series, B = win first game P(A) = .60, P(B | A) = .70, P(B | AC ) =.25

P(BC ) = .18 + .30 = .48 P(A | BC ) =

P(A and B C ) C

P( B )

.18  .375 .48

6.98

P(PT) = .28 + .052 = .332 P(R | PT) =

P(R and PT ) .28   .843 P(PT ) .332


6.99

P(PT) = .0046 + .0269 = .0315 P(H | PT) =

P(H and PT ) .0046   .1460 P(PT ) .0315

6.100 Sensitivity = P(PT | H) = .920 Specificity = P(NT | H C ) = .973 Positive predictive value = P(H | PT) = .1460 Negative predictive value = P H C | NT) =

6.101

P(H C and NT ) .9681 .9681    .9996 P( NT ) .0004  .9681 .9685


P(PT) = .0164 + .6233 = .6397 P(NT) = .0036 + .3567 = .3603 P(C | PT) =

P(C and PT ) .0164   .0256 P(PT ) .6397

P(C | NT) =

P(C and NT ) .0036   .0010 P( NT ) .3603

6.102 P(Light| Myopic) =

P( Light and Myopic ) .1008   .4000 .2520 P( Myopic )

6.103 P(Italy | Bad thing) = Number of Italy and Bad thing / Number of bad thing) = 273.6/1100.8 = .2485 6.104 P(Greece | Disapprove) = Number of Greece and Disapprove / Number Disapprove) = 238.7/820.25 = .2910 6.105 P(Managerial/Professional | Student debt) = P(Managerial/Professional and Student debt) / P(Student Debt) = .1440/.3456 = .4167 6.106 P(BBA | Pass) = P(BBA and Pass) / P(Pass) = .3034/.7376 = .4113 6.107a P(Good Day) = Number of Good Day / Number of Respondents = (525(.30) + 650(.17) + 390(.41)) / 1565 = .2734 b P(United States | Typical Day) = Number of United States and Typical Day) / Number Typical Day = 191.1/1042.1 = .1834 6.108 a P(Marketing A) = .053 + .237 = .290 b P(Marketing A | Statistics not A) =

P(Marketing A and Statistics not A) .23 .237    .290 P(Statistics not A) .237  .580 .817

c Yes, the probabilities in Parts a and b are the same. 6.109 Define events: A = win contract A and B = win contract B


a P(A and B) = .12 b P(A and BC) + P(AC and B ) = .18 + .14 = .32 c P(A and B) + P(A and BC ) + P(AC and B ) = .12 + .18 + .14 = .44 6.110 a P(second) = .05 + .14 = .19 b P(successful | –8 or less) =

P(successfuland  8 or less) .15 .15    .517 P(8 or less) .15  .14 .29

c No, because P(successful) = .66 + .15 = .81, which is not equal to P(successful | –8 or less) . 6.111 Define events: A = woman, B = drug is effective

P(B) = .528 + .221 = .749

6.112 P(AC | B) =

P(A C and B) .221   .295 P(B) .749

6.113 P(Idle roughly) = P(at least one spark plug malfunctions) = 1– P(all function) = 1 – (.90 4 ) = 1-.6561 = .3439


6.114

P(no sale) = .65 + .175 = .825 6.115 a P(pass) = .86 + .03 = .89 b P(pass | miss 5 or more classes) =

P(pass and miss 5 or more classes) .03 .03    .250 P(miss 5 or more classes) .09  .03 .12

c P(pass | miss less than 5 classes) =

P(pass and miss less than 5 classes) .86 .86   .977 .  P(miss less than 5 classes) .86  .02 .88

d No since P(pass)  P(pass | miss 5 or more classes) 6.116 Define events: R = reoffend, D = detained

a P(D) = P(R and D) + P(R C and D) = .1107 + .2263 = .3370 P(R| D) =

P(R and D) .1107   .3285 P ( D) .3370

b P(D C ) = P(R and D C ) + P(R C and D C ) = .1593 + .5037 = .6630 P(R| D C ) =

P(R and D C ) P( D C )

.1593  .2403 .6630


6.117 a P(excellent) = .27 + .22 = .49 b P(excellent | man) =

P(man and excellent ) .22 .22    .44 P(man ) .22  .10  .12  .06 .50

c P(man | excellent) =

P(man and excellent ) .22 .22    .449 P(excellent ) .27  .22 .49

d No, since P(excellent)  P(excellent | man)

6.118

P(R) = .0176 + .5888 = .6064 P(S | R) =

P(S and R ) .5888   .9710 P( R ) .6064

6.119 Define events: A1 = Low-income earner, A 2 = medium-income earner, A 3 = high-income earner, B = die of a heart attack, BC survive a heart attack


P(BC ) = .1848 + .4459 + .2790 = .9097 P( A1 | BC ) =

P(A1 and B C ) C

P( B )

.1848  .2031 .9097

6.120 Define the events: A1 = envelope containing two Maui brochures is selected, A 2 = envelope containing two Oahu brochures is selected, A 3 = envelope containing one Maui and one Oahu brochures is selected. B = a Maui brochure is removed from the selected envelope.

P(B) = 1/3 + 0 + 1/6 = 1/2 P( A1 | B) =

P(A 1 and B) 1 / 3   2/3 P(B) 1/ 2


6.121 Define events: A = purchase extended warranty, B = regular price a P(A | B) =

P(A and B) .21 .21    .2692 P(B) .21  .57 .78

b P(A) = .21 + .14 = .35 c No, because P(A)  P(A | B) 6.122 Define events: A = company fail, B = predict bankruptcy

P(B) = .068 + .2392 = .3072 P(A | B) =

P(A and B) .068   .2214 P(B) .3072

6.123 Define events: A = job security is an important issue, B = pension benefits is an important issue P(A) = .74, P(B) = .65, P(A | B) = .60 a P(A and B) = P(B)P(A | B) = (.65)(.60) = .39 b P(A or B) = .74 + .65 – .39 = 1 6.124 Probabilities of outcomes: P(HH) = .25, P(HT) = .25, P(TH) = .25, P(TT) = .25 P(TT | HH is not possible) = .25/(.25 + .25 + .25) = .333 6.125 P(T) = .5 Case 6.1 1. P(Curtain A) = 1/3, P(Curtain B) = 1/3 2. P(Curtain A) = 1/3, P(Curtain B) = 2/3 Switch to Curtain B and double your probability of winning the car.


Case 6.2 Probability

Bases

Probability

Joint

of outcome

Occupied

Outs

of scoring

Probability

1

.75

2nd

1

.42

.3150

2

.10

1st

1

.26

.0260

3

.10

none

2

.07

.0070

4

.05

1st and 2nd

0

.59

.0295

Outcome

P(scoring) = .3775 Because the probability of scoring with a runner on first base with no outs (.39) is greater than the probability of scoring after bunting (.3775) you should not bunt. Case 6.3 0 outs: Probability of scoring any runs from first base = .39 Probability of scoring from second base = probability of successful steal  probability of scoring any runs from second base = (.68)(.57) = .3876 Decision: Do not attempt to steal. 1 out: Probability of scoring any runs from first base = .26 Probability of scoring from second base = probability of successful steal  probability of scoring any runs from second base = (.68)  (.42) = .2856 Decision: Attempt to steal. 2 outs: Probability of scoring any runs from first base = .13 Probability of scoring from second base = probability of successful steal  probability of scoring any runs from second base = (.68)  (.24) = .1632 Decision: Attempt to steal.


Case 6.4

Age 25: P(D) = 1/1,300 P(D and PT) = (1/1,300)(.624) = .00048 P(D and NT) = (1/1,300)(.376) = .00029 P( D C and PT) = (1,299/1,300)(.04) = .03997 P( D C and NT) = (1,299/1,300)(.96) = .95926 P(PT) = .00048 + .03997 = .04045 P(NT) = .00029 + .95926 = .95955 P(D | PT) = .00048/.04045 = .01187 P(D | NT) = .00029/.95955 = .00030 Age 30: P(D) = 1/900 P(D and PT) = (1/900)(.710) = .00079 P(D and NT) = (1/900)(.290) = .00032 P( D C and PT) = (899/900)(.082) = .08190 P( D C and NT) = (899/900)(.918) = .91698 P(PT) = .00079 + .08190 = .08269 P(NT) = .00032 + .91698 = .91730 P(D | PT) = .00079/.08269 = .00955 P(D | NT) = .00032/.91730 = .00035 Age 35: P(D) = 1/350 P(D and PT) = (1/350)(.731) = .00209


P(D and NT) = (1/350)(.269) = .00077 P( D C and PT) = (349/350)(.178) = .17749 P( D C and NT) = (349/350)(.822) = .81965 P(PT) = .00209 + .17749 = .17958 P(NT) = .00077 + .81965 = .82042 P(D | PT) = .00209/.17958 = .01163 P(D | NT) = .00077/.82042 = .00094 Age 40: P(D) = 1/100 P(D and PT) = (1/100)(.971) = .00971 P(D and NT) = (1/100)(.029) = .00029 P( D C and PT) = (99/100)(.343) = .33957 P( D C and NT) = (99/100)(.657) = .65043 P(PT) = .00971 + .33957 = .34928 P(NT) = .00029 + .65043 = .65072 P(D | PT) = .00971/.34928 = .02780 P(D | NT) = .00029/.65072 = .00045 Age 45: P(D) = 1/25 P(D and PT) = (1/25)(.971) = .03884 P(D and NT) = (1/25)(.029) = .00116 P( D C and PT) = (24/25)(.343) = .32928 P( D C and NT) = (24/25)(.657) = .63072 P(PT) = .03884 + .32928 = .36812 P(NT) = .00116 + .63072 = .63188 P(D | PT) = .03884/.36812 = .10551 P(D | NT) = .00116/.63188 = .00184 Age 49: P(D) = 1/12 P(D and PT) = (1/12)(.971) = .08092 P(D and NT) = (1/12)(.029) = .00242 P( D C and PT) = (11/12)(.343) = .31442 P( D C and NT) = (11/12)(.657) = .60255 P(PT) = .08092 + .31442 = .39533 P(NT) = .00242 + .60255 = .60467 P(D | PT) = .08092/.39533 = .20468 P(D | NT) = .00242/.60467 = .00400


Case 6.5 The probability that 23 people have different birthdays is .4927. The probability that at least two people have the same birthday is 1 − .4927 = .5073.


Chapter 7 7.1 a 0, 1, 2, … b Yes, we can identify the first value (0), the second (1), and so on. c It is finite, because the number of cars is finite. d The variable is discrete because it is countable. 7.2 a any value between 0 and several hundred miles b No, because we cannot identify the second value or any other value larger than 0. c No, uncountable means infinite. d The variable is continuous. 7.3 a The values in cents are 0 ,1 ,2, … b Yes, because we can identify the first ,second, etc. c Yes, it is finite because students cannot earn an infinite amount of money. d Technically, the variable is discrete. 7.4 a 0, 1, 2, …, 100 b Yes. c Yes, there are 101 values. d The variable is discrete because it is countable. 7.5 a No the sum of probabilities is not equal to 1. b Yes, because the probabilities lie between 0 and 1 and sum to 1. c No, because the probabilities do not sum to 1. 7.6 P(x) = 1/6 for x = 1, 2, 3, 4, 5, and 6 7.7 a

x 0 1 2 3 4 5

P(x) 1218/101,501 = .012 32,379/101,501 = .319 37,961/101,501 = .374 19,387/101,501 = .191 7714/101,501 = .076 2842/101,501 = .028

b (i) P(X  2) = P(0) + P(1) + P(2) = .012 + .319 + .374 = .705 (ii) P(X > 2) = P(3) + P(4) + P(5) = .191 + .076 + .028 = .295 (iii) P(X  4) = P(4) + P(5) = .076 + .028 = .104

187


7.8 a P(2  X  5) = P(2) + P(3) + P(4) + P(5) = .310 + .340 + .220 + .080 = .950 P(X > 5) = P(6) + P(7) = .019 + .001 = .020 P(X < 4) = P(0) + P(1) + P(2) + P(3) = .005 + .025 + .310 + .340 = .680 b. b. E(X) =

 xP (x) = 0(.005) + 1(.025) + 2(.310) + 4(.340) +5(.080) + 6(.019) + 7(.001) =

3.066 c.  2 = V(X) =

 (x  ) P(x) = (0–3.066) (.005) + (1–3.066) (.025) + (2–3.066) (.310) 2

2

2

2

2

2

2

2

+ (3–3.066) (.340) + (4–3.066) (.220) + (5–3.066) (.080) + (6–3.066) (.019) 2

+ (7–3.066) (.001) = 1.178

=

 2  1.178 = 1.085

7.9 P(0) = P(1) = P(2) = . . . = P(10) = 1/11 = .091 7.10 a P(X > 0) = P(2) + P(6) + P(8) = .3 + .4 + .1 = .8 b P(X  1) = P(2) + P(6) + P(8) = .3 + .4 + .1 = .8 c P(X  2) = P(2) + P(6) + P(8) = .3 + .4 + .1 = .8 d P(2  X  5) = P(2) = .3 7.11a P(3  X  6) = P(3) + P(4) + P(5) + P(6) = .04 + .28+ .42 + .21 = .95 b. P(X > 6) = P(X  7) = P(7) + P(8) = .02 + .02 = .04 c. P(X < 3) = P(X  2) = P(0) + P(1) + P(2) = 0 + 0 + .01 = .01

7.12 P(Losing 6 in a row) = .5 6 = .0156

7.13 a P(X < 2) = P(0) + P(1) = .05 + .43 = .48 b P(X > 1) = P(2) + P(3) = .31 + .21 = .52

188


7.14

a P(HH) = .25 b P(HT) = .25 c P(TH) = .25 d P(TT) = .25 7.15 a P(0 heads) = P(TT) = .25 b P(1 head) = P(HT) + P(TH) = .25 + .25 = .50 c P(2 heads) = P(HH) = .25 d P(at least 1 head) = P(1 head) + P(2 heads) = .50 + .25 = .75 7.16

189


7.17 a P(2 heads) = P(HHT) + P(HTH) + P(THH) = .125 + .125 + .125 = .375 b P(1 heads) = P(HTT) + P(THT) = P(TTH) = .125 + .125 + .125 = .375 c P(at least 1 head) = P(1 head) + P(2 heads) + P(3 heads) = .375 + .375 + .125 = .875 d P(at least 2 heads) = P(2 heads) + P(3 heads) = .375 + .125 = .500

7.18a.  = E(X) =

 2 = V(X) =

 xP (x) = –2(.59) +5(.15) + 7(.25) +8(.01) = 1.40

 (x  ) P(x) = (–2–1.4) (.59) + (5–1.4) (.15) + (7–1.4) (.25) + (8–1.4) (.01) 2

2

2

2

2

= 17.04 b.

x

–2

5

7

8

y

–10

25

35

40

P(y)

.59

.15

.25

.01

 yP( y) = –10(.59) + 25(.15) + 35(.25) + 40(.01) = 7.00 V(Y) =  ( y  ) P( y) = (–10–7.00) (.59) + (25–7.00) (.15) + (35–7.00) (.25)

c. E(Y) =

2

2

2

2

2

+ (40–7.00) (.01) = 426.00 d. E(Y) = E(5X) = 5E(X) = 5(1.4) = 7.00 V(Y) = V(5X) = 5 2 V(X) = 25(17.04) = 426.00.

7.19a  = E(X) =

 2 = V(X) =

 xP (x) = 0(.4) + 1(.3) + 2(.2) + 3(.1) = 1.0

 (x  ) P(x) = (0–1.0) (.4) + (1–1.0) (.3) + (2–1.0) (.2) + (3–1.0) (.1) 2

2

2

2

2

= 1.0

= b.

 2  1.0 = 1.0

x

0

1

2

3

y

2

5

8

11

P(y)

.4

.3

.2

.1

 yP( y) = 2(.4) + 5(.3) + 8(.2) + 11(.1) = 5.0  =V(Y) =  ( y  ) P( y) = (2 – 5) (.4) + (5 – 5) (.3) + (8 – 5) (.2) + (11 – 5) (.1) = 9.0

c. E(Y) = 2

=

2

2

2

 2  9.0 = 3.0

d. E(Y) = E(3X + 2) = 3E(X) + 2 = 3(1) + 2 = 5.0

 2 = V(Y) = V(3X + 2) = V(3X) = 3 2 V(X) = 9(1) = 9.0.

=

 2  9.0 = 3.0

The parameters are identical. 190

2

2


7.20a. P(X  2) = P(2) + P(3) = .4 + .2 = .6

 xP (x) = 0(.1) + 1(.3) + 2(.4) + 3(.2) = 1.7  = V(X) =  ( x  ) P( x ) = (0–1.7) (.1) + (1–1.7) (.3) + (2–1.7) (.4) + (3–1.7) (.2) = .81 b.  = E(X) =

2

2

2

2

2

2

7.21 E(Profit) = E(5X) = 3E(X) = 3(1.7) = 5.1 V(Profit) = V(3X) = 3 2 V(X) = 9(.81) = 7.29 7.22 a P(X > 4) = P(5) + P(6) + P(7) = .20 + .10 + .10 = .40 b P(X  2) = 1– P(X  1) = 1 – P(1) = 1 – .05 = .95

7.23  = E(X) =

 2 = V(X) =

 xP (x) = 1(.05) + 2(.15) + 3(.15) + 4(.25) + 5(.20) + 6(.10) + 7(.10) = 4.1

 (x  ) P(x) = (1–4.1) (.05) + (2–4.1) (.15) + (3–4.1) (.15) + (4–4.1) (.25) 2

2

2

2

2

2

2

2

+ (5–4.1) (.20) + (6–4.1) (.10) + (7–4.1) (.10) = 2.69 7.24 Y = .25X; E(Y) = .25E(X) = .25(4.1) = 1.025 2

V(Y) = V(.25X) = (.25) (2.69) = .168 7.25 a. x

1

2

3

4

5

6

7

y

.25

.50

.75

1.00

1.25

1.50

1.75

P(y)

.05

.15

.15

.25

.20

.10

.10

b. E(Y) =

 yP( y) = .25(.05) + .50(.15) + .75(.15) +1.00(.25) + 1.25(.20) + 1.50(.10) + 1.75(.10)

= 1.025 V(Y) =

 (y  ) P(y) = (.25–1.025) (.05) + (.50–1.025) (.15) + (.75–1.025) (.15) 2

2

2

2

2

2

2

+ (1.00–1.025) (.25) + (1.25–1.025) (.20) + (1.50–1.025) (.10) + (1.75–1.1025) (.10) = .168 c. The answers are identical. 7.26 a P(4) = .06 b P(8) = 0 c P(0) = .35 d P(X  1) = 1 – P(0) = 1 – .35 = .65 7.27 a P(X  20) = P(20) + P(25) + P(30) + P(40) + P(50) + P(75) + P(100) = .08 + .05 + .04 + .04 + .03 + .03 + .01 = .28 191

2


b P(X = 60) = 0 c P(X > 50) = P(75) + P(100) = .03 + .01 = .04 d P(X > 100) = 0 7.28 a P(X = 3) = P(3) = .21 b P(X  5) = P(5) + P(6) + P(7) + P(8) = .12 + .08 + .06 + .05 = .31 c P(5  X  7) = P(5) + P(6) + P(7) = .12 + .08 + .06 = .26 7.29 a P(X > 1) = P(2) + P(3) + P(4) = .17 + .06 + .01 = .24 b P(X = 0) = .45 c P(1  X  3) = P(1) + P(2) + P(3) = .31 + .17 + .06 = .54

7.30  = E(X) =

 2 = V(X) =

 xP (x) = 0(.04) + 1(.19) + 2(.22) + 3(.28) + 4(.12) + 5(.09) + 6(.06) = 2.76

 (x  ) P(x) = (1–2.76) (.04) + (2–2.76) (.19) + (3–2.76) (.28) 2

2

2

2

2

2

2

+ (4–2.76) (.12) + (5–2.76) (.09) + (6–2.76) (.06) = 2.302

=

 2  2.302 = 1.517

7.31 Y = 10X; E(Y) = E(10X) = 10E(X) = 10(2.76) = 27.6 2

V(Y) = V(10X) = 10 V(X) =100(2.302) = 230.2

=

 2  230 .2 = 15.17

7.32  = E(X) =

 xP (x) = 1(.24) + 2(.18) + 3(.13) + 4(.10) + 5(.07) + 6(.04) + 7(.04) + 8(.20) =

3.86

 2 = V(X) =

 (x  ) P(x) = (1–3.86) (.24) + (2–3.86) (.18) + (3–3.86) (.13) + (4–3.86) 2

2

2

2

(.10) 2

2

2

2

+ (5–3.86) (.07) +(6–3.86) (.04) + (7–3.86) (.04) + (8–3.86) (.20) = 6.78

=

 2  6.78 = 2.60

7.33 Revenue = 2.50X; E(Revenue) = E(2.50X) = 2.50E(X) = 2.50(3.86) = 9.65 V(Revenue) = V(2.50X) = 2.50 2 (V(X) = 6.25(6.78) = 42.38

=

 2  42 .38 = 6.51

7.34 E(Value of coin) = 400(.40) + 900(.30) + 100(.30) = 460. Take the $500. 192

2


7.35  = E(X) =

 2 = V(X) =

 xP (x) = 0(.10) + 1(.20) + 2(.25) + 3(.25) + 4(.20) = 2.25

 (x  ) P(x) = (0–2.25) (.10) + (1–2.25) (.20) + (2–2.25) (.25) + (3–2.25) 2

2

2

2

2

(.13) 2

+ (4–2.25) (.20) = 1.59

=

 2  1.59 = 1.26

7.36 E(damage costs) = .01(400) + .02(200) + .10(100) + .87(0) = 18. The owner should pay up to $18 for the device.

7.37 E(X) =

 xP (x) = 1,000,000(1/10,000,000) + 200,000(1/1,000,000) + 50,000(1/500,000)

+ 10,000(1/50,000) + 1,000(1/10,000) = .1 + .2 + .1 + .2 + .1 = .7 Expected payoff = 70 cents.

7.38  = E(X) =

 2 = V(X) =

 xP (x) = 1(.05) + 2(.12) + 3(.20) + 4(.30) + 5(.15) + 6(.10) + 7 (.08) = 4.00

 (x  ) P(x) = (1–4.0) (.05) + (2–4.0) (.12) + (3–4.0) (.20) + (4–4.0) (.30) 2

2

2

2

2

2

2

2

+ (5–4.0) (.15) +(6–4.0) (.10) + (7–4.0) (.08) = 2.40 7.39 Y = .25X; E(Y) = E(.25X) = .25E(X) = .25(4.0) = 1.0 2

V(Y) = V(.25X) = (.25) V(X) =.0625(2.40) = .15

7.40  = E(X) =

 xP (x) = 0(.10) + 1(.25) + 2(.40) + 3(.20) + 4(.05) = 1.85

7.41 Profit = 4X; Expected profit = E(4X) = 4E(X) = 4(1.85) = $7.40 7.42a P(X > 4) = P(5) + P(6) + P(7) + P(8) = .04 + .15 + .03 + .10 = .32 b P(X < 5) = P(1) + P(2) + P(3) + P(4) = .03 + .32 + .05 + .28 = .68 P(4≤ X ≤ 6) = P(4) + P(5) + P(6) = .28 + .04 + .15 = .47

7.43  = E(X) =

 xP (x) = 1(.03) + 2(.32) + 3(.05) + 4(.28) + 5(.04) + 6 (.15) + 7 (.03) + 8(.10)

= 4.05

193


 2 = V(X) =

 (x  ) P(x) = (1–4.05) (.03) + (2–4.05) (.32) + (3–4.05) (.05) + (4–4.05) 2

2

2

2

2

2

2

(.28) + (5–4.05) (.04) +(6–4.05) (.15) + (7–4.05) (.03) + (8–4.05)2 (.10) = 4.11

=

 2  4.11 = 2.03

7.44a P(X > 3) = P(4) + P(5) + P(6) = .02 + .01 + .01 = .04 b P(0) = .78 c P(3 ≤ X ≤ 5) = P(3) + P(4) + P(5) = .03 + .02 + .01 = .06 7.45 a

b

x

P(x)

1

.6

2

.4

y

P(y)

1

.6

2

.4

 xP (x) = 1(.6) + 2(.4) = 1.4  = V(X) =  ( x  ) P( x ) = (1–1.4) (.6) + (2–1.4) (.4) = .24

c  = E(X) =

2

2

2

2

d  = 1.4,  2 = .24

7.46 a

  xyP (x, y) = (1)(1)(.5) + (1)(2)(.1) + (2)(1)(.1) + (2)(2)(.3) = 2.1 all x all y

COV(X, Y) =

  xyP (x, y) –   = 2.1 – (1.4)(1.4) = .14 x

y

all x all y

 x   2x  .24 = .49,  y   2y  .24 = .49



.14 COV (X, Y) = = .58 (.49 )(. 49 ) xy

7.47 E(X + Y) = E(X) + E(Y) = 1.4 + 1.4 = 2.8 V(X + Y) = V(X) + V(Y) + 2COV(X, Y) = .24 + .24 + 2(.14) = .76 7.48 a

x+y

P(x + y)

2

.5

194

2


3

.2

4

.3

 (x  y)P(x  y) = 2(.5) + 3(.2) + 4(.3) = 2.8 ] P( x  y) = (2–2.8) (.5) + (3–2.8) (.2) + (4–2.8) (.3) = = V(X+Y) =  [( x  y)  

b  x y = E(X+Y) =  2x y

xy

2

2

2

.76 c Yes 7.49 a

b

x

P(x)

1

.4

2

.6

y

P(y)

1

.7

2

.3

 xP (x) = 1(.4) + 2(.6) = 1.6  = V(X) =  ( x  ) P( x ) = (1–1.6) (.4) + (2–1.6) (.6) = .24 d  = E(Y) =  yP( y) = 1(.7) + 2(.3) = 1.3  = V(Y) =  ( y  ) P( y) = (1–1.3) (.7) + (2–1.3) (.3) = .21 c  = E(X) = 2

2

2

2

2

2

2

2

7.50 a

  xyP (x, y) = (1)(1)(.28) + (1)(2)(.12) + (2)(1)(.42) + (2)(2)(.18) = 2.08 all x all y

COV(X, Y) =

  xyP (x, y) –   = 2.08 – (1.6)(1.3) = 0 x

y

all x all y

 x   2x  .24 = .49,  y   2y  .21 = .46



0 COV (X, Y) = =0 (.49 )(. 46 ) xy

7.51 E(X + Y) = E(X) + E(Y) = 1.6 + 1.3 = 2.9 V(X + Y) = V(X) + V(Y) + 2COV(X, Y) = .24 + .21 + 2(0) = .45 7.52 a x + y

P(x + y)

2

.28

3

.54

4

.18

b  x y = E(X+Y) =

 (x  y)P(x  y) = 2(.28) + 3(.54) + 4(.18) = 2.9 195

2


 2x y = V(X+Y) =

 [( x  y)  

xy ]

2

2

2

= .45 c Yes 7.53 a x

P(x)

y

P(y)

1

.7

1

.6

2

.2

2

.4

3

.1

 xP (x) = 1(.7) + 2(.2) + 3(.1) = 1.4  = V(X) =  ( x  ) P( x ) = (1–1.4) (.7) + (2–1.4) (.2) + (3–1.4) (.1) = .44  = E(Y) =  yP( y) = 1(.6) + 2(.4) = 1.4  = V(Y) =  ( y  ) P( y) = (1–1.4) (.6) + (2–1.4) (.4) = .24   xyP (x, y) = (1)(1)(.42) + (1)(2)(.28) + (2)(1)(.12) + (2)(2)(.08) + (3)(1)(.06) +

b  x = E(X) = 2

2

2

2

2

2

2

y

2

all x all y

(3)(2)(.04) = 1.96 COV(X, Y) =

  xyP (x, y) –   = 1.94 – (1.4)(1.4) = 0 x

y

all x all y

 x   2x  .44 = .66,  y   2y  .24 = .49



c

0 COV (X, Y) = =0 xy (.66 )(. 49 )

x+y

P(x + y)

2

.42

3

.40

4

.14

5

.04

7.54

7.55

x y

0

1

2

1

.42

.21

.07

2

.18

.09

.03

2

P( x  y) = (2–2.9) (.28) + (3–2.9) (.54) + (4–2.9) (.18)

x

196

2


y

0

1

1

.04

.16

2

.08

.32

3

.08

.32

7.56 a

b

Refrigerators, x

P(x)

0

.22

1

.49

2

.29

Stoves, y

P(y)

0

.34

1

.39

2

.27

 xP (x) = 0(.22) + 1(.49) + 2(.29) = 1.07  = V(X) =  ( x  ) P( x ) = (0–1.07) (.22) + (1–1.07) (.49) + (2–1.07) (.29) = .505 d  = E(Y) =  yP( y) = 0(.34) + 1(.39) + 2(.27) = .93  = V(Y) =  ( y  ) P( y) = (0–.93) (.34) + (1–.93) (.39) + (2–.93) (.27) = .605 e   xyP ( x, y) = (0)(0)(.08) + (0)(1)(.09) + (0)(2)(.05) + (1)(0)(.14) + (1)(1)(.17) c  x = E(X) =

2

2

2

2

2

y

2

2

2

2

all x all y

+ (1)(2)(.18) + (2)(0)(.12) + (2)(1)(.13) + (2)(2)(.04) = .95 COV(X, Y) =

  xyP (x, y) –   = .95 – (1.07)(.93) = –.045 x

y

all x all y

 x   2x  .505 = .711,  y   2y  .605 = .778



.045 COV (X, Y) = = –.081 xy (.711)(. 778 )

7.57 a

Bottles, x

P(x)

0

.72

1

.28

Cartons, y

P(y)

0

.81

1

.19

b

 xP (x) = 0(.72) + 1(.28) = .28  = V(X) =  ( x  ) P( x ) = (0–.28) (.72) + (1–.28) (.28) = .202

c  x = E(X) = 2

2

2

2

197

2


 yP( y) = 0(.81) + 1(.19) = .19  = V(Y) =  ( y  ) P( y) = (0–.19) (.81) + (1–.19) (.19) = .154 e   xyP ( x, y) = (0)(0)(.63) + (0)(1)(.09) + (1)(0)(.18) + (1)(1)(.10) = .100

d  y = E(Y) =

2

2

2

2

all x all y

COV(X, Y) =

  xyP (x, y) –   = .100 – (.28)(.19) = .0468 x

y

all _ x all _ y

 x   2x  .202 = .449,  y   2y  .154 = .392



.0468 COV (X, Y) = = .266 (.449 )(. 392 ) xy

7.58 a P(X = 1 | Y = 0) = P(X =1 and Y = 0)/P(Y = 0) = .14/.34 = .412 b P(Y = 0 | X = 1) = P(X =1 and Y = 0)/P(X = 1) = .14/.49 = .286 c P(X = 2 | Y = 2) = P(X =2 and Y = 2)/P(Y = 2) = .04/.27 = .148 7.59 a P(CMD = 0 and SD = 2) = .06 b P(CMD = 2 and SD = 0) = 0 c P(CMD ≥ 1 and SD ≥ 1) = .07 + .01 + .10 + .05 + .04 + .02 = .29 7.60 a P(CMD = 1 and SD = 2) = .10 b. P(SD = 2 | CMD = 1) = .10/.24 = .4167 c P(CMD = 1 | SD = 2) = .10/.21 = .4762 7.61 a P(CMD = 0) = .68, P(CMD = 1) = .24, P(CMD = 2) = .08

 xP (x) = 0(.68) + 1(.24) + 2(.08) = .40  = V(X) =  ( x  ) P( x ) = (0–.40) (.68) + (1–.40) (.24) + (2-.40) (.08) = .40

b  = E(X) =

2

2

=

2

2

2

 2  .40 = .632

7.62a. P(SD = 0) = .45, P(SD=1) = .23, P(SD = 2) = .21, P(SD = 3) = .11

 xP (x) = 0(.45) + 1(.23) + 2(.21) + 3(.11) = .98  = V(X) =  ( x  ) P( x ) = (0–.98) (.45) + (1–.98) (.23) + (2-.98) (.21) + (3-.98) (.11)

b.  = E(X) =

2

2

2

2

= 1.100

=

 2  1.100 = 1.049

198

2

2


7.63a P(Home Team > Visiting Team) = .11 + .09 + .10 + .05 + .02 + .01 = .38 b P(Tie) = .14 + .10 + .04 + 0 = .28 c P(Home Team < Visiting Team) = .12 + .09 + .03 + .07 + .02 + .01 = .34 7.64a. P(HT = 0) = .38, P(HT = 1) =.30, P(HT = 2) = .19, P(HT = 3) = .13

 xP (x) = 0(.38) + 1(.30) + 2(.19) + 3(.12) = 1.07  = V(X) =  ( x  ) P( x ) = (0–1.07) (.38) + (1–1.07) (.30) + (2-1.07) (.19) + (3-1.07) (.12)

b.  = E(X) =

2

2

2

2

2

2

= 1.085

 2  1.085 = 1.042

=

7.65a. P(VT = 0) = .44, P(VT = 1) = .29, P(VT = 2) = .21, P(VT = 3) = .06

 xP (x) = 0(.44) + 1(.29) + 2(.21) + 3(.06) = .89  = V(X) =  ( x  ) P( x ) = (0–.89) (.44) + (1–.89) (.29) + (2-.89) (.21) + (3-.89) (.06)

b.  = E(X) =

2

2

2

2

2

2

= .878

=

 2  .878 = .937

7.66 a. P(T = 0) = .14, P(T = 1) = .23, P(T = 2) = .28, P(T = 3) = .25, P(T = 4) = .08, P(T = 5) = .02, P(T = 6) = 0

 xP (x) = 0(.14) + 1(.23) + 2(.28) + 3(.25) + 4(.08) + 5(.02) + 6(0) = 1.96  = V(X) =  ( x  ) P( x ) = (0–1.96) (.14) + (1–1.96) (.23) + (2-1.96) (.28) + (3-1.96) (25) +

b.  = E(X) =

2

2

2

2

2

2

(4-1.96)2(.08) + (5-1.96)2(.02) + (6-1.96)2(0) = 1.538

= c.

 2  1.538 = 1.240

  xyP (x, y) = (0)(0)(.14) + (0)(1)(.12) + (0)(2)(.09) + (0)(3)(.03) + (1)(0)(.11) + all x all y

(1)(1)(.10) + (1)(2)(.07) + (1)(3)(.02) + (2)(0)(.09) + (2)(1)(.05) + (2)(2)(.04) + (2)(3)(.01) + (3)(0)(.10) + (3)(1)(.02) + (3)(2)(.01) + (3)(3)(0) = .74

COV(X, Y) =

  xyP (x, y) –   = .74 – (1.07)(.89) = -.212 x

y

all x all y



COV ( X , Y )

7.67 E

 x y

=

.212 = –.217 (1.085 )(. 937 )

 X    E(X ) = 18 + 12 + 27 + 8 = 65 i

i

199


V

 X    V(X ) = 8 + 5 + 6 + 2 = 21 i

i

 X    E(X ) = 35 + 20 + 20 + 50 + 20 = 145 V X    V(X ) = 8 + 5 + 4 + 12 + 2 = 31

7.68 E

i

i

i

i

 X    E(X ) = 8 + 14 + 5 + 3 + 30 + 30 + 10 = 100 V X    V(X ) = 2 + 5 + 1 + 1 + 8 +10 + 3 = 30

7.69 E

i

i

i

i

 X    E(X ) = 10 + 3 + 30 + 5 + 100 + 20 = 168 V X    V(X ) = 9 + 0 + 100 + 1 + 400 + 64 = 574

7.70 E

i

i

i

i

7.71 The expected value does not change. The standard deviation decreases. 7.2 E(Rp) = w1E(R1) + w2E(R2) = (.30)(.12) + (.70)(.25) = .2110 a. V(Rp) =

w12 12 + w22  22 + 2 w1 w2  1  2

= (.30)2 (.02)2  (.70)2 (.152 )  2(.30)(.70)(.5)(.02)(.15) = .0117  R p  .0117 = .1081

b. V(Rp) =

w12 12 + w22  22 + 2 w1 w2  1  2

= (.30) 2 (.02) 2  (.70) 2 (.152 )  2(.30)(.70)(.2)(.02)(.15) = .0113  R p  .0113 = .1064

c. V(Rp) =

w12 12 + w22  22 + 2 w1 w2  1  2

= (.30) 2 (.02) 2  (.70) 2 (.152 )  2(.30)(.70)(0)(.02)(.15) = .0111  R p  .0111 = .1052

7.73 a She should choose stock 2 because its expected value is higher. b. She should choose stock 1 because its standard deviation is smaller. 7.74 E(Rp) = w1E(R1) + w2E(R2) = (.60)(.09) + (.40)(.13) = .1060 V(Rp) =

w12 12 + w22  22 + 2 w1 w2  1  2 = (.60) 2 (.15) 2  (.40) 2 (.212 )  2(.60)(.40)(.4)(.15)(.21) = .0212

200


 R p  .0212 = .1456

7.75 E(Rp) = w1E(R1) + w2E(R2) = (.30)(.09) + (.70)(.13) = .1180 V(Rp) =

w12 12 + w22  22 + 2 w1 w2  1  2 = (.30) 2 (.15) 2  (.70) 2 (.212 )  2(.30)(.70)(.4)(.15)(.21) = .0289

 R p  .0289 = .1700

The statistics used in Exercises 7.66 to 7.83 were computed by Excel. The variances were taken from the variance-covariance matrix. As a result they are the population parameters. To convert to statistics multiply the variance of the portfolio returns by n/(n–1).

7.76 a .0074 .0667 b .0056 .0675 c .0064 .0590 d (a) e (b)

7.77 a .0158 .0425 b .0153 .0461 c .0143 .0439 d (a) has the largest mean and the smallest standard deviation. 7.78 a .0080 .0394 b .0084 .0397 c .0059 .0437 d (c) has the smallest mean and the largest standard deviation. 7.79 a .0016 .0327 b .0120 .0307 c .0121 .0315 d (c) e (b) 7.80 a .0101 .0338 b .0097 .0339 201


c .0110 .0360 d (c) e (a) 7.81 a .0080 .0448 b .0062 .0472 c .0098 .0428 d (c) has the largest mean and the smallest standard deviation. 7.82 a .0103 .0276 b .0085 .0282 c .0120 .0315 d (c) e (a) 7.84 .0100 .0267 7.85 Because all the stocks are bank stocks there is very little diversification. 7.86

7.87 a .0131 .0364 b .0118 .0424 c .0127 .0531 d (a) has the largest mean and the smallest standard deviation. 7.88 a .0119 .0350 b .0135 .0326 c .0142 .0372 d (c) e (b)

202


7.89 a .0096 .0350 b .0115 .0536 c .0093 .0357 d (b) e (a) 7.90 a .0065 .0277 b .0096 .0303 c .0047 .0371 d (b) e (a) 7.91 a .0077 .0362 b .0102 .0341 c .0086 .0324 d (b) e (c) 7.93 .0100 .0315 7.94 a .0128 .0456 b .0161 .0491 c .0117 .0437 d (b) e (c) 7.95 a .0154 .0464 b .0172 .0465 c .0165 .0519 d (b) e (a) 7.96 a .0186 .0468 b .0185 .0592 c .0213 .0752 d (c) e (a)

203


7.97 a .0147 .0454 b .0212 .0463 c .0119 .0482 d (b) e (a) 7.99 .0200 .0444

7.100 P(X = x) =

n! p x (1  p) n x x! (n  x )!

a P(X = 3) =

10! (.3) 3 (1  .3)103 = .2668 3! (10  3)!

b P(X = 5) =

10! (.3) 5 (1  .3)105 = .1029 5! (10  5)!

c P(X = 8) =

10! (.3) 8 (1  .3)108 = .0014 8!(10  8)!

7.101 a P(X = 3) = P(X  3) – P(X  2) = .6496 – .3828 = .2668 b P(X = 5) = P(X  5) – P(X  4) = .9527 – .8497 = .1030 c P(X = 8) = P(X  8) – P(X  7) = .9999 – .9984 = .0015 7.102 a .26683 b .10292 c .00145

7.103 P(X = x) =

n! p x (1  p) n x x! (n  x )!

a. P(X = 2) =

6! (.2) 2 (1  .2) 62 = .2458 2! (6  2)!

b. P(X = 3) =

6! (.2) 3 (1  .2) 63 = .0819 3! (6  3)!

c. P(X = 5) =

6! (.2) 5 (1  .2)5 = .0015 5! (6  5)!

7.104 a P(X = 2) = P(X  2) – P(X  1) = .9011 – .6554 = .2457 b P(X = 3) = P(X  3) – P(X  2) = .9830 −.9011 = .0819 204


c P(X = 5) = P(X  5) – P(X  4) = .9999 – 9984 = .0015 7.105 a .24576 b .08192 c .00154 7.106 a P(X = 18) = P(X  18) – P(X  17) = .6593 – .4882 = .1711 b P(X = 15) = P(X  15) – P(X  14) =.1894 – .0978 = .0916 c P(X  20) = .9095 d P(X  16) = 1 – P(X  15) = 1 – .1894 = .8106 7.107 a .17119 b .09164 c .90953 d .81056 7.108 Binomial distribution with p = .25 a P(X = 1) =

4! (.25)1(1  .25)41 = .4219 1! (4  1)!

b Table 1 with n = 8: p(2) = P(X  2) – P(X  1) = .6785 – .3671 = .3114 c Excel: p(3) = .25810 7.109 Table 1 with n = 25 and p = .3: P(X  10) = .9022 7.110 Table 1 with n = 25 and p = .90 a P(X = 20) = P(X  20) – P(X  19) = .0980 – .0334 = .0646 b P(X  20) = 1 – P(X  19) = 1 – .0334 = .9666 c P(X  24) = .9282 d E(X) = np = 25(.90) = 22.5 7.111 Table 1 with n = 25 and p = .75: P(X  15) = 1 – P(X  14) = 1 – .0297 = .9703

7.112 P(X = 0) = P(X = 1) =

4! (.7) 0 (1  .7) 40 = .0081 0!( 4  0)!

4! (.7)1 (1  .7) 41 = .0756 1!( 4  1)!

205


P(X = 2) =

4! (.7) 2 (1  .7) 42 = .2646 2!( 4  2)!

P(X = 3) =

4! (.7)3 (1  .7) 43 = .4116 3!(4  3)!

P(X = 4) =

4! (.7) 4 (1  .7) 44 = .2401 4!( 4  4)!

7.113 Excel with n = 5 and p = .27: a. P(X ≤ 2) = .8743 b P(X = 0) = .2073 c P(X ≥ 3) = 1 – P(≤ 2) = 1-.8743 = .1257 7.114 Excel with n = 10 and p = .30 a P(X = 3) = .2668 b .P(X ≥ 3) = 1 – P(X ≤ 2) = 1 - .3828 = .6172 7.115 Excel with n = 10 and p = .25 a. P(X ≤ 3) = .7759 b. E(X) = np = 100(.25) = 25

7.116 P(X = 20) =

20! (.75) 20 (1  .75) 2020 = .00317 20! (20  20 )!

7.117 Excel with n = 4 and p = .72 a. P(X = 4) = .2687 b. P(X = 2) = .2439 c E(X) = np = 4(.72) = 2.88 7.118 Excel with n = 10 and p = .29 a. P(X ≤ 3) = .6761 b. P(X = 0) = .0326 c. E(X) = np = 10(.29) = 2.9 7.119a Excel with n = 10 and p = 244/495: P(X  5) = 1 – P(X  4) = 1 – .39447 = .60553 b E(X) = np =100(244/495) = 49.29

206


7.120 a P(X = 2) =

5! (.45) 2 (1  .45) 52 = .3369 2! (5  2)!

b Excel with n = 25 and p = .45: P(X  10) = 1 – P(X  9) = 1 – .24237 = .75763 7.121 a Table 1 with n = 5 and p = .5: P(X = 2) = P(X  2) – P(X  1) = .5 – .1875 = .3125 b: Table 1 with n = 25 and p = .5: P(X  10) = 1 – P(X  9) = 1 – .1148 = .8852

7.122 a P(X = 2) =

5! (.52)2 (1  .52)52 = .2990 2!(5  2)!

b Excel with n = 25 and p = .52: P(X  10) = 1 – P(X  9) = 1 – .08033 = .91967 7.123 a Excel with n = 25 and p = 2/38: P(X  2) = 1 – P(X  1) = 1 – .61826 = .38174 b Excel with n = 25 and p = 2/38: P(X = 0)) = .25880 c Excel with n = 25 and p = 18/38: P(X  15) = 1 – P(X  14) = 1 – .85645 = .14355 d Excel with n = 25 and p = 18/38: P(X  10) = .29680 7.124 a Excel with n = 100 and p = .52: P(X  50) = 1 – P(X  49) = 1 – .30815 = .69185 b Excel with n = 100 and p = .36: P(X  30) = .12519 c Excel with n = 100 and p = .06: P(X  5) = .44069 7.125 Excel with n = 20 and p = .38: P(X  10) = 1 – P(X  9) = 1 – .81032 = .18968 7.126a. Excel with n = 10 and p = .23: P(X  5) = 1 – P(X  4) = 1 – .94308 = .05692 b. Excel with n = 25 and p = .23: P(X  5) = .47015 7.127 Excel with n = 100 and p = .53 a P(X > 50) = 1 - P(X ≤ 50) = 1 - .3078 = .6922 b p(X > 60) = 1 – P(X ≤ 60) = 1 - .9341 = .0659 cE(X) = np = 100(.53) = 53

7.128 a P(X = 0) =

e 2 2 0 e   x = = .1353 0! x!

b P(X = 3) =

e 2 2 3 e   x = = .1804 3! x!

c P(X = 5) =

e 2 2 5 e   x = = .0361 5! x! 207


7.129a P(X = 0) =

e   x e .5 .5 0 = = .6065 x! 0!

b P(X = 1) =

e .5 .51 e   x = = .3033 1! x!

c P(X = 2) =

e .5 .5 2 e   x = = .0758 2! x!

7.130 a Table 2 with  = 3.5: P(X = 0) = P(X  0) = .0302 b Table 2 with  = 3.5: P(X  5) = 1 – P(X  4) = 1 – .7254 = .2746 c Table 2 with  = 3.5/7: P(X = 1) = P(X  1) – P(X  0) = .9098 – .6065 = .3033

7.131 a P(X = 5 with  = 14/3) = b. P(X = 1 with  = 14/3) =

e e   x = x!

7.132 a P(X = 0 with  = 2) = b P(X = 10 with  = 14) =

e e   x = x! 1 / 3

14 / 3

(14 / 3) 5 = .1734 5!

(1 / 3)1 = .2388 1!

e   x e 2 (2) 0 = = .1353 0! x!

e   x e 14 (14)10 = = .0663 10! x!

7.133 a Table 2 with  = 5: P(X  10) = 1 – P(X  9) = 1 – .9682 = .0318 b Table 2 with  = 10: P(X  20) = 1 – P(X  19) = 1 – .9965 = .0035 7.134 a Excel with  = 30: P(X  35) = 1 – P(X  34) = 1 – .79731 = .20269 b Excel with  = 15:P(X  12 = .26761 7.135 a Excel with  = 1.8: P(X  3) = 1 – P(X  2) = 1 – .73062 = .26938 b Table 2 with  = 9: P(10  X  15) = P(X  15) – P(X  9) = .9780 – .5874 = .3906

7.136 P(X = 0 with  = 80/200) =

e   x e .4 (.4) 0 = =.6703 0! x!

7.137 Poisson with µ = 2.5 a P(X ≥ 5) = 1 – P(X ≤ 4) = 1 - .8912 = .1088 b P(0) = .0821 208


c P(X ≤ 3) =.7576 7.138 Poisson with µ = 4 a P(X ≥ 5) = 1 – P(X ≤ 4) = 1 - .6288 = .3712 b P(X ≤ 3) = .4335 c P(X = 4) = .1954 7.139 Poisson with µ = 5 a P(X ≤ 3) = .2650 b P(X ≥ 6) = 1 – P(X ≤ 5) = 1 – .6160 = .3840 c P(X ≤ 8) = .9319 7.140 Poisson with µ = 1 a P(X = 0) = .3679 b P(X ≥ 5) = 1 – P(X ≤ 4) = 1 - .9963 = .0037 7.141 Poisson with µ = 2 a P(X = 0) = .1353 b P(X ≥ 4) = 1 – P(X ≤ 3) = 1 – .8571 = . 1429 c P(X ≤ 2) = .6767 7.142 a Table 2 with  = 1.5: P(X  2) = 1 – P(X  1) = 1 – .5578 = .4422 b Table 2  = 6: P(X < 4) = P(X  3) = .1512

7.143 a P(X = 1 with  = 5) =

e   x e 5 (5)1 = = .0337 1! x!

b Table 2 with  = 15: P(X > 20) = P(X  21) = 1 – P(X  20) = 1 – .9170 = .0830

7.144 a P(X = 0 with  = 1.5) =

e   x e 1.5 (1.5) 0 = = .2231 0! x!

b Table 2 with  = 4.5: P(X  5) = .7029 c Table 2 with  = 3.0: P(X  3) = 1 – P(X  2 = 1 – .4232 = .5768

7.145 Binomial with n = 25 and p = .20 a P(X ≤ 5) = .6167 b P(X ≥ 7) = 1 – P(X ≤ 6) = 1 – .7800 = .2200 c P(X = 5) = .1960 209


7.146 Binomial with n = 20 and p = .15 a P(X = 3) = .2428 b P(X ≤ 5) = .9327 c P(X ≥ 3) = 1 – P(X ≤ 2) = 1 - .4049 = .5951

7.147 P(X = 5) =

5! (.774) 5 (1  .774) 55 = .2778 5! (5  5)!

7.148 a E(X) = np = 40(.02) = .8 b P(X = 0) =

40! (.02)0 (1  .02)400 = .4457 0! (40  0)!

 xP (x) = 0(.48) + 1(.35) + 2(.08) + 3(.05) + 4(.04) = .82

7.149 a  = E(X) =

 2 = V(X) =

 (x  ) P(x) = (0–.82) (.48) + (1–.82) (.35) + (2–.82) (.08) 2

2

2

2

2

2

+ (3–.82) (.05) + (4–.82) (.04) = 1.0876

=

 2  1.0876 = 1.0429

7.150 a P(X = 10 with  = 8) =

e   x e 8 (8)10 = = .0993 10! x!

b Table 2 with  = 8: P(X > 5) = P(X  6) = 1 – P(X  5) = 1 – .1912 = .8088 c Table 2 with  = 8: P(X < 12) = P(X  11) = .8881

7.151 Binomial with n = 10 and p = .80 a P(X ≥ 9) = 1 – P(X ≤ 8) = 1 – .6242 = .3758 b P(X = 8) = .3020 c P(X ≤ 7) = .3222 7.152 Table 1 with n = 10 and p = .3: P(X > 5) = P(X  6) = 1 – P(X  5) = 1 – .9527 = .0473

7.153 a  = E(X) =

 2 = V(X) =

 xP (x) = 0(.05) + 1(.16) + 2(.41) + 3(.27) + 4(.07) + 5(.04) = 2.27

 (x  ) P(x) = (0–2.27) (.05) + (1–2.27) (.16) + (2–2.27) (.41) 2

2

2

2

2

2

+ (3–2.27) (.27) + (4–2.27) (.07) + (5–2.27) (.04) = 1.1971

=

 2  1.1971 = 1.0941

210

2


7.154 Poisson with µ = 5 a P(X = 1) = .0337 b P(X ≤ 5) = .6160 c P(X ≥ 8) = 1 – P(X ≤ 7) = 1 - .8666 = .1334 7.155 Binomial with n = 25 and p = .10 a P(X ≥ 3) = 1 – P(X ≤ 2) = 1 - .5371 = .4629 b P(X = 2) = .2659 c P(X ≤ 3) = .7636 7.156 Binomial with n = 10 and p = .50 a P(X = 6) = .2051 bP(X ≥ 8) = 1 – P(X ≤ 7) = 1 – .9453 = .0547 c P(X ≤ 4) = .3770 7.157 Poisson with µ = 1.5 a P(X = 1) = .3347 b P(X ≥ 3) = 1 – P(X ≤ 2) = 1 – .8088 = .1912 c P(X ≤ 4) = .9814 7.158 Table 1 with n = 10 and p = .20: P(X  6) = 1 – P(X  5) = 1 – .9936 = .0064

7.159 a P(X = 2) =

10! (.05) 2 (1  .05)10 2 = .0746 2!(10  2)!

b Excel with n = 400 and p = .05: P(X = 25) = .04455 c .05 7.160 a Excel with n = 80 and p = .70: P(X > 65) = P(X  66) = 1 – P(X  65) = 1 – .99207 = .00793 b E(X) = np = 80(.70) = 56 c =

np(1  p)  80 (.70 )(1  .70 ) = 4.10

7.161 a Excel with  = 9.6: P(X ≥ 10) = 1 – P(X ≤ 9) = 1 − .5089 = .4911 b. Excel with µ = 6: P(X ≤ 5) = .4457 c Excel with  = 2.3: P(X ≥ 3) = 1 – P(X ≤ 2) = 1 − .5960 = .4040

211


7.162 Table 1 with n = 25 and p = .40: a P(X = 10) = P(X  10) – P(X  9) = .5858 – .4246 = .1612 b P(X < 5) = P(X  4) = .0095 c P(X > 15) = P(X  16) = 1 – P(X  15) = 1 – .9868 = .0132 7.163 Excel with n = 100 and p = .45: a P(X > 50) = P(X  49) = 1 – P(X  50) = 1 – .86542 = .13458 b P(X < 44) = P(X  43) = .38277 c P(X = 45) = .07999

 xP (x) = 0(.36) + 1(.22) + 2(.20) + 3(.09) + 4(.08) + 5(.05) = 1.46

7.164 a.  = E(X) =

 2 = V(X) =

 (x  ) P(x) = (0–1.46) (.36) + (1–1.46) (.22) + (2–1.46) (.20) 2

2

2

2

2

2

2

+ (3–1.46) (.09) + (4–1.46) (.08) + (5–1.46) (.05) = 2.23

=

 2  2.23 = 1.49

 xP (x) = 0(.15) + 1(.18) + 2(.23) + 3(.26) + 4(.10) + 5(.08) = 2.22  = V(X) =  ( x  ) P( x ) = (0–2.22) (.15) + (1–2.22) (.18) + (2–2.22) (.23)

b.  = E(X) =

2

2

2

2

2

2

2

2

+ (3–2.22) (.26) + (4–2.22) (.10) + (5–2.22) (.08) = 2.11

=

 2  2.11 = 1.45

7.165a Binomial with n = 5 and p = .350: P(X ≥ 1) = 1 – P(0) = 1 - .1160 = .8840 b Probability of at least 1 hit in 56 consecutive games = (.8840) 56 = .00100 7.166 Binomial with n = 10 and p = .80 a P(X = 10) = .1074 b P(X ≥ 8) = 1 – P(X ≤ 7) = 1 - .3222 = .6778 c P(X ≤ 8) = .6242 7.167 Binomial with n = 20 and p = .60 a P(X ≥ 15) = 1 – P(X ≤ 14) = 1 – .8744 = .1256 b P(X ≤ 12) = .5841 c P(X = 12) = .1797

7.168 Binomial with n = 5 and p = .01. (using Excel)

212


x

p(x)

0

.95099

1

.04803

2

.00097

3

.00001

4

0

5

0

7.169 Binomial with n = 100 and p = .01 a P(X = 0) = .3660 b P(X = 1) = .3697 c P(X = 2) = .1849 7.170 p = .08755 because P(X  1) = 1– P(X = 0 with n = 10 and p = .08755) = 1– .40 = .60 Case 7.1 Expected number of runs without bunting = .85. If batter bunts: Bases Outcome

Expected Number

Probability

Occupied

Outs

of Runs

1

.75

2nd

1

.69

.5175

2

.10

1st

1

.52

.0520

3

.10

none

2

.10

.0100

4

.05

1st and 2nd

0

1.46

.0730

Expected number of runs = .6255 Decision: Don’t bunt.

213


Chapter 8 8.1a. P(X > 45) 

(60  45)  2 (75  60 )  2  = .0800 50 15 50 15

b. P(10 < X < 40)  c. P(X < 25) 

(15  [30 ])  6 (0  [15])  10 (15  0)  17 (25  15)  7    = .7533 50 15 50 15 50  15 50 15

d. P(35 < X < 65) 

8.2a. P(X > 45) 

(45  35)  6 (60  45 )  2 (65  60 )  2   = .1333 50 15 50  15 50  15

(60  45 )  3 (75  60 )  3  = .1200 50  15 50  15

b. P(10 < X < 40)  c. P(X < 25) 

(15  10 )  17 (30  15 )  7 (40  30 )  6   = .3333 50  15 50  15 50  15

(15  10 )  16 (30  15 )  8 (40  30 )  8   = .4800 50  15 50 15 50 15

(30  [45 ])  5 (15  [30 ])  5 (0  [15 ])  2 (15  0)  16 (25  15 )  8     50 15 50 15 50  15 50  15 50  15

= .6667 d. P(35 < X < 65) 

(45  35 )  8 (60  45 )  3 (65  60 )  3   = .1867 50  15 50  15 50  15

8.3a. P(55 < X < 80) 

(60  55 )  16 (70  60 )  5 (80  70 )  24   = .6167 60  10 60  10 60  10

b. P(X > 65) 

(70  65 )  5 (80  70 )  24 (90  80 )  7 (100  90 )  1    = .5750 60  10 60  10 60  10 60  10

c. P(X < 85) 

(50  40 )  7 (60  50 )  16 (70  60 )  5 (80  70 )  24 (85  80 )  7     60  10 60  10 60  10 60  10 60  10

= .9250 d. P(75 < X < 85) 

(80  75 )  24 (85  80 )  7  = .2583 60  10 60  10

215


8.4 a.

b. P(X > 25) = 0 c. P(10 < X < 15) = (15  10 )

1 = .25 20

d. P(5.0 < X < 5.1) = (5.1  5)

8.5a. f(x) =

1 = .005 20

1 1 = (60  20 ) 40

b. P(35 < X < 45) = (45–35)

20 < x < 60

1 = .25 40

216


c.

8.6 f(x) =

1 1  30 < x < 60 (60  30 ) 30

a. P(X > 55) = (60  55 )

1 = .1667 30

b. P(30 < X < 40) = (40  30 )

1 = .3333 30

c. P(X = 37.23) = 0

8.7

1  (60  30 )  7.5 ; The first quartile = 30 + 7.5 = 37.5 minutes 4

8.8 .10  (60  30)  3 ; The top decile = 60–3 = 57 minutes

8.9 f(x) =

1 1  (175  110 ) 65

a. P(X > 150) = (175  150 )

110 < x < 175

1 = .3846 65

b. P(120 < X < 160) = (160  120 )

1 = .6154 65

8.10 .20(175–110) = 13. Bottom 20% lie below (110 + 13) = 123

Note: The area in a triangle is height X base /2

8.11a.

217


b. P(0 < X < 2) = (.5)(2–0)(1) = 1.0 c. P(X > 1) = (.5)(2 – 1)(.5) = .25 d. P(X < .5) = 1 – P(X > .5) = 1 – (.5)(.75)(2–.5) = 1 – .5625 = .4375 e. P(X = 1.5) = 0

8.12 a

b. P(2 < X < 4) = P(X < 4) – P(X < 2) = (.5)(3/8)(4–1) – (.5)(1/8)(2–1) = .5625 – .0625 = .5 c. P(X < 3) = (.5)(2/8)(3–1) = .25

8.13a. 218


b. P(1 < X < 3) = P(X < 3) – P(X < 1) =

1 1 1 3   (3  0)    (1  0) = .18 – .02 = .16 2 25 2 25

c. P(4 < X < 8) = P(4 < X < 5) + P(5 < X < 8) P(4 < X < 5)= P(X < 5) – P(X <4) =

1 5 1 4   (5  0)    (4  0) = .5 – .32 = .18 2 25 2 25

P(5 < X < 8) = P(X > 5) – P(X > 8) =

1 2 1 5   (10  5)    (10  8) = .5 – .08 = .42 2 25 2 25

P(4 < X < 8) = .18 + .42 = .60 d. P(X < 7) = 1 – P(X > 7) P(X > 7) =

1 3   (10  7) = .18 2 25

P(X < 7) = 1 – .18 = .82 e. P(X > 3) = 1 – P(X < 3) P(X < 3) =

1 3   (3  0) = .18 2 25

P(X > 3) = 1 – .18 = .82 8.14 a. f(x) = .10 – .005x 0  x  20 b. P(X > 10) = (.5)(.05)(20–10) = .25 c. P(6 < X < 12) = P(X > 6) – PX > 12) = (.5)(.07)(20–6) – (.5)(.04)(20–12) = .49 – .16 = .33 8.15 b P(X < 8) = P(0 < X < 1) + P(1 < X < 8) = .40(1) + .05(8-1) = .40 + .35 = .75 c P(.4 < X < 10) = P(.4 < X < 1) + P(1 < X < 10) = .4(1-.4) + .05(10-1) = .24 + .45 = .69

8.16 b P(X < 5.5) = P(0 < X < 2) + P(2 < X < 5) + P(5 < X < 5.5) = .10(2-0) + .20(5 – 2) + .15(5.5 – 5) = .20 + .60 + .075 = .875 c P(X > 3.5) = P(3.5 < X < 5) +P(5 < X < 6) + P(6 < X < 7) = .20(5-3.5) + .15(6-5) + .05(7-6) = .30 + .15 + .05 = .50 d P(1 < X < 6.5) = P(1 < X < 2) = P(2 < X < 5) + P(5 < X < 6) + P(6 < X < 6.5)

219


= .10(2-1) + .20(5-2) + .15(6-5) + .05(6.5-6) = .10 + .60 + .15 + .025 = .875

8.17 c P(X < 2) = .4(2-0)/2 = .40 d P(X < 3) = P(0 < X < 2) + P(2 < X < 3) = .40 + .40(3-2) = .80 e P(1 < X < 2) = P(0 < X < 2) – P(0 < X < 1) = .4(2-0)/2 - .2(1-0)/2 = .30 P(1 < X < 2.5) = P(1 < X < 2) + P(2 < X < 2.5) = .30 + .4(2.5-2) = .50 8.18 b P(X < 2) = P(0 < X < 4) – P(2 < X < 4) = .4(4-0)/2 - .20(4-2)/2 = .8 - .20 = .60 c P(X > 5) = P(4 < X < 6) – P(4 < X < 5) = .20(6-4)/2 - .10(5-4)/2 = .20 - .05 = .15 d P(2.5 < X < 5.5) = (2.5 < X < 4) + P(4 < X < 5.5) = .15(4-2.5)/2 + .15(5.5-4)/2 = .1125 + .1125 = .2250

8.19 P( Z < 1.60) = .9452

8.20 P(Z < 1.61) = .9463

8.21 P(Z < 1.65) = .9505 8.22 P(Z < −1.39) = .0823 8.23 P(Z < −1.80) = .0359 8.24 P(Z < − 2.16) = .0154 8.25 P(–1.30 < Z < 0.70) = P( Z < 0.70) − P(Z < −1.30) = .7580− .0968 = .6612 8.26 P(Z > –1.24) = 1 – P(Z < −1.24) = 1 − .1075 = .8925 8.27 P(Z < 2.23) = .9871 8.28 P(Z > 1.87) = 1 – P(Z < 1.87) = 1 – .9693 = .0307 8.29 P(Z < 2.57) = ..9949 8.30 P(1.04 < Z < 2.03) = P(Z < 2.03) – P(Z < 1.04) = .9788 – .8508 = .1280 8.31 P(–0.71 < Z < –0.33) = P(Z < −.33) – P(Z < −.71) = .3707 – .2389 = .1318

220


8.32 P(Z > 3.09) = 1 – P(Z < 3.09) = 1 – .9990 = .0010 8.33 P(Z > 0) = 1 – P(Z < 0) = 1 − .5 = .5 8.34 P(Z > 4.0) = 0 8.35 P(Z < z.03) = 1 – .03 = .9700; z.03 = 1.88 8.36 P(Z < z.065) = 1 – .065 = .9350; z.065 = 1.51 8.37 P(Z < z.28 ) = 1 – .28 = .7200; z.28 = .58

 X   145  100   8.38 P(X > 145) = P  = P(Z > 2.25) = 1 – P(Z < 2.25) = 1 – .9878 = .0122 20   

8.39 P(Z < z.15 ) = 1 – .15 = .8500; z .15 = 1.04; z.15 

x  250 x ; 1.04  ; x = 291.6 40 

 800  1,000 X   1,100  1,000    8.40 P(800 < X < 1100) = P  = P(–.8 < Z < .4) 250  250   = P(Z < .4) − P(Z < −.8) = .6554 − .2119 = .4435

8.41 P(Z <  z .08 ) = .0800; z .08  1.41;  z .08 

x  x  50 ;  1.41  ; x = 38.72  8

 5  6.3 X   10  6.3    8.42 a P(5 < X < 10) = P  = P(–.59 < Z > 1.68) 2.2    2.2 = P(Z < 1.68) − P(Z < −.59) = .9535 − .2776 = .6759

 X   7  6.3   b P(X > 7) = P  = P(Z > .32) = 1 – P(Z < .32) = 1 – .6255 = .3745 2.2     X   4  6.3   c P(X < 4) = P  = P(Z < –1.05) = .1469 2.2   

8.43 P(Z < z .10 ) = 1 – .10 = .9000; z .10 = 1.28; z .10  Calls last at least 9.116 minutes. 221

x  x  6 .3 ; 1.28  ; x = 9.116  2 .2


 X   5,000  5,100   8.44 P(X > 5,000) = P  = P(Z > –.5) = 1 −P(Z < −.5) = 1 − .3085 = .6915 200   

8.45 P(Z <  z .02 ) = .02; z .02  2.05;  z .02 

x  x  5100 ;  2.05  ; x = 4690;  200

8.46 z.75 = -.67: Q1 = (-.67)(300) + 1000 = 799, z.50 = 0: Q2 = 1000, z.25 = .67: Q3 = (.67)(300) + 1000 = 1201

 X   30 ,000  25,000    = P(Z > 1.00) = 1 – P(Z < 1.00) = 1 – .8413 8.47 a P(X > 30,000) = P 5,000    = .1587

 X   22 ,500  25,000    = P(Z < –0.50) = .3085 b P(X < 22,500) = P 5,000   

 20 ,000  25,000 X   32 ,000  25,000     c P(20,000 < X < 32,000) = P 5,000  5,000   = P(–1.00 < Z < 1.40) = P(Z < 1.40) − P(Z < −1.00) = .9192 − 1587 = .7605

 X   210  200  8.48 P(X > 210) = P   = P(Z > .5) = 1 −P(Z < .5) = 1 − .6915 = .3085 20    8.49 85th percentile = (z.15)(5) + 68 = 1.04(5) + 68 = 73.2 8.50 First quintile = (z.80)σ + µ = (-.84)(10,000) + 50,000 = 41,600 Second quintile = (z.60)σ + µ = (-.25)(10,000) + 50,000 = 47,500 Third quintile = (z.40)σ + µ = (.25)(10,000) + 50,000 = 52,500 Fourth quintile = (z.20)σ + µ = (.84)(10,000) + 50,000 = 58,400

 X   70,000  82,000    = P(Z < –1.88) = .0301 8.51 a P(X < 70,000) = P 6,400     X   100 ,000  82 ,000   b P(X > 100,000) = P  = P(Z > 2.81) = 1 – P(Z < 2.81) = 1 – .9975 = 6,400    .0025

222


8.52 Top 5%: P(Z < z .05 ) = 1 – .05 = .9500; z .05 = 1.645; z .05 

x  x  32 ; 1.645  ;x=  1 .5

34.4675 Bottom 5%: P(Z <  z .05 ) = .0500; z .05  1.645 ;  z .05 

x  x  32 ;  1.645  ;  1.5

x = 29.5325

 X   36  32   8.53 a P(X > 36) = P  = P(Z > 2.67) = 1 – P(Z < 2.67) = 1 – .9962 = .0038 1.5   

 X   34  32   b P(X < 34) = P  = P(Z < 1.33) = .9082 1.5     30  32 X   33  32    c P(30 < X < 33) = P  = P(–1.33 < Z < .67)  1.5   1.5 = P(Z < .67) − P(Z < −1.33) = .7486 − .0918 = .6568

 X   8  7.2   8.54 P(X > 8) = P  = P (Z > 1.20) = 1 – P(Z < 1.20) = 1 – .8849 = .1151 .667   

8.55 P(Z < z .25 ) = .7500; z .25 = .67; z .25 

x  7.2 x  ; .67  ; x = 7.65 hours .667 

 X   10  7.5   8.56 a P(X > 10) = P  = P Z > 1.19) = 1 – P(Z < 1.19) = 1 – .8830 = .1170 2.1     7  7. 5 X   9  7 . 5    b P(7 < X < 9) = P  = P(–.24 < Z < .71) 2.1    2.1 = P(Z < .71) − P(0 < Z < −.24) = .7611 − .4052 = .3559

 X   3  7.5   c P(X < 3) = P  = P Z < –2.14) = .0162 2.1    d P(Z < – z .05 ) = .0500;  z .05 = –1.645;  z .05 

x  x  7. 5 ;  1.645  ; x = 4.05 hours 2 .1 

 X   12 ,000  11,500   8.57 a P(X > 12,000) = P  = P Z > .63) = 1– P(Z < .63) = 1 – .7357 = 800    .2643

 X   10 ,000  11,500   b P(X < 10,000) = P  = P(Z < –1.88) = .0301 800   

223


8.58 P(Z < – z .01 ) = .0100;  z .05 = –2.33;  z .01 

x  x  11,500 ;  2.33  ; x = 9,636  800

 X   4900  6000  8.59 a P(X > 4900) = P   = P Z > -.92) = 1 – P(Z < -.92) = 1 – .1788 = 1200    .8212

 3800  6000 X   5700  6000  b P(3800 < X < 5700) = P    = P(–1.83 < Z < -.25)  1200 1200   = P(Z < -.25) − P(Z < −1.83) = .4013 − .0336 = .3677

 X   6500  6000  c P(X < 6500) = P   = P Z < .42) = .6628 1200     X   145  155  8.60 P(X < 145) = P   = P Z < -1.11) = .1335 9     24  26 X   28  26    8.61 a P(24 < X < 28) = P  = P(–.80 < Z < .80) 2. 5    2. 5 = P(Z < .80) – P(Z < −.80) = .7881 − .2119 = .5762

 X   28  26   b P(X > 28) = P  = P(Z > .80) = 1 – P(Z < .80) = 1 – .7881 = .2119 2.5     X   24  26   c P(X < 24) = P  = P(Z < –.80) =.2119 2.5   

 X   30  27   8.62 a P(X > 30) = P  = P(Z > .43) = 1 – P(Z < .43) = 1 – .6664 = .3336 7   

 X   40  27   b P(X > 40) = P  = P(Z > 1.86) = 1 – P(Z < 1.86) = 1 – .9686 = .0314 7     X   15  27   c P(X < 15) = P  = P(Z < –1.71) = .0436 7    d P(Z < z .20 ) = 1 – .20 = .8000; z .20 = .84; z .20 

x  x  27 ; .84  ; x = 32.88  7

 X   4  7.5   8.63 a P(X < 4) = P  = P(Z < –2.92) = .0018 1.2   

224


 7  7.5 X   10  7.5    b P(7 < X < 10) = P  = P(–.42 < Z < 2.08)  1.2   1.2 = P(Z < 2.08) − P(Z < −.42) = .9812 − .3372 = .6440

 X   10  16 .40   8.64a P(X < 10) = P  = P(Z < –2.33) = .0099 2.75    b P(Z < – z .10 ) = .1000; – z .10 = –1.28;  z .10 

x  x  16 .40 ;  1.28  ; x = 12.88  2.75

8.65 A: P(Z < z .10 ) = 1– .10 = .9000; z .10 = 1.28; z .10  B: P(Z < z .40 ) = 1 –.40 = .6000; z .40 = .25; z .40  C: P(Z < – z .20 ) = .2000; z .20  .84;  z .20 

x  70 x  ; 1.28  ; x = 82.8  10

x  70 x  ; .25  ; x = 72.5  10

x  x  70 ;  .84  ; x = 61.6;  10

D: P(Z < – z .05 ) = .0500; z .05  1.645 ;  z .05 

x  x  70 ;  1.645  ; x = 53.55  10

8.66 P(Z < z .02 ) = 1 – .02 = .9800; z .02 = 2.05; z .02 

x  100 x  ; 2.05  ; x = 132.80  16

(rounded to 133)

 X   64,000  50,000  8.67 P(x > 64,000) = P   = P(Z > 1.75) = 1 – P(Z < 1.75) = 1 – .9599 8000    = .0401

 X   500 ,000  325 ,600   8.68 P(x > 500,000) = P  = P(Z > 1.74) = 1 – P(Z < 1.74) = 1 – .9591 100 ,000    = .0409

8.69 P(Z < z .06 ) = 1 – .06 = .9400; z .06 = 1.55; z .06 

ROP   ROP  200 ; 1.55  ;  30

ROP = 246.5 (rounded to 247)

8.70 P(Z < z.20 ) = 1 – .20 = .8000; z .20 = .84; z .20 

225

x  150 x  ; .84  ; x = 171  25


8.71 P(Z < z .30 ) = 1 – .30 = .7000; z .30 = .52; z .30 

x  850 x  ; .52  ; x = 896.8  90

(rounded to 897)

8.72 P( Z < z .40 ) = 1– .40 = .6000; z .40 = .25; z .40 

x  850 x  ; .25  ; .x = 872.5  90

(rounded to 873)

 X   25  14  8.73 a P(X > 25 = P   = P(Z > .61) = 1 – P(Z < .61) = 1 - .7291 = .2709 18   

 X   0  14  b P(X < 0) = P   = P(Z < –.78) = .2177 18    8.74 First quintile = (z.80)σ + µ = (-.84)(30,000) + 250,000 = 224,800 Second quintile = (z.60)σ + µ = (-.25)(30,000) + 250,000 = 242,500 Third quintile = (z.40)σ + µ = (.25)(30,000) + 250,000 = 257,500 Fourth quintile = (z.20)σ + µ = (.84)(30,000) + 250,000 = 275,200 8.75 From Exercise 7.57:  = 65,  2 = 21, and  = 4.58

 X   60  65   P(X > 60) = P  = P(Z > –1.09) = 1 − P(Z < −1.09) = 1 −.1379 = .8621 4.58   

 X   150  145   8.76 P(X < 150) = P  = P(Z < .90) = .8159 5.57   

 X   105  100  8.77 a P(X > 105) = P   = P(Z > .90) = 1- P(Z < .90) = 1 - .8159 = .1841 5.48     X   92  100  b P(X > 92) = P   = P(Z > -1.46) = 1- P(Z < -1.46) = 1 - .0721 = .9279 5.48   

 95  100 X   112  100  c P(95 < X < 112) = P    = P(–.91 < Z < 2.19) 5.48    5.48 = P(Z < 2.19) − P(Z < −.91) = .9857 − .1814 = .8043 8.78 z.75 = -.67: Q1 = (-.67)(23.96) + 168 = 152.95, z.50 = 0: Q2 = 168, z.25 = .67: Q3 = (.67)(23.96) + 168 = 185.05

226


8.79

Exponential Distribution (Lambda = 3)

f(x)

3,5 3 2,5 2 1,5 1 0,5 0 0

0,5

1 X

1,5

2

8.80

Exponential Distribution (Lambda = .25) 0,3 0,25

f(x)

0,2 0,15 0,1 0,05 0 0

5

10 X

8.81 a P(X  1)  e ..5(1)  e ..5 = .6065 a P(X  .4)  e ..5(.4)  e ..2 = .8187 c P(X  .5)  1  e ..5(..5)  1  e ..25 = 1 – .7788 = .2212 d P(X  2)  1  e .5(2)  1  e 1 = 1 – .3679 = .6321

8.82 a P(X  2)  e ..3(2)  e .6 = .5488 b P(X  4)  1  e ..3(4)  1  e 1.2 = 1 – .3012 = .6988

227

15

20


c P(1  X  2)  e ..3(1)  e ..3(2)  e ..3  e .6 = .7408 – .5488 = .1920 d P(X = 3) = 0 8.83  = 6 kilograms/hour = .1 kilogram/minute P(X  15)  e .1(15)  e 1.5 = .2231

8.84   1 /   25 hours;  = .04 breakdowns/hour P(X  50)  e .04(50)  e 2 = .1353

8.85  = 10 trucks/hour = .167 truck/minute P(X  15)  e .167(15)  e 2.5 = .0821

8.86   1 /  = 5 minutes;  = .2 customer/minute P(X  10)  1  e .2(10)  1  e 2 = 1– .1353 = .8647

8.87   1 /  = 2.7 minutes;  = .37 service/minute P(X  3)  1  e .37(3)  1  e 1.11 = 1– .3296 = .6704

8.88   1 /  = 7.5 minutes;  = .133 service/minute P(X  5)  1  e .133(5)  1  e .665 = 1– .5143 = .4857

8.89   1 /  = 125 seconds;  = .008 transactions/second = .48 transactions/minute P(X  3)  e .48(3)  e 1.44 = .2369

8.90   1 /  = 6 minutes;  = .167 customers/minute P(X  10)  e .167(10)  e 1.67 = .1889

8.91 x = 30 and P(X > x) = .01. λ=

 ln[ P( X  x)]  ln[. 01] (4.605 ) = = = .1535 x 30 30

The service rate should be .1535 trucks per minute or 9.21 trucks per hour.

8.92 x = 30 and P(X > x) = .10.

228


λ=

 ln[ P( X  x)]  ln[. 10 ] (2.303 ) = = = .0768 x 30 30

The service rate should be .0768 cars per minute or 4.61 cars per hour.

8.93

a 1.341

b 1.319

c 1.988

d 1.653

8.94

a 2.750

b 1.282

c 2.132

d 2.528

8.95

a 1.3406

b 1.3195

c 1.9890

d 1.6527

8.96

a 1.6556

b 2.6810

c 1.9600

d 1.6602

8.97

a .0189

b .0341

c .0927

d .0324

8.98

a .1744

b .0231

c .0251

d .0267

8.99

a 9.24

b 136

c 9.39

d 37.5

8.100

a 17.3

b 50.9

c 2.71

d 53.5

8.101

a 73.3441

b 102.946

c 16.3382

d 24.7690

8.102

a 33.5705

b 866.911

c 24.3976

d 261.058

8.103

a .2688

b 1.0

c .9903

d 1.0

8.104

a .4881

b .9158

c .9988

d .9077

8.105

a 4.35

b 8.89

c 3.29

d 2.50

8.106

a 2.84

b 1.93

c 3.60

d 3.37

8.107

a 1.4857

b 1.7633

c 1.8200

d 1.1587

8.108

a 1.5204

b 1.5943

c 2.8397

d 1.1670

8.109

a .0510

b .1634

c .0222

d .2133

229


8.110

a .1050

b .1576

c .0001

230

d .0044


Chapter 9 9.91a. 1/6 b. 1/6 9.10 a P( X  1) =P(1,1)= 1/36 b P( X  6) = P(6,6) = 1/36

9.11a P( X = 1) = (1/6) 5 = .0001286 b P( X = 6) = (1/6) 5 = .0001286

9.12 The variance of X is smaller than the variance of X.

9.13 The sampling distribution of the mean is normal with a mean of 40 and a standard deviation of 12/ 100 = 1.2.

9.14 No, because the sample mean is approximately normally distributed.

 X   1050  1000   = P(Z > 1.00) = 1 – P(Z < 1.00) = 1 – .8413 9.15 a P( X  1050 ) = P   200 / 16  / n = .1587

 X   960  1000   = P(Z < –.80) = .2119 b P( X  960 )  P   200 / 16  / n  X   1100  1000   = P(Z > 2.00) = 1 – P(Z < 2.00) = 1 – .9772 = .0228 c P( X  1100 )  P   200 / 16  / n

 X   1050  1000   = P(Z > 1.25) = 1 – P(Z < 1.25) = 1 – .8944 9.16 a P( X  1050 ) = P   200 / 25  / n = .1056

 X   960  1000   = P(Z < –1.00) = .1587 b P( X  960 )  P   200 / 25  / n  X   1100  1000   = P(Z > 2.50) = 1 – P(Z < 2.50) = 1 – .9938 = .0062 c P( X  1100 )  P   200 / 25  / n

229


 X   1050  1000   = P(Z > 2.50) = 1 – P(Z < 2.50) = 1 – .9938 9.17 a P( X  1050 ) = P   200 / 100  / n = .0062

 X   960  1000   = P(Z < –2.00) = .0228 b P( X  960 )  P     / n 200 / 100 

 X   1100  1000   = P(Z > 5.00) = 0 c P( X  1100 )  P   200 / 100  / n

 49  50 X   52  50   = P(–.40 < Z < .80) 9.18 a P(49  X  52 )  P    / n 5/ 4   5/ 4 = P(Z < .80) − P(Z < −.40) = .7881 −.3446 = .4435

 49  50 X   52  50   = P(–.80 < Z < 1.60) b P(49  X  52 )  P     5 / 16  / n 5 / 16  = P(Z < 1.60) − P(Z < −.80) = .9452 −.2119 = .7333

 49  50 X   52  50   = P(–1.00 < Z < 2.00) c P(49  X  52 )  P     5 / 25  / n 5 / 25  = P(Z < 2.00) − P(Z < −1.00) = .9772 −.1587 = .8185

 49  50 X   52  50   = P(–.20 < Z < .40) 9.19 a P(49  X  52 )  P     10 / 4  / n 10 / 4  = P(Z < .40) − P(Z < −.20) = .6554 −.4207= .2347

 49  50 X  52  50   = P(–.40 < Z < .80) b P(49  X  52 )  P     10 / 16  / n 10 / 16  = P(Z < .80) − P(Z < −.40) = .7881 −.3446= .4435

 49  50 X  52  50   = P(–.50 < Z < 1.00) c P(49  X  52 )  P     10 / 25  / n 10 / 25  = P(Z < 1.00) − P(Z < −.50) = .8413 −.3085= .5328

 49  50 X   52  50   = P(–.10 < Z < .20) 9.20 a P(49  X  52 )  P     20 / 4  / n 20 / 4  = P(Z < .20) − P(Z < −.10) = .5793 −.4602= .1191

 49  50 X  52  50   = P(–.20 < Z < .40) b P(49  X  52 )  P     20 / 16  / n 20 / 16  = P(Z < .40) − P(Z < −.20) = .6554 −.4207= .2347 230


 49  50 X  52  50   = P(–.25 < Z < .50) c P(49  X  52 )  P     20 / 25  / n 20 / 25  = P(Z < .50) − P(Z < −.25) = .6915 −.4013= .2902

9.21 a

Nn = N 1

1,000  100 = .9492 1,000  1

b

Nn = N 1

3,000  100 = .9834 3,000  1

c

Nn = N 1

5,000  100 = .9900 5,000  1

d. The finite population correction factor is approximately 1.

9.22 a  x =

b x = c x =

 n  n

 n

Nn = N 1

500 1,000

10 ,000  1,000 = 15.00 10 ,000  1

500 Nn = N 1 500

10,000  500 = 21.80 10,000  1

500 Nn = N 1 100

10,000  100 = 49.75 10,000  1

 X   66  64   9.23 a P(X > 66) = P  = P(Z > 1.00) = 1 – P(Z < 1.00) = 1 – .8413 = .1587 2   

 X   66  64   = P(Z > 2.00) = 1 – P(Z < 2.00) = 1 – .9772 = .0228 b P( X  66 )  P   2/ 4  / n

 X  66  64   = P(Z > 10.00) = 0 c P( X  66 )  P     / n 2 / 100 

9.24 We can answer part (c) and possibly part (b) depending on how nonnormal the population is.  X   120  117   9.25 a P(X > 120) = P  = P(Z > 0.58) = 1 – P(Z < .58) = 1 – .7190 = .2810 5 .2   

 X   120  117   = P(Z > 1.15) = 1 – P(Z < 1.15) = 1 – .8749 = .1251 b P( X  120 )  P   5.2 / 4  / n 4

c [P(X >120)] =[.2810]

4

= .00623

231


 X   262 ,000  250 ,000   = P(Z > 2.4) = 1 – P(Z < 2.4) 9.26 P( X  262 ,000 )  P  50,000 / 100   / n = 1 - .9918 = .0082

9.27 No, unless the population is extremely nonnormal.

 X   60  52   9.28 a P(X > 60) = P  = P(Z > 1.33) = 1 – P(Z < 1.33) = 1 – .9082 = .0918 6   

 X   60  52   = P(Z > 2.31) = 1 – P(Z < 2.31) = 1 – .9896 = .0104 b P( X  60 )  P   6/ 3  / n 3

3

c [P(X >60)] =[.0918] = .00077  X   12  10   9.29 a P(X > 12) = P  = P(Z > .67) = 1 – P(Z < .67) = 1 – .7486 = .2514 3   

 X   11  10   = P(Z > 1.67) = 1 – P(Z < 1.67) b P( X  275 / 25 )  P( X  11)  P     / n 3 / 25  = 1 – .9525 = 0475  X   75  78   9.30 a P(X < 75) = P  = P(Z < –.50) = .3085 6   

 X   75  78   = P(Z < –3.54) = 1 – P(Z < 3.54) = 1 – 1 = 0 b P( X  75)  P     / n 6 / 50 

 X  76  9.31 a P(X > 7) = P  = P(Z > .67) = 1 – P(Z < .67) = 1 – .7486 = .2514 1 .5   

 X  76   = P(Z > 1.49) = 1 – P(Z < 1.49) = 1 – .9319 = .0681 b P( X  7)  P     / n 1.5 / 5  5

5

c [P(X >7)] =[.2514] = .00100

 X   5.97  6.05   = P(Z < –2.67) =.0038 9.32 a P( X  5.97 )  P   .18 / 36  / n b It appears to be false.

232


 X   625  600   = P(Z > .50) = 1 – P(Z < .50) 9.33 P( X  10,000 / 16 )  P( X  625 )  P     / n 200 / 16  = 1 – .6915 = .3085 9.34 The professor needs to know the mean and standard deviation of the population of the weights of elevator users and that the distribution is not extremely nonnormal.

 X   71 .25  75   = P(Z > –1.50) 9.35 P( X  1,140 / 16 )  P( X  71 .25 )  P   10 / 16  / n = 1 − P(Z < −1.50) = 1 − 0668 = .9332

 X  5  4.8   = P(Z > 1.19) 9.36 P(Total time > 300) = P( X  300 / 60 )  P(X  5)  P     / n 1.3 / 60  = 1 – P(Z < 1.19) = 1 – .8830 = .1170 9.37 No because the central limit theorem says that the sample mean is approximately normally distributed.

 X   1.92  2.0   9.38 P(Total number of cups > 240) = P( X  240 / 125 )  P( X  1.92 )  P     / n .6 / 125  = P(Z > –1.49) = 1 − P(Z < −1.49) = 1 − .0681 = .9319

 X   300  275   9.39 P(Total number of faxes > 1500) = P( X  1500 / 5)  P( X  300 )  P   75 / 5  / n = P(Z > .75) = 1 – P(Z < .75) = 1 – .7734 = .2266

9.40 a The sample mean is normally distributed with a mean of 2800 and a standard deviation of 200.

 2500  2800 X   2900  2800   = P(-1.5 < Z < .5) b P(2500  X  2900 )  P   / n 400 / 4   400 / 4 = P(Z < .5) – P(Z < -1.5) = .6915 – .0668 = .6247 9.41 We could not answer the question because the sample size is small.

233


 P̂  p  9.42a P( P̂ > .60) = P  p(1  p) / n 

  = P(Z > 3.46) = 0 (.5)(1  .5) / 300 

 P̂  p  b. P( P̂ > .60) = P  p(1  p) / n 

.60  .55

.60  .5

  = P(Z > 1.74) = 1 – P(Z < 1.74) (.55 )(1  .55 ) / 300 

= 1 – .9591 = .0409

  = P(Z > 0) = 1 – P(Z < 0) = 1 − .5 = .5 (.6)(1  .6) / 300 

 P̂  p  c. P( P̂ > .60) = P  p(1  p) / n 

.60  .6

 P̂  p  9.43a P( P̂ < .22) = P  p(1  p) / n 

  = P(Z < –1.55) = .0606 (.25 )(1  .25 ) / 500  .22  .25

 P̂  p  b. P( P̂ < .22) = P  p(1  p) / n 

  = P(Z < –1.96) = .0250 (.25 )(1  .25 ) / 800 

 P̂  p  c. P( P̂ < .22) = P  p(1  p) / n 

  = P(Z < –2.19) = .0143 (.25 )(1  .25) / 1000 

.22  .25

.22  .25

 P̂  p  9.44 P( P̂ < .75) = P  p(1  p) / n 

  = P(Z < –1.25) = .1056 (.80 )(1  .80 ) / 100 

 P̂  p  9.45 P( P̂ > .35)= P  p(1  p) / n 

  = P(Z > –.79) = .1 − P(Z < −.79) (.40 )(1  .40 ) / 60 

.75  .80

.35  .40

1 − .2148= .7852

 P̂  p  9.46 P( P̂ < .49) = P  p(1  p) / n 

  = P(Z < –2.70) = .0035 (.55 )(1  .55 ) / 500 

 P̂  p  9.47 P( P̂ > .04)= P  p(1  p) / n 

  = P(Z > 4.04) = 1 – P(Z < 4.04) (.02 )(1  .02 ) / 800 

.49  .55

.04  .02

= 1 – 1= 0; The defective rate appears to be larger than 2%.

 P̂  p  9.48 a P( P̂ < .50) = P  p(1  p) / n 

  = P(Z < –1.20) =.1151; the claim may (.53)(1  .53) / 400  .50  .53

be true

234


 P̂  p  b P( P̂ < .50) = P  p(1  p) / n 

  = P(Z < –1.90) = .0287; the claim (.53)(1  .53) / 1,000  .50  .53

appears to be false

 P̂  p  9.49 P( P̂ > .10) = P  p(1  p) / n 

  = P(Z > –1.15) = 1 – P(Z < – 1.15) (.14 )(1  .14 ) / 100  .10  .14

= 1 – .1251 = .8749

 P̂  p  9.50 P( P̂ > .05)= P  p(1  p) / n 

  = P(Z > 2.34) = 1 – P(Z < 2.34) (.03)(1  .03) / 400  .05  .03

= 1 – .9904 = .0096; The commercial appears to be dishonest

 P̂  p  9.51 P( P̂ > .32) = P  p(1  p) / n 

  = P(Z > 1.38) = 1 – P(Z < 1.38) (.30 )(1  .30 ) / 1,000  .32  .30

= 1 – .9162 = .0838

 P̂  p  9.52 a P( P̂ < .45) = P  p(1  p) / n 

  = P(Z < –2.45) = .0071 (.50 )(1  .50 ) / 600  .45  .50

b The claim appears to be false.

 P̂  p  9.53 P( P̂ < .75) = P  p(1  p) / n 

  = P(Z < –2.34) = .0096 (.80 )(1  .80 ) / 350 

 P̂  p  9.54 P( P̂ < .70) = P  p(1  p) / n 

  = P(Z < –2.48) = .0066 (.75)(1  .75 ) / 460 

 P̂  p  9.55 P( P̂ > .28) = P  p(1  p) / n 

  = P(Z > 2.40) = 1 – P(Z < 2.40) (.25 )(1  .25 ) / 1200 

.75  .80

.70  .75

.28  .25

= 1 – .9918 = .0082 9.56 The claim appears to be false.

235


 9.57 P( P̂ > .15) = P  

Pˆ  p p(1  p) / n

  = P(Z > 1.19) = 1 – P(Z < 1.19) (.13)(1  .13) / 400  .15  .13

= 1 – .8830 = .1170

 9.58 P( P̂ > .22) = P  

Pˆ  p p (1  p) / n

  = P(Z > 1.12) = 1 – P(Z < 1.12) (.20 )(1  .20 ) / 500  .22  .20

= 1 – .8686 = .1314

 9.59 P( P̂ < .08) = P  

 9.60 P( P̂ < .08) = P  

Pˆ  p p(1  p) / n

Pˆ  p p(1  p) / n

  = P(Z < -.82) = .2061 (.10 )(1  .10 ) / 150 

  = P(Z < -2.40) = .0082 (.15)(1  .15) / 150 

.08  .10

.08  .15

     ( X  X )  (   ) 25  (280  270 )  1 2 1 2  = P(Z > 1.21) 9.61 P( X 1  X 2  25 )  P   25 2 30 2  12  22      10 10  n1 n 2 

= 1 – P(Z < 1.21) = 1 – .8869 = .1131      ( X  X )  (   ) 25  (280  270 )  2 1 2  = P(Z > 2.72) 9.62 P( X 1  X 2  25 )  P 1  2 2  1  2 25 2 30 2       50 50  n1 n 2 

= 1 – P(Z < 2.72) = 1 – .9967 = .0033      ( X  X )  (   ) 25  (280  270 )  2 1 2  = P(Z > 3.84) 9.63 P( X 1  X 2  25 )  P 1  2 2  25 2 30 2  1  2      100 100  n1 n 2 

= 1 – P(Z < 3.84) = 1 – 1= 0

236


     ( X  X )  (   ) 0  (40  38 )  2 1 2  = P(Z > –1.00) 9.64 P( X 1  X 2  0)  P 1  2 2 2  2  6 8 1  2      25 25  n n 2 1 

= 1 – P(Z < –1.00) = 1 – .1587 = .8413      ( X  X )  (   ) 0  (40  38 )  2 1 2  = P(Z > –.50) = 1 – P(Z < –.50) 9.65 P( X 1  X 2  0)  P 1  2 2  12 2 16 2  1  2      25 25  n1 n 2 

= 1 – .3085 = .6915      ( X  X )  (   ) 0  (140  138 )  2 1 2  = P(Z > –1.00) 9.66 P( X 1  X 2  0)  P 1   12  22 62 82       25 25  n1 n 2 

= 1 – P(Z < –1.00) = 1 – .1587 = .8413      ( X  X )  (   ) 0  (75  65 )  2 1 2  = P(Z > –.77) = 1 – P(Z < −.77) 9.67 P( X 1  X 2  0)  P 1  2 2  1  2 20 2 21 2       5 5  n1 n 2 

= 1 – .2206 = .7794      ( X  X )  (   ) 0  (73  77 )  1 2 1 2  = P(Z > .51) = 1 – P(Z < .51) 9.68 P( X 1  X 2  0)  P   12  22 12 2 10 2       4 4  n1 n 2 

= 1 – .6950 = .3050      ( X  X )  (   ) 0  (18  15 )  2 1 2  = P(Z > –2.24) = 1– P(Z < –2.24) 9.69 P( X 1  X 2  0)  P 1  2 2  32 32  1  2      10 10  n1 n 2 

= 1 – .0125 = .9875

237


     ( X  X )  (   ) 0  (10  15 )  2 1 2  = P(Z < 5.89) = 1 9.70 P( X 1  X 2  0)  P 1  2 2 2  2  3 3 1  2      25 25  n n 2 1 

238


Chapter 10 10.1 A point estimator is a single value; an interval estimator is a range of values.

10.2 An unbiased estimator of a parameter is an estimator whose expected value equals the parameter.

10.3

10.4

10.5 An unbiased estimator is consistent if the difference between the estimator and the parameter grows smaller as the sample size grows.

10.6

239


10.7 If there are two unbiased estimators of a parameter, the one whose variance is smaller is relatively efficient.

10.8

10.13 a. x  z  / 2  / n = 100  1.645(25/ 50 ) = 100  5.82; LCL = 94.18, UCL = 105.82 b. x  z  / 2  / n = 100  1.96(25/ 50 ) = 100  6.93; LCL = 93.07, UCL = 106.93 c. x  z  / 2  / n = 100  2.575(25/ 50 ) = 100  9.11; LCL = 90.89, UCL = 109.11 d. The interval widens.

10.14 a. x  z  / 2  / n = 200  1.96(50/ 25 ) = 200  19.60; LCL = 180.40, UCL = 219.60 b. x  z  / 2  / n = 200  1.96(25/ 25 ) = 200  9.80; LCL = 190.20, UCL = 209.80 c. x  z  / 2  / n = 200  1.96(10/ 25 ) = 200  3.92; LCL = 196.08, UCL = 203.92 d. The interval narrows.

10.15 a. x  z  / 2  / n = 80  1.96(5/ 25 ) = 80  1.96; LCL = 78.04, UCL = 81.96 b. x  z  / 2  / n = 80  1.96(5/ 100 ) = 80  .98; LCL = 79.02, UCL = 80.98 c. x  z  / 2  / n = 80  1.96(5/ 400 ) = 80  .49; LCL = 79.51, UCL = 80.49 d. The interval narrows.

10.16 a. x  z  / 2  / n = 500  2.33(12/ 50 ) = 500  3.95; LCL = 496.05, UCL = 503.95 b. x  z  / 2  / n = 500  1.96(12/ 50 ) = 500  3.33; LCL = 496.67, UCL = 503.33 c. x  z  / 2  / n = 500  1.645(12/ 50 ) = 500  2.79; LCL = 497.21, UCL = 502.79 d. The interval narrows.

240


10.17 a. x  z  / 2  / n = 500  2.575(15/ 25 ) = 500  7.73; LCL = 492.27, UCL = 507.73 b. x  z  / 2  / n = 500  2.575(30/ 25 ) = 500  15.45; LCL = 484.55, UCL = 515.45 c. x  z  / 2  / n = 500  2.575(60/ 25 ) = 500  30.91; LCL = 469.09, UCL = 530.91 d. The interval widens.

10.18 a. x  z  / 2  / n = 10  1.645(5/ 100 ) = 10  .82; LCL = 9.18, UCL = 10.82 b. x  z  / 2  / n = 10  1.645(5/ 25 ) = 10  1.64; LCL = 8.36, UCL = 11.64 c. x  z  / 2  / n = 10  1.645(5/ 10 ) = 10  2.60; LCL = 7.40, UCL = 12.60 d. The interval widens.

10.19 a. x  z  / 2  / n = 100  1.96(20/ 25 ) = 100  7.84; LCL = 92.16, UCL = 107.84 b. x  z  / 2  / n = 200  1.96(20/ 25 ) = 200  7.84; LCL = 192.16, UCL = 207.84 c. x  z  / 2  / n = 500  1.96(20/ 25 ) = 500  7.84; LCL = 492.16, UCL = 507.84 d. The width of the interval is unchanged.

10.20 a. x  z  / 2  / n = 400  2.575(5/ 100 ) = 400  1.29; LCL = 398.71, UCL = 401.29 b. x  z  / 2  / n = 200  2.575(5/ 100 ) = 200  1.29; LCL = 198.71, UCL = 201.29 c. x  z  / 2  / n = 100  2.575(5/ 100 ) = 100  1.29; LCL = 98.71, UCL = 101.29 d. The width of the interval is unchanged.

10.21 Yes, because the expected value of the sample median is equal to the population mean.

10.22 The variance decreases as the sample size increases, which means that the difference between the estimator and the parameter grows smaller as the sample size grows larger.

10.23 Because the variance of the sample mean is less than the variance of the sample median, the sample mean is relatively more efficient than the sample median.

10.24 a sample median  z  / 2

1.2533 

= 500  1.645

1.2533 (12 ) 50

n

241

= 500  3.50


b. The 90% confidence interval estimate of the population mean using the sample mean is 500  2.79. The 90% confidence interval of the population mean using the sample median is wider than that using the sample mean because the variance of the sample median is larger. The median is calculated by placing all the observations in order. Thus, the median loses the potential information contained in the actual values in the sample. This results in a wider interval estimate.

10.25 x  z  / 2  / n = 6.89  1.645(2/ 9 ) = 6.89  1.10; LCL = 5.79, UCL = 7.99

10.26 x  z  / 2  / n = 43.75  1.96(10/ 8 ) = 43.75  6.93; LCL = 36.82, UCL = 50.68 We estimate that the mean age of men who frequent bars lies between 36.82 and 50.68. This type of estimate is correct 95% of the time.

10.27 x  z  / 2  / n = 22.83  1.96(12/ 12 ) = 22.83  6.79; LCL = 16.04, UCL = 29.62

10.28 x  z  / 2  / n = 9.85  1.645(8/ 20 ) = 9.85  2.94; LCL = 6.91, UCL = 12.79

10.29 x  z  / 2  / n = 68.6  1.96(15/ 15 ) = 68.6  7.59; LCL = 61.01, UCL = 76.19 We estimate that the mean number of cars sold annually by all used car salespersons lies between 61.01 and 76.19. This type of estimate is correct 95% of the time.

10.30 x  z  / 2  / n = 16.9  2.575(5/ 10 ) = 16.9  4.07; LCL = 12.83, UCL = 20.97

10.31 x  z  / 2  / n = 147.33  1.96(40/ 15 ) = 147.33  20.24; LCL = 127.09, UCL = 167.57

10.32 x  z  / 2  / n = 13.15  1.645(6/ 13 ) = 13.15  2.74; LCL = 10.41, UCL = 15.89

10.33 x  z  / 2  / n = 75.625  2.575(15/ 16 ) = 75.625  9.656; LCL = 65.969, UCL = 85.281

10.34 x  z  / 2  / n = 252.38  1.96(30/ 400 ) = 252.38  2.94; LCL = 249.44, UCL = 255.32

242


10.35 x  z  / 2  / n = 1,810.16  1.96(400/ 64 ) = 1,810.16  98.00; LCL = 1,712.16, UCL = 1,908.16

10.36 x  z  / 2  / n = 12.10  1.645(2.1/ 200 ) = 12.10  .24; LCL = 11.86, UCL = 12.34. We estimate that the mean rate of return on all real estate investments lies between 11.86% and 12.34%. This type of estimate is correct 90% of the time.

10.37 x  z  / 2  / n = 10.21  2.575(2.2/ 100 ) = 10.21  .57; LCL = 9.64, UCL = 10.78

10.38 x  z  / 2  / n = .510  2.575(.1/ 250 ) = .510  .016; LCL = .494, UCL = .526. We estimate that the mean growth rate of this type of grass lies between .494 and .526 inch . This type of estimate is correct 99% of the time.

10.39 x  z  / 2  / n = 26.81  1.96(1.3/ 50 ) = 26.81  .36; LCL = 26.45, UCL = 27.17. We estimate that the mean time to assemble a cell phone lies between 26.45 and 27.17 minutes. This type of estimate is correct 95% of the time.

10.40 x  z  / 2  / n = 19.28  1.645(6/ 250 ) = 19.28  .62; LCL = 18.66, UCL = 19.90. We estimate that the mean leisure time per week of Japanese middle managers lies between 18.66 and 19.90 hours. This type of estimate is correct 90% of the time.

10.41 x  z  / 2  / n = 15.00  2.575(2.3/ 100 ) = 15.00  .59; LCL = 14.41, UCL = 15.59. We estimate that the mean pulse-recovery time lies between 14.41 and 15.59 minutes. This type of estimate is correct 99% of the time.

10.42 x  z  / 2  / n = 585,063  1.645(30,000/ 80 ) = 585,063  5,518; LCL = 579,545, UCL = 590,581. We estimate that the mean annual income of all company presidents lies between $579,545 and $590,581. This type of estimate is correct 90% of the time.

10.43 x  z  / 2  / n = 109.6  1.96(20/ 200 ) = 109.60  2.77; LCL = 106.8, UCL = 112.4

243


10.44 x  z  / 2  / n = 227.48  2.575(50/ 300 ) = 227.48  7.43; LCL = 220.05, UCL = 234.91

10.45 x  z  / 2  / n = 314,245  1.645(75,000/ 150 ) = 314,245  10,074; LCL = 304,171, UCL = 324,319

10.46 x  z  / 2  / n = 27.19  1.96(8/ 100 ) = 27.19  1.57; LCL = 25.62, UCL = 28.76

2

2

z   1.645  50  10.47 a. n    / 2  =   = 68 B 10     2

2

z   1.645 100  b. n    / 2  =   = 271 B 10     2

2

z   1.96  50  c. n    / 2  =   = 97  10   B  2

2

z   1.645  50  d. n    / 2  =   = 17 20    B 

10.48 a The sample size increases. b The sample size increases. c The sample size decreases.

2

2

z   2.575  250  10.49 a. n    / 2  =   = 166 B 50     2

2

2

2

z   2.575  50  b. n    / 2  =   =7 B 50    

z   1.96  250  c. n    / 2  =   = 97 B 50     2

2

z   2.575  250  d. n    / 2  =   = 4,145 10    B 

10.50 a The sample size decreases. b The sample size decreases. c The sample size increases.

244


2

2

z   1.645 10  10.51 a. n    / 2  =   = 271 1    B 

b. 150  1

10.52 a. x  z  / 2 b. x  z  / 2

 150  1.645

5

 150  .5

271

n

 150  1.645

n

20

 150  2

271

10.53 a. The width of the confidence interval estimate is equal to what was specified. b. The width of the confidence interval estimate is smaller than what was specified. c. The width of the confidence interval estimate is larger than what was specified.

2

2

z   1.96  200  10.54 a. n    / 2  =   = 1,537 10    W 

b. 500  10

10.55 a. x  z  / 2 b. x  z  / 2

 500  1.96

100

 500  5

1537

n

 500  1.96

n

400

 500  20

1537

10.56 a The width of the confidence interval estimate is equal to what was specified. b The width of the confidence interval estimate is smaller than what was specified. c The width of the confidence interval estimate is larger than what was specified.

2

2

10.57

z   1.645  10  n    / 2  =   = 68 2    B 

10.58

z   2.575  360  n    / 2  =   = 2,149 20    B 

10.59

z   1.96  12  n    / 2  =   = 139 2    B 

2

2

2

2

245


2

2

10.60

z   1.645  20  n    / 2  =   = 1,083 1    B 

10.61

z   1.96  25  n    / 2  =   = 97 5    B 

10.62

z   1.96  15  n    / 2  =   = 217 2    B 

2

2

2

2

246


Chapter 11 11.1

H 0 : The drug is not safe and effective H 1 : The drug is safe and effective

11.2

H 0 : I will complete the Ph.D. H 1 : I will not be able to complete the Ph.D.

11.3

H 0 : The batter will hit one deep H 1 : The batter will not hit one deep

11.4

H 0 : Risky investment is more successful H 1 : Risky investment is not more successful

11.5

H 1 : The plane is on fire H 1 : The plane is not on fire

11.6 The defendant in both cases was O. J. Simpson. The verdicts were logical because in the criminal trial the amount of evidence to convict is greater than the amount of evidence required in a civil trial. The two juries concluded that there was enough (preponderance of) evidence in the civil trial, but not enough evidence (beyond a reasonable doubt) in the criminal trial.

All p-values and probabilities of Type II errors were calculated manually using Table 3 in Appendix B.

11.7 Rejection region: z < z .005  2.575 or z > z.005 = 2.575

z

x  / n

980  1000

 1.00

200 / 100

p-value = 2P(Z < –1.00) = 2(.1587) = .3174 There is not enough evidence to infer that   1000.

247


11.8 Rejection region: z > z.03 = 1.88

z

x  / n

51  50

 .60

5/ 9

p-value = P(Z > .60) = 1 – .7257 = .2743 There is not enough evidence to infer that  > 50.

11.9 Rejection region: z < z.10  1.28 z

x  / n

14 .3  15 2 / 25

 1.75

p-value = P(Z < –1.75) = .0401 There is enough evidence to infer that  < 15.

11.10 Rejection region: z < z .025  1.96 or z > z.025 = 1.96 z

x  / n

100  100 10 / 100

0

p-value = 2P(Z > 0) = 2(.5) = 1.00

248


There is not enough evidence to infer that   100.

11.11 Rejection region: z > z.01 = 2.33 z

x  / n

80  70 20 / 100

 5.00

p-value = p(z > 5.00) = 0 There is enough evidence to infer that  > 70.

11.12 Rejection region: z < z.05  1.645 z

x  / n

48  50 15 / 100

 1.33

p-value = P(Z < –1.33) = .0918 There is not enough evidence to infer that  < 50.

11.13 z 

x

/ n

240  250 40 / 70

 2.09

p-value = P(Z < -2.09) = .0182

249


11.14 z 

x

/ n

1525  1500 220 / 125

 1.27

p-value = 2P(Z > 1.27) = 2(1-P(Z < 1.27) = 2(1 - .8980) = 2(.1020) = .2040

11.15 z 

x

/ n

8.5  7.5

 3.65

1.5 / 30

p-value = P(Z > 3.65) = 0.

11.16 z 

x

1.5  0

/ n

 1.50

10 / 100

p-value = P(Z > 1.50) = 1 – P(Z < 1.5) = 1 - .9332 = .0668

11.17 z 

x

/ n

2.3  0 25 / 400

 1.84

p-value = P(Z < -1.84) = .0329

11.18 z 

x

/ n

5.5  0 50 / 90

 1.04

p-value = 2P(Z < -1.04) = 2(.1492) = .2984

11.19 z 

x

/ n

4  (5) 5 / 25

 1.00

p-value =2P(Z > 1.00) = 2(1 – P(Z < 1.00) = 2(1 - .8413) = .3174

11.20 a. No because the test statistic will be negative. b The p-value will be larger than .5.

11.21a. z 

x  / n

52  50

 1.20

5/ 9

p-value = P(Z > 1.20) = 1 – .8849 = .1151 b. z 

x  / n

52  50 5 / 25

 2.00

p-value = P(Z > 2.00) = .5 – .4772 = .0228. c. z 

x  / n

52  50 5 / 100

 4.00

p-value = P(Z > 4.00) = 0.

250


d. The value of the test statistic increases and the p-value decreases.

11.22a. z 

x  / n

190  200

 .60

50 / 9

p-value = P(Z < –.60) = .5 – .2257 = .2743 b. z 

x 

190  200

/ n

 1.00

30 / 9

p-value = P(Z < –1.00) = .1587 c z

x 

/ n

190  200

 3.00

10 / 9

p-value = P(Z < –3.00) = .0013 d. The value of the test statistic decreases and the p-value decreases. x 

11.23 a. z 

/ n

21  20 5 / 25

 1.00

p-value = 2P(Z > 1.00) = 2(1 – .8413) = .3174 b. z 

x  / n

22  20

 2.00

5 / 25

p-value = 2P(Z > 2.00) = 2(1 – .9772) = .0456 c. z 

x 

/ n

23  20

 3.00

5 / 25

p-value = 2P(Z > 3.00) = 2(1 – .9987) = .0026 d. The value of the test statistic increases and the p-value decreases. x 

11.24 a. z 

/ n

99  100 8 / 100

 1.25

p-value = 2P(Z < –1.25) = 2(.1056) = .2112 b. z 

x  / n

99  100

 .88

8 / 50

p-value = 2P(Z < –.88) = 2(.1894) = .3788 c. z 

x  / n

99  100

 .56

8 / 20

p-value = 2P(Z < –.56) = 2(.2877) = .5754 d. The value of the test statistic increases and the p-value increases.

11.25 a. z 

x  / n

990  1000

 4.00

25 / 100 251


p-value = P(Z < –4.00) = 0 b. z 

x  / n

990  1000

 2.00

50 / 100

p-value = P(Z < –2.00) = .0228 c. z 

x 

/ n

990  1000

 1.00

100 / 100

p-value = P(Z < –1.00) = .1587 d. d. The value of the test statistic increases and the p-value increases.

x 

11.26 a. z 

/ n

72  60

 3.00

20 / 25

p-value = P(Z > 3.00) = 1 – .9987 = .0013 b. z 

x  / n

68  60

 2.00

20 / 25

p-value = P(Z > 2.00) = 1 – .9772 = .0228 c. z 

x  / n

64  60

 1.00

20 / 25

p-value = P(Z > 1.00) = 1 – .8413 = .1587 d. The value of the test statistic decreases and the p-value increases.

11.27 a z 

x  / n

178  170 65 / 200

 1.74

p-valu.e = P(Z > 1.74) = 1 – .9591 = .0409 b. z 

x  / n

178  170 65 / 100

 1.23

p-value = P(Z > 1.23) = 1 – .8907 = .1093 c. The value of the test statistic increases and the p-value decreases.

11.28 a z 

x  / n

178  170 35 / 400

 4.57

p-value = P(Z > 4.57) = 0. b z

x  / n

178  170 100 / 400

 1.60

p-value = P(Z > 1.60) = 1 – .9452 = .0548 The value of the test statistic decreases and the p-value increases.

11.29 a Yes, because the test statistic will be positive.

252


b The p-value will be larger than .5. x 

11.30 a z 

/ n

21.63  22

 .62

6 / 100

p-value = P(Z < –.62) = .2676 bz 

x  / n

21 .63  22 6 / 500

 1.38

p-value = P(Z < –1.38) = .0838 The value of the test statistic decreases and the p-value decreases.

11.31 a z 

x  / n

21 .63  22

3 / 220

 1.83

p-value = P(Z < –1.83) = .0336 bz 

x  / n

21.63  22 12 / 220

 .46

p-value = P(Z < –.46) = .3228 The value of the test statistic increases and the p-value increases.

11.32

x  22

z

x

p-value

6 / 220 22.0 21.8 21.6 21.4 21.2 21.0 20.8 20.6 20.4

0 –.49 –.99 –1.48 –1.98 –2.47 –2.97 –3.46 –3.96

x 

11.33 a z 

/ n

.5 .3121 .1611 .0694 .0239 .0068 .0015 0 0

17 .55  17 .09

 .84

3.87 / 50

p-value = 2P(Z > .84) = 2(1 – .7995) = 2(.2005) = .4010 bz 

x  / n

17.55  17.09 3.87 / 400

 2.38

p-value = 2P(Z > 2.38) = 2(1 – .9913) = 2(.0087) = .0174 The value of the test statistic increases and the p-value decreases.

11.34 a z 

x  / n

17 .55  17 .09

 2.30

2 / 100

p-value = 2P(Z > 2.30) = 2(1 – .9893) = 2(.0107) = .0214 253


x 

bz

/ n

17 .55  17 .09

 .46

10 / 100

p-value = 2P(Z > .46) = 2(1 – .6772) = 2(.3228) = .6456 The value of the test statistic decreases and the p-value increases.

11.35a

11.36

x  17 .09

x

z

15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0

3.87 / 100 –5.40 –4.11 –2.82 –1.52 –.23 1.06 2.35 3.64 4.94

p-value 0 0 .0048 .1286 .8180 .2892 .0188 0 0

H0 :  = 5

H1 :  > 5 z

x  / n

65 1.5 / 10

 2.11

p-value = P(Z > 2.11) = 1 – .9826 = .0174 There is enough evidence to infer that the mean is greater than 5 cases.

11.37

H 0 :  = 50

H1 :  > 50 z

x  / n

59 .17  50 10 / 18

 3.89

p-value = P(Z > 3.89) = 0 There is enough evidence to infer that the mean is greater than 50 minutes.

11.38

H 0 :  = 12

H1 :  < 12 z

x  / n

11.00  12 3 / 15

 1.29

p-value = P(Z < –1.29) = .0985 There is enough evidence to infer that the average number of golf balls lost is less than 12.

11.39

H 0 :  = 36

254


H1 :  < 36 z

x  / n

34 .25  36 8 / 12

 .76

p-value = P(Z < –.76) = .2236 There is not enough evidence to infer that the average student spent less time than recommended.

11.40

H0 :  = 6

H1 :  > 6 z

x  / n

6.60  6

 .95

2 / 10

p-value = P(Z > .95) = 1 – .8289 = .1711 There is not enough evidence to infer that the mean time spent putting on the 18 th green is greater than 6 minutes.

11.41

H 0 :  = .50

H1 :   .50 z

x  / n

.493  .50 .05 / 10

 .44

p-value = 2P(Z < –.44) = 2(.3300) = .6600 There is not enough evidence to infer that the mean diameter is not .50 inch.

11.42

H 0 :  = 25

H1 :  > 25 z

x  / n

30 .22  25 12 / 18

 1.85

p-value = P(Z > 1.85) = 1 – .9678 =.0322 There is not enough evidence to conclude that the manager is correct.

11.43

H 0 :  = 5,000

H1 :  > 5,000 z

x  / n

5,065  5,000

 1.62

400 / 100

p-value = P(Z > 1.62) = 1 – .9474 =.0526 There is not enough evidence to conclude that the claim is true.

255


11.44

H 0 :  = 30,000

H1 :  < 30,000

z

x  / n

29 ,120  30 ,000

 2.06

8,000 / 350

p-value = P(Z < –2.06) = .0197 There is enough evidence to infer that the president is correct

11.45

H 0 :  = 560

H1 :  > 560 z

x  / n

569 .0  560

 .80

50 / 20

p-value = P(Z > .80) = 1 – .7881 = .2119 There is not enough evidence to conclude that the dean’s claim is true.

11.46a

H 0 :  = 17.85

H1 :  > 17.85 z

x  / n

19 .13  17 .85

 1.65

3.87 / 25

p-value = P(Z > 1.65) = 1 – .9505 = .0495 There is enough evidence to infer that the campaign was successful. b We must assume that the population standard deviation is unchanged.

11.47

H0 :  = 0

H1 :  < 0 z

x  / n

1.20  0

 1.41

6 / 50

p-value = P(Z < –1.41) = .0793 There is not enough evidence to conclude that the safety equipment is effective.

11.48

H 0 :  = 55

H1 :  > 55 z

x  / n

55 .80  55

 2.26

5 / 200

p-value = P(Z > 2.26) = 1 – .9881 = .0119 There is not enough evidence to support the officer’s belief. 256


11.49

H0 :  = 4

H1 :  > 4 z

x  / n

5.04  4

 4.90

1.5 / 50

p-value = P(Z > 4.90) = 0 There is enough evidence to infer that the expert is correct.

11.50

H 0 :  = 20

H1 :  < 20 z

x  / n

19 .39  20

 1.22

3 / 36

p-value = P(Z < –1.22) = .1112 There is not enough evidence to infer that the manager is correct.

11.51

H 0 :  = 100

H1 :  > 100 z

x  / n

105 .7  100

 2.25

16 / 40

p-value = P(Z > 2.25) = 1 – .9878 = .0122 There is not enough evidence to infer that the site is acceptable.

11.52

H0 :  = 4

H1 :   4 z

x  / n

4.84  4

 3.33

2 / 63

p-value = 2P(Z > 3.33) = 0 There is enough evidence to infer that the average Alpine skier does not ski 4 times per year.

11.53

H0 :  = 5

H1 :  > 5 z

x  / n

5.64  5 2 / 25

 1.60

p-value = P(Z > 1.60) = 1 – .9452 = .0548 There is enough evidence to infer that the golf professional’s claim is true.

257


11.54

H 0 :  = 32

H1 :  < 32 z

x  / n

29 .92  32

 2.73

8 / 110

p-value = P(Z < –2.73) = 1– .9968 = .0032 There is enough evidence to infer that there has been a decrease in the mean time away from desks. A type I error occurs when we conclude that the plan decreases the mean time away from desks when it actually does not. This error is quite expensive. Consequently we demand a low pvalue. The p-value is small enough to infer that there has been a decrease.

11.55

H 0 :  = 230

H1 :  > 230 z

x  / n

231 .56  230 10 / 100

 1.56

p-value = P(Z > 1.56) = 1 – .9406 = .0594 There is not enough evidence to infer that Nike is correct. 11.56

H 0 :  = 10

H1 :  > 10 z

x

/ n

10.44  10 3 / 174

 1.93

p-value = P(Z > 1.93) = 1 – .9732 = .0268 There is enough evidence to infer that heating costs increased faster than inflation.

11.57

H 0 :  = 30

H1 :   30 z

x

/ n

29 .51  30

 1.63

5 / 277

p-value = 2P(Z < -1.63) = 2(.0516) = .1032 There is not enough evidence to infer that the average mean monthly expenditures on bakery products is not equal to $30.

11.58

H 0 :  = 125,000

H1 :  > 125,000

258


z

x

/ n

126 ,837  125 ,000 25,000 / 410

 1.49

p-value = P(Z > 1.49) = 1 – .9319 = .0681 There is not enough evidence to infer that mean value of 401k accounts is greater than $125,000.

11.59

H 0 :  = 7500

H1 :  > 7500 z

x

/ n

7625  7500 1200 / 163

 1.33

p-value = P(Z > 1.33) = 1 – .9082 = .0918 There is not enough evidence to infer that the mean income exceeds $7500.

11.60 Rejection region:

x  200

x 

> z / 2 or

/ n

> z .025  1.96 or

10 / 100 x > 201.96 or x < 198.04

x  200 10 / 100

x  / n

< z  / 2

< –1.96

 = P(198.04 < x < 201.96 given  = 203)

 198 .04  203 x  201 .96  203   = P( –4.96 < z < –1.04) = .1492 – 0 = .1492 = P    / n 10 / 100   10 / 100

11.61 Rejection region:

x  / n

> z

x  1000

> z.01  2.33 50 / 25 x > 1023.3

 x   1023 .3  1050   = P(z < –2.67) = .0038   = P( x < 1023.3 given  = 1050) = P 50 / 25   / n

11.62 Rejection region:

x  50 10 / 40 x < 47.40

x  / n

< z 

< z.05  1.645

 x  47 .40  48   = P(z > –.38) = 1 − .3520 = .6480   = P( x > 47.40 given  = 48) = P  10 / 40  / n

11.63 Exercise 11.48 259


Exercise 11.49

Exercise 11.50

260


11.64 a. Rejection region:

x  100 20 / 100 x > 102.56

x  / n

> z

> z .10  1.28

 x   102 .56  102   = P(z < .28) = .6103   = P( x < 102.56 given  = 102) = P  20 / 100  / n x  b. Rejection region: > z / n x  100 > z .02  2.55 20 / 100 x > 104.11  x   104 .11  102   = P(z < 1.06) = .8554   = P( x < 104.11 given  = 102) = P  20 / 100  / n c.  increases.

11.65 a. Rejection region:

x  / n

< z 

x  40

< z .05  1.645 5 / 25 x < 38.36

 x  38 .36  37   = P(z > 1.36) = 1 – .9131 = .0869   = P( x > 38.36 given  = 37) = P  5 / 25  / n x  < z  b. Rejection region: / n x  40 < z .15  1.04 5 / 25 x < 38.96  x  38 .96  37   = P(z > 1.96) = 1 – .9750 = .0250   = P( x > 38.96 given  = 37) = P  5 / 25  / n c.  decreases.

11.66 Exercise 11.52 a

261


Exercise 11.52 b

Exercise 11.53 a

262


Exercise 11.53 b

11.67 a. Rejection region:

x  / n

< z 

x  200

< z .10  1.28 30 / 25 x < 192.31  x   192 .31  196   = P(z > –.62) = 1 − .2676 = .7324   = P( x > 192.31 given  = 196) = P  30 / 25  / n x  b. Rejection region: < z  / n x  200 < z .10  1.28 30 / 100 x < 196.16  x   196 .16  196   = P(z > .05) = 1 – .5199 = .4801   = P( x > 196.16 given  = 196) = P  30 / 100  / n

263


c.  decreases.

11.68 a. Rejection region:

x  / n

> z

x  300

> z .05  1.645 50 / 81 x > 309.14  x   309 .14  310   = P(z < –.15) = .4404   = P( x < 309.14 given  = 310) = P  50 / 81  / n x  > z b. Rejection region: / n x  300 > z .05  1.645 50 / 36 x > 313.71  x   313 .71  310   = P(z < .45) = .6736   = P( x < 313.71 given  = 310) = P  50 / 36  / n c.  increases.

11.69 Exercise 11.55 a

Exercise 11.55 b

264


Exercise 11.56 a

Exercise 11.56 b

265


11.70

11.71 266


11.72

H 0 :  = 170

H1 :  < 170 A Type I error occurs when we conclude that the new system is not cost effective when it actually is. A Type II error occurs when we conclude that the new system is cost effective when it actually is not.

The test statistic is the same. However, the p-value equals 1 minus the p-value calculated Example 11.1. That is, p-value = 1 – .0069 = .9931 We conclude that there is no evidence to infer that the mean is less than 170. That is, there is no evidence to infer that the new system will not be cost effective.

11.73 Rejection region:

x0 6 / 50

x  / n

<  z

< z .10  1.28

x < –1.09

 x   1.09  (2)   = P(z > 1.07) = 1 – .8577 = .1423   = P( x > –1.09 given  = –2) = P  6 / 50  / n  can be decreased by increasing  and/or increasing the sample size.

267


11.74 Rejection region:

x  22

x  / n

< z 

< z .10  1.28

6 / 220 x < 21.48

 x  21 .48  21   = P(z > 1.19) =1 – .8830 = .1170   = P( x > 21.48 given  = 21) = P  6 / 220  / n

The company can decide whether the sample size and significance level are appropriate.

11.75 Rejection region:

x  / n

> z

x  100

> z .01  2.33 16 / 40 x > 105.89  x   105 .89  104   = P(z < .75) = .7734   = P( x < 105.89 given  = 104) = P  16 / 40  / n

11.76 Rejection region:

x  32

x  / n

<  z

< z .05  1.645

8 / 110 x < 30.75

 x  30 .75  30 )   = P(z > .98) = 1 – .8365 = .1635   = P( x > 30.75 given  = 30) = P  8 / 110  / n  can be decreased by increasing  and/or increasing the sample size.

11.77 i Rejection region:

x  10 3 / 100

x  / n

<  z

< z.01  2.33

x < 9.30

 x  9.30  9   = P(z > 1) = 1 – .8413 = .1587   = P( x > 9.30 given  = 9) = P    / n 3 / 100  ii Rejection region:

x  10 3 / 75

x  / n

<  z

< z.05  1.645

268


x < 9.43

 x  9.43  9   = P(z > 1.24) = 1 – .8925 = .1075   = P( x > 9.43 given  = 9) = P    / n 3 / 75 

iii Rejection region:

x  10 3 / 50

x  / n

<  z

< z.10  1.28

x < 9.46

 x  9.46  9   = P(z > 1.08) = 1 – .8599 = .1401   = P( x > 9.46 given  = 9) = P    / n 3 / 50  Plan ii has the lowest probability of a type II error.

11.78 A Type I error occurs when we conclude that the site is feasible when it is not. The consequence of this decision is to conduct further testing. A Type II error occurs when we do not conclude that a site is feasible when it actually is. We will do no further testing on this site, and as a result we will not build on a good site. If there are few other possible sits, this could be an expensive mistake.

11.79

H 0 :   20 H1 :   25

Rejection region:

x  / n

> z

x  20

> z .01  2.33 8 / 25 x > 23.72

 x  23 .72  25   = P(z < –.80) = .2119   = P( x < 23.72 given  = 25) = P  8 / 25  / n The process can be improved by increasing the sample size.

269


Chapter 12 12.1a x  t  / 2 s / n = 70  2.004(12/ 56 ) = 70 ± 3.21; LCL = 66.79, UCL = 73.21 b x  t  / 2 s / n = 35 ± 2.004(12/ 56 ) = 30 ± 3.21; LCL = 26.79, UCL = 33.21 c. The interval width does not change.

12.2a x  t  / 2 s / n = 50 ± 1.711(10/ 25 ) = 50 ± 3.42; LCL = 46.58, UCL = 53.42 b x  t  / 2 s / n = 100 ± 1.711(10/ 25 ) = 100 ± 3.42; LCL = 96.58, UCL = 103.42 c. The interval width does not change.

12.3 a x  t  / 2 s / n = 510 ± 2.064(125/ 25 ) = 510 ± 51.60; LCL = 458.40, UCL = 561.60 b x  t  / 2 s / n = 510 ± 2.009(125/ 50 ) = 510 ± 35.51; LCL = 474.49, UCL = 545.51 c x  t  / 2 s / n = 510 ± 1.984(125/ 100 ) = 510 ± 24.80; LCL = 485.20, UCL = 534.80 d. The interval narrows.

12.4 a x  t  / 2 s / n = 1,500 ± 1.984(300/ 100 ) = 1,500 ± 59.52; LCL = 1,440.48, UCL = 1,559.52 b x  t  / 2 s / n = 1,500 ± 1.984(200/ 100 ) = 1,500 ± 39.68; LCL = 1,460.32, UCL = 1,539.68 c x  t  / 2 s / n = 1,500 ± 1.984(100/ 100 ) = 1,500 ± 19.84; LCL = 1,480.16, UCL = 1,519.84 d. The interval narrows.

12.5 a x  t  / 2 s / n = 700 ± 1.645(100/ 400 ) = 700 ± 8.23; LCL = 691.77, UCL = 708.23 b x  t  / 2 s / n = 700 ± 1.96(100/ 400 ) = 700 ± 9.80; LCL = 690.20, UCL = 709.80 a x  t  / 2 s / n = 700 ± 2.576(100/ 400 ) = 700 ± 12.88; LCL = 687.12, UCL = 712.88 d. The interval widens.

12.6 a x  t  / 2 s / n = 10 ± 1.984(1/ 100 ) = 10 ± .20; LCL = 9.80, UCL = 10.20 b x  t  / 2 s / n = 10 ± 1.984(4/ 100 ) = 10 ± .79; LCL = 9.21, UCL = 10.79 c x  t  / 2 s / n = 10 ± 1.984(10/ 100 ) = 10 ± 1.98; LCL = 8.02, UCL = 11.98 d The interval widens.

251


12.7 a x  t  / 2 s / n = 120 ± 2.009(15/ 51 ) = 120 ± 4.22; LCL = 115.78, UCL = 124.22 b x  t  / 2 s / n = 120 ± 1.676(15/ 51 ) = 120 ± 3.52; LCL = 116.48, UCL = 123.52 c x  t  / 2 s / n = 120 ± 1.299(15/ 51 ) = 120 ± 2.73; LCL = 117.27, UCL = 122.73 d The interval narrows.

12.8 a x  t  / 2 s / n = 63 ± 1.990(8/ 81 ) = 63 ± 1.77; LCL = 61.23, UCL = 64.77 b x  t  / 2 s / n = 63 ± 2.000(8/ 64 ) = 63 ± 2.00; LCL = 61.00, UCL = 65.00 c x  t  / 2 s / n = 63 ± 2.030(8/ 36 ) = 63 ± 2.71; LCL = 60.29, UCL = 65.71 d The interval widens.

H 0 :  = 20

12.9

H1 :  > 20 a Rejection region: t  t , n 1  t.05,9 = 1.833

x 

t

s/ n

23  20

 1.05, p-value = .1597. There is not enough evidence to infer that the population mean is

9 / 10

greater than 20. b Rejection region: t  t , n 1  t.05, 29 = 1.699

x 

t

s/ n

23  20

 1.83, p-value = .0391. There is enough evidence to infer that the population mean is greater

9 / 30

than 20. c Rejection region: t  t , n 1  t .05, 49  1.676

x 

t

s/ n

23  20

 2.36, p-value = .0112. There is enough evidence to infer that the population mean is greater

9 / 50

than 20. d As the sample size increases the test statistic increases [and the p-value decreases].

12.10

H 0 :  = 180

H1 :   180 Rejection region: t   t  / 2, n 1   t .025,199  1.972 or t  t  / 2,n 1  t .025,199 = 1.972 at 

x 

s/ n

175  180

 3.21, p-value = .0015. There is enough evidence to infer that the population mean is not

22 / 200

equal to 180.

252


x 

b t

s/ n

175  180

 1.57 , p-value = .1177. There is not enough evidence to infer that the population mean is

45 / 200

not equal to 180.

x 

c t

s/ n

175  180

 1.18, p-value = .2400. There is not enough evidence to infer that the population mean is

60 / 200

not equal to 180. d. As s increases, the test statistic increases and the p-value increases. 12.11 Rejection region: t   t ,n 1   t .05,99  1.660

x 

at 

145  150

s/ n

 1.00, p-value = .1599. There is not enough evidence to infer that the population mean is

50 / 100

less than 150.

x 

bt 

s/ n

140  150

 2.00, p-value = .0241. There is enough evidence to infer that the population mean is less

50 / 100

than 150.

x 

ct 

s/ n

135  150

 3.00, p-value = .0017. There is enough evidence to infer that the population mean is less

50 / 100

than 150 d The test statistics decreases and the p-value decreases.

12.12

H0: µ = 50 H1: µ ≠ 50

a Rejection region: t   t  / 2,n 1   t .05, 24  1.711 or t  t  / 2,n 1  t .05, 24  1.711

t

x 

s/ n

52  50

 .67 , p-value = .5113. There is not enough evidence to infer that the population mean is not

15 / 25

equal to 50. b Rejection region: t   t  / 2,n 1   t .05,14  1.761 or t  t  / 2, n 1  t .05,14  1.761

t

x 

s/ n

52  50

 .52, p-value = .6136. There is not enough evidence to infer that the population mean is not

15 / 15

equal to 50. c Rejection region: t   t  / 2,n 1   t .05, 4  2.132 or t  t  / 2,n 1  t .05, 4  2.132

t

x 

s/ n

52  50

 .30, p-value = .7804. There is not enough evidence to infer that the population mean is not

15 / 5

equal to 50. d The test statistic decreases and the p-value increases.

253


12.13 Rejection region: t   t , n 1   t .10, 49  1.299 at 

x 

s/ n

585  600

 2.36, p-value = .0112. There is enough evidence to infer that the population mean is less

45 / 50

than 600. bt 

x 

s/ n

590  600

 1.57 , p-value = .0613. There is enough evidence to infer that the population mean is less

45 / 50

than 600. ct 

x 

s/ n

595  600

 .79, p-value = .2179. There is not enough evidence to infer that the population mean is

45 / 50

less than 600. d The test statistic increases and the p-value increases. 12.14 Rejection region: t  t , n 1  t .01,99  2.364 at 

x 

s/ n

106  100

 1.71, p-value = .0448. There is not enough evidence to infer that the population mean is

35 / 100

greater than 100. bt 

x 

s/ n

106  100

 2.40, p-value = .0091. There is enough evidence to infer that the population mean is

25 / 100

greater than 100. ct 

x  s/ n

106  100

 4.00, p-value = .0001. There is enough evidence to infer that the population mean is

15 / 100

greater than 100 d The test statistic increases and the p-value decreases.

12.15 a x  t  / 2 s / n = 40  2.365(10/ 8 ) = 40 ± 8.36; LCL = 31.64, UCL = 48.36 b x  z  / 2  / n = 40  1.96(10/ 8 ) = 40 ± 6.93; LCL = 33.07, UCL = 46.93 c The student t distribution is more widely dispersed than the standard normal; thus, z  / 2 is smaller than t  / 2 .

12.16 a x  t  / 2 s / n = 175  2.132(30/ 5 ) = 175± 28.60; LCL = 146.40, UCL = 203.60 b x  z  / 2  / n = 175  1.645(30/ 5 ) = 175± 22.07; LCL = 152.93, UCL = 197.07 c The student t distribution is more widely dispersed than the standard normal; thus, z  / 2 is smaller than t  / 2 .

12.17 a x  t  / 2 s / n = 15,500  1.645(9,950/ 1,000 ) = 15,500 ± 517.59; LCL = 14,982.41, UCL = 16,017.59 b x  z  / 2  / n = 15,500  1.645(9,950/ 1,000 ) = 15,500 ± 517.59; LCL = 14,982.41, UCL = 16,017.59 254


c With n = 1,000 the student t distribution with 999 degrees of freedom is almost identical to the standard normal distribution.

12.18 a x  t  / 2 s / n = 350  2.576(100/ 500 ) = 350± 11.52; LCL = 338.48, UCL = 361.52 b x  z  / 2  / n = 350  2.575(100/ 500 ) = 350 ± 11.52; LCL = 338.48, UCL = 361.52 c With n = 500 the student t distribution with 999 degrees of freedom is almost identical to the standard normal distribution.

12.19

H0: µ = 70 H1: µ > 70

a Rejection region: t  t , n 1  t .05,10 = 1.812

t

x 

74 .5  70

s/ n

 1.66, p-value = .0641. There is not enough evidence to infer that the population mean is

9 / 11

greater than 70. b Rejection region: z  z   z .05 = 1.645

z

x  / n

74 .5  70

9 / 11

 1.66, p-value = P(Z > 1.66) = 1 – P(Z< 1.66) = 1 – .9515 = .0485. There is enough

evidence to infer that the population mean is greater than 70. c The Student t distribution is more dispersed than the standard normal.

12.20

H0: µ = 110 H1: µ < 110

a Rejection region: t   t , n 1   t .10,9 = –1.383

t

x 

103  110

s/ n

 1.30, p-value = .1126. There is not enough evidence to infer that the population mean is

17 / 10

less than 110. b Rejection region: z  z   z .10 = –1.28

z

x  / n

103  110 17 / 10

 1.30, p-value = P(Z < –1.30) = .0968. There is enough evidence to infer that the

population mean is less than 110. c The Student t distribution is more dispersed than the standard normal.

12.21

H0: µ = 15 H1: µ < 15

a Rejection region: t   t ,n 1   t .05,1499 = –1.645

255


t

x 

14  15

s/ n

 1.55, p-value = .0608. There is not enough evidence to infer that the population mean is

25 / 1,500

less than 15. b Rejection region: z  z   z .05 = –1.645

z

x  / n

14  15

25 / 1,500

 1.55, p-value = P(Z < –1.55) = .0606. There is not enough evidence to infer that the

population mean is less than 15. c With n = 1,500 the student t distribution with 1,499 degrees of freedom is almost identical to the standard normal distribution. 12.22 a. Rejection region: t  t  , n 1  t .05,999 = 1.645

t

x 

405  400

s/ n

 1.58, p-value = .0569. There is not enough evidence to infer that the population mean is

100 / 1,000

less than 15. b Rejection region: z  z   z .05 = 1.645

z

x

/ n

405  400 100 / 1,000

 1.58, p-value = P( Z > 1.58) = 1 – .9429 = .0571. There is not enough evidence to infer

that the population mean is less than 15. c With n = 1,000 the student t distribution with 999 degrees of freedom is almost identical to the standard normal distribution.

12.23

H0: µ = 6 H1: µ < 6

a Rejection region: t   t , n 1   t .05,11 = –1.796

t

x 

s/ n

5.69  6

 .68, p-value = .2554. There is not enough evidence to support the courier’s

1.58 / 12

advertisement.

12.24 x  t  / 2 s / n = 24,051  2.145(17,386/ 15 ) = 24,051  9,628; LCL = 14,422, UCL = 33,680

12.25

H0: µ = 20 H1: µ > 20

Rejection region: t  t , n 1  t .05,19  1.729

t

x  s/ n

20 .85  20

 .56, p-value = .2902. There is not enough evidence to support the doctor’s claim.

6.76 / 20 256


12.26

H 0: µ = 8 H 1: µ < 8

Rejection region: t   t ,n 1   t.01,17  2.567

t

x 

7.91  8 .085 / 18

s/ n

 –4.49, p-value = .0002. There is enough evidence to conclude that the average container is

mislabeled.

12.27 x  t  / 2 s / n = 18.13  2.145(9.75/ 15 ) = 18.13± 5.40; LCL = 12.73, UCL =23.53

12.28 x  t  / 2 s / n = 26.67  1.796(16.52/ 12 ) = 26.67± 8.56; LCL = 18.11, UCL = 35.23

12.29 x  t  / 2 s / n = 17.70  2.262(9.08/ 10 ) = 17.70± 6.49; LCL = 11.21, UCL =24.19

12.30

H0: µ = 10 H1: µ < 10

Rejection region: t   t ,n 1   t .10,9  1.383

t

x  s/ n

7.10  10

 2.45, p-value = .0185. There is enough evidence to infer that the mean proportion of

3.75 / 10

returns is less than 10%.

12.31 x  t  / 2 s / n = 7.15  1.972(1.65/ 200 ) = 7.15 ± .23; LCL = 6.92, UCL = 7.38

12.32 x  t  / 2 s / n = 4.66  2.576(2.37/ 240 ) = 4.66 ± .39; LCL = 4.27, UCL = 5.05 Total number: LCL = 100 million (4.27) = 427 million, UCL = 100 million (5.05) = 505 million

12.33 x  t  / 2 s / n =17.00  1.975(4.31/ 162 = 17.00 ±.67; LCL = 16.33, UCL = 17.67 Total number: LCL = 50 million (16.33) = 816.5 million, UCL = 50 million (17.67) = 883.5 million

12.34 x  t  / 2 s / n = 15,137  1.96(5,263/ 306 = 15,137 ±590; LCL = 14,547, UCL = 15,727 Total credit card debt: LCL = 50 million (14,547) = $727,350 million, UCL = 50 million (15,727) = $786,350 million

12.35a. x  t  / 2 s / n = 59.04  1.980(20.62/ 122 ) = 59.04 ± 3.70; LCL = 55.34, UCL = 62.74 257


Total spent on other products: LCL = 2800(55.34 = $154,952, UCL = 2800(62.74) = $175,672

12.36 x  t  / 2 s / n = 2.67  1.973(2.50/ 188 ) = 2.67 ± .36; LCL = 2.31, UCL = 3.03

12.37 x  t  / 2 s / n = 29.69  1.96(7.88/ 475 ) = 44.14 ± .71; LCL = 43.43, UCL = 44.85

12.38 x  t  / 2 s / n = 591.87  1.972(125.06/ 205 ) = 591.87 ± 17.22; LCL = 574.65, UCL = 609.09 Total cost of congestion: LCL = 171 million (574.65) = $98,265 million, UCL = 171 million (609.09) = $104,154 million

12.39 x  t  / 2 s / n = 3.79  1.960(4.25/ 564 ) = 3.79 ± .35; LCL = 3.44, UCL = 4.14 Package of 10: LCL = 13.65(10) = 136.5 days, UCL = 14.23(10) = 142.3 days.

12.40

H0: µ = 15 H1: µ > 15

Rejection region: t  t ,n 1  t.05,115  1.658

t

x 

s/ n

15 .27  15 5.72 / 116

 .51, p-value = .3061. There is not enough evidence to infer that the mean number of

commercials is greater than 15.

12.41 x  t  / 2 s / n = 4.34  1.960(4.22/ 950 ) = 4.34 ± .27; LCL = 4.07, UCL = 4.61 Total number of visits: LCL = 307,439,000(4.07) = 1,251,276,730

12.42

UCL = 307,439,000 (4.61) = 1,417,293,790

H0: µ = 85 H1: µ > 85

Rejection region: t  t  , n 1  t .05,84  1.663

t

x  s/ n

89 .27  85 17 .30 / 85

 2.28, p-value = .0127. There is enough evidence to infer that an e-grocery will be

successful.

12.43 x  t  / 2 s / n = 15.02  1.990(8.31/ 83 ) = 15.02  1.82; LCL = 13.20, UCL = 16.84

12.44 x  t  / 2 s / n = 96,100  1.960(34,468/ 473 ) = 96,100 ± 3106; LCL = 92,994, UCL = 99,206 Total amount of debt: LCL = 7 million(92,994) = 650,958 million 258

UCL = 7 million(99,206) = 694,442 million


12.45 x  t  / 2 s / n = 1.507  1.645(.640/ 473 ) = 1.507 ± .048; LCL = 1.459, UCL = 1.555

12.46 x  t  / 2 s / n = 27,852  1.96(9252/ 347 ) = 27,852 ± 977; LCL = 26,875, UCL = 28,829 Total: LCL = 43.3million (26,875) = $1,163,687,500,000, UCL = 43.3 million (28,829) = $1,248,295,700,000

12.47 x  t  / 2 s / n = 354.55  2.576(90.32/ 681 ) = 354.55 ± 8.94; LCL = 345.61, UCL = 363.49

12.48a x  t  / 2 s / n = 25,228  1.96(5544/ 184 ) = 25,228 ± 806; LCL = 24,422, UCL = 26,034 b x  t  / 2 s / n = 27,751  1.96(6098/ 184 ) = 27,751 ± 887; LCL = 26,864, UCL = 28,638

12.49 x  t  / 2 s / n = 366,203  1.96(122,277/ 452 ) = 366,203 ± 11,303; LCL = 354,900, UCL = 377,506 Total: 2,500,000(354,900) = $887,250,000,000, UCL = 2,500,000(377,506) = $943,765,000,000.

12.50 x  t  / 2 s / n = 46,699  1.96(9032/ 608 ) = 46,699 ± 719; LCL = 45,980, UCL = 47,418 Total: LCL = 112,236(45,980) = $5,160,611,280, UCL = 112,236(47,418) = 5,322,006,648.

12.51 x  t  / 2 s / n = 7.31  1.973(5.58/ 178 ) = 7.31 ± .83; LCL = 6.48, UCL = 8.14

12.52 x  t  / 2 s / n = 1157.77  1.645(396.51/ 325 ) = 1157.77 ± 36.28; LCL = 1121.49, UCL = 1194.05

12.53 x  t  / 2 s / n = 530.69  1.96(97.17/ 485 ) = 530.69 ± 8.67; LCL = 522.02, UCL = 539.36 Total: LCL = 24,000,000(522.02) = $12,528,480,000, UCL = 24,000,000(539.36) = $12,944,640,000

12.54a.

b The required condition is that the variable is normally distributed. c The histogram is somewhat bell shaped. 259


12.55a H0: µ = 12 H1: µ > 12

p-value = 0. There is enough evidence to infer that the average American completed high school. b The required condition is that the variable is normally distributed. c The histogram is bell shaped.

12.56

12.57a H0: µ = 40 H1: µ > 40

p-value = 0. There is enough evidence to infer that the average American is working more than 40 hours per week. b The required condition is that the variable is normally distributed. c The histogram is bell shaped.

12.58

260


12.59

H0: µ = 50.4 H1: µ > 50.4

p-value = 0. There is enough evidence to support the expectation. The required condition is that the variable is normally distributed. The histogram is bell shaped.

12.60a.

b The required condition is that the variable is normally distributed. c The histogram is positively skewed.

12.61a.

b The required condition is that the variable is normally distributed. c The histogram is positively skewed.

12.62a.

261


b The required condition is that the variable is normally distributed. The histogram is positively skewed.

12.63

12.64

12.65

12.66

H0: µ = 2625 H1: µ < 2625

262


p-value = 0. There is enough evidence to infer that the average middle class household spends less than $2625.

12.67

12.68

H0: µ = 12.9 H1: µ > 12.9

p-value = 0. There is enough evidence to infer that the average middle class head of household has more education that 12.9 years,

12.69

12.70

H0: σ2 = 300 H1: σ2 ≠ 300

263


a Rejection region:  2  12 / 2,n 1   .2975,99  74 .2 or  2   2 / 2,n 1   .2025,99  130

2 

(n  1)s 2 

2

=

(100  1)( 220 ) = 72.60, p-value = .0427. There is enough evidence to infer that the population 300

variance differs from 300. b Rejection region:  2  12 / 2,n 1   .2975,49  32.4 or  2   2 / 2,n 1   .2025,49  71 .4

2 

(n  1)s 2 

2

=

(50  1)( 220 ) = 35.93, p-value = .1643. There is not enough evidence to infer that the population 300

variance differs from 300. c Decreasing the sample size decreases the test statistic and increases the p-value of the test. H0: σ2 = 100

12.71

H1: σ2 < 100 a Rejection region:  2  12,n 1   .299,49  29 .7

2 

(n  1)s 2 

2

=

(50  1)(80 ) = 39.20, p-value = .1596. There is not enough evidence to infer that the population 100

variance is less than 100. b Rejection region:  2  12,n 1   .299,99  70 .1

2 

(n  1)s 2 2

=

(100  1)(80 ) = 79.20, p-value = .0714. There is not enough evidence to infer that the population 100

variance is less than 100. c Increasing the sample size increases the test statistic and decreases the p-value.

12.72 a LCL =

UCL =

 2 / 2,n 1

(n  1)s 2 12 / 2, n 1

b LCL =

UCL =

(n  1)s 2

(n  1)s 2  2 / 2,n 1

(n  1)s 2 12 / 2, n 1

=

=

=

=

(n  1)s 2  .205,14

=

(15  1)(12 ) = 7.09 23 .7

(n  1)s 2 (15  1)(12 ) = 25.57 6.57  .295,14

(n  1)s 2  .205, 29 (n  1)s 2  .295, 29

=

(30  1)(12 ) = 8.17 42 .6

=

(30  1)(12 ) = 19.66 17 .7

c Increasing the sample size narrows the interval.

264


12.73 LCL =

UCL =

(n  1)s 2

=

 2 / 2,n 1

(n  1)s 2 12 / 2, n 1

=

(n  1)s 2  .205,7

(n  1)s 2  .295,7

=

=

(8  1)(. 00093 ) = .00046, 14 .1

(8  1)(. 00093 ) = .00300 2.17

H0: σ2 = 250

12.74

H1: σ2 < 250 Rejection region:  2  12,n 1   .290,9  4.17

2 

(n  1)s 2 

2

=

(10  1)( 210 .22 )  7.57 , p-value = .4218. There is not enough evidence to infer that the population 250

variance has decreased. H0: σ2 = 23

12.75

H1: σ2 ≠ 23 Rejection region:  2  12 / 2,n 1   .295,7  2.17 or  2   2 / 2,n 1   .205,7  14 .1

2 

(n  1)s 2 

2

=

(8  1)(16 .50 )  5.02 , p-value = .6854. There is not enough evidence to infer that the population 23

variance has changed.

12.76 LCL =

UCL =

(n  1)s 2  2 / 2,n 1

(n  1)s 2 12 / 2, n 1

=

=

(n  1)s 2  .2025,9

(n  1)s 2  .2975,9

=

=

(10  1)(15 .43)  7.31 19 .0

(10  1)(15 .43)  51 .43 2.70

12.77 a H0: σ2 = 250 H1: σ2 ≠ 250 Rejection region:  2  12 / 2,n 1   .2975,24  12.4 or  2   2 / 2,n 1   .2025,24  39.4

2 

(n  1)s 2 

2

=

(25  1)( 270 .58 ) = 25.98, p-value = .7088. There is not enough evidence to infer that the 250

population variance is not equal to 250. b Demand is required to be normally distributed. c The histogram (not shown here) is approximately bell shaped.

265


H0: σ2 = 18

12.78

H1: σ2 > 18 Rejection region:  2   2 ,n 1   .210,244  272.704 (from Excel)

2 

(n  1)s 2 

2

=

(245  1)( 22 .56 ) = 305.81; p-value = .0044. There is enough evidence to infer that the population 18

variance is greater than 18.

12.79 LCL =

UCL =

(n  1)s 2

=

 2 / 2,n 1

(n  1)s 2 12 / 2, n 1

=

(n  1)s 2  .205,89

(n  1)s 2  .295,89

=

=

(90  1)( 4.72 )  3.72 113

(90  1)( 4.72 )  6.08 69 .1

H0: σ2 = 200

12.80

H1: σ2 < 200 Rejection region:  2  12,n 1   .295,99  77.9

2 

(n  1)s 2 

2

=

(100  1)(174 .47 ) = 86.36; p-value = .1863. There is not enough evidence to infer that the 200

population variance is less than 200. Replace the bulbs as they burn out.

12.81 LCL =

UCL =

(n  1)s 2  2 / 2,n 1

(n  1)s 2 12 / 2, n 1

=

=

(n  1)s 2  .2025, 24

(n  1)s 2  .2975, 24

=

=

(25  1)(19 .68)  11 .99 39 .4

(25  1)(19 .68 )  38 .09 12 .4

12.82 a p̂  z  / 2 p̂(1  p̂) / n = .48 ±1.96 .48 (1  .48) / 200 = .48  .0692 b p̂  z  / 2 p̂(1  p̂) / n = .48 ±1.96 .48 (1  .48 ) / 500 = .48  .0438 c p̂  z  / 2 p̂(1  p̂) / n = .48 ±1.96 .48 (1  .48 ) / 1000 = .48  .0310 d The interval narrows. 12.83 a p̂  z  / 2 p̂(1  p̂) / n = .50 ±1.96 .50 (1  .50 ) / 400 = .50  .0490 b p̂  z  / 2 p̂(1  p̂) / n = .33 ±1.96 .33(1  .33) / 400 = .33  .0461

266


c p̂  z  / 2 p̂(1  p̂) / n = .10 ±1.96 .10 (1  .10 ) / 400 = .10  .0294 d The interval narrows.

12.84

H0: p = .60 H1: p > .60

a z

b z

c z

p̂  p p(1  p) / n p̂  p p(1  p) / n p̂  p p(1  p) / n

=

=

=

.63  .60 .60 (1  .60 ) / 100 .63  .60 .60 (1  .60 ) / 200 .63  .60 .60 (1  .60 ) / 400

= .61, p-value = P(Z > .61) = 1 – .7291 =.2709

= .87, p-value = P(Z > .87) = 1 – .8078 = .1922 = 1.22, p-value = P(Z > 1.22) = 1 – .8888 = .1112

d The p-value decreases.

12.85 a z 

b z

c z

p̂  p p(1  p) / n

p̂  p p(1  p) / n p̂  p p(1  p) / n

=

=

.73  .70

=

.70 (1  .70 ) / 100

.72  .70 .70 (1  .70 ) / 100 .71  .70 .70 (1  .70 ) / 100

= .65, p-value = P(Z > .65) = 1 – .7422 =.2578

= .44, p-value = P(Z > .44) = 1 – .6700 =.3300

= .22, p-value = P(Z > .22) = 1 – .5871 =.4129

d. The z statistic decreases and the p-value increases.

2

2

z  1.645 .5(1  .5)  p̂(1  p̂)   = 752 12.86 n =   / 2 =      .03 B    

12.87a. .5  .03 b. Yes, because the sample size was chosen to produce this interval.

12.88a. p̂  z  / 2 p̂(1  p̂) / n = .75 ±1.645 .75 (1  .75) / 752 = .75  .0260 b. The interval is narrower. c. Yes, because the interval estimate is better than specified. 2

2

z  1.645 .75 (1  .75 )  p̂(1  p̂)   = 564 12.89 n =   / 2 =      B .03    

267


12.90a. .75  .03 b. Yes, because the sample size was chosen to produce this interval.

12.91a. p̂  z  / 2 p̂(1  p̂) / n = .92 ±1.645 .92 (1  .92 ) / 564 = .92  .0188 b. The interval is narrower. c. Yes, because the interval estimate is better than specified.

12.92a. p̂  z  / 2 p̂(1  p̂) / n = .5 ±1.645 .5(1  .5) / 564 = .5  .0346 b. The interval is wider. c. No because the interval estimate is wider (worse) than specified.

12.93 p̂ = 259/373 = .69

p̂  z  / 2 p̂(1  p̂) / n = .69  1.96 .69 (1  .69 ) / 373 = .69 ± .0469; LCL = .6431, UCL = .7369

12.94

H0: p = .25 H1: p < .25

p̂ = 41/200 = .205

z

p̂  p p(1  p) / n

=

.205  .25 .25(1  .25) / 200

 1.47 , p-value = P(Z < –1.47) = .0708. There is enough evidence to

support the officer’s belief.

12.95 p̂ = 204/314 = .65

p̂  z  / 2 p̂(1  p̂) / n = .65  1.645 .65 (1  .65 ) / 314 = .65  .0443; LCL = .6057, UCL = .6943

12.96

H0: p = .92 H1: p > .92

p̂ = 153/165 = .927

z

p̂  p p(1  p) / n

=

.927  .92 .92 (1  .92 ) / 165

 .33, p-value = P(Z > .33) = 1 – .6293 =.3707. There is not enough evidence

to conclude that the airline’s on-time performance has improved.

268


12.97 p̂ = 97/344 = .28

p̂  z  / 2 p̂(1  p̂) / n = .28 ±1.96 .28 (1  .28 ) / 344 = .28 ± .0474; LCL = .2326, UCL = .3274

12.98 p̂ = 68/400 = .17

p̂  z  / 2 p̂(1  p̂) / n = .17 ±1.96 .17 (1  .17 ) / 400 = .17 ± .0368; LCL = .1332, UCL = .2068

12.99 LCL = .1332(1,000,000)(3.00) = $399,600, UCL = .2068(1,000,000)(3.00) = $620,400

p 12.100 ~

x2 1 2   .0147 n  4 200  4

~ .0147 (1  .0147 ) p (1  ~ p) ~ = .0147  1.96 = .0147 ± .0165; LCL = 0 (increased from –.0018), UCL p  z / 2 200  4 n4 = .0312

p 12.101 ~

x2 3 2   .0132 n  4 374  4

~ .0132 (1  .0132 ) p (1  ~ p) ~ = .0132  1.645 = .0132 ± .0097; LCL = .0035, UCL = .0229 p  z / 2 374  4 n4

p 12.102 ~

x2 1 2   .0077 n  4 385  4

~ .0077 (1  .0077 ) p (1  ~ p) ~ = .0077  2.575 = .0077 ± .0114; LCL = 0 (increased from –.0037), UCL p  z / 2 385  4 n4 = .0191 12.103 p̂  z  / 2 p̂(1  p̂) / n = .2680  1.96 .2680 (1  .2680 ) / 974 = .2680 ± .0278; LCL = .2402, UCL = .2958 Number: LCL = 234,564,000(.2402) = 56,342,273, UCL = 234,564,000(.2958) = 69,384,031

12.104 p̂  z  / 2 p̂(1  p̂) / n = .1192 ±1.96 .1192 (1  .1192 ) / 562 = .1192 ± .0268; LCL = .0924, UCL = .1460

12.105 p̂  z  / 2 p̂(1  p̂) / n = .1056 ±1.96 .1056 (1  .1056 ) / 521 = .1056 ± .0264; LCL = .0792, UCL = .1320

12.106 LCL = 75,000(.0792) =5,940, UCL = 75,000(.1320) = 9,900

269


12.107 p̂  z  / 2 p̂(1  p̂) / n = .1202 ±1.96 .1202 (1  .1202 ) / 391 = .1202 ± .0322; LCL = .0880, UCL = .1524

12.108 H0: p = .90 H1: p < .90

z

p̂  p p(1  p) / n

=

.8644  .90 .90 (1  .90 ) / 177

= –1.58, p-value = P(Z < –1.58) = .0571. There is not enough evidence to

infer that the satisfaction rate is less than 90%.

12.109 p̂  z  / 2 p̂(1  p̂) / n = .2333 ±1.96 .2333 (1  .2333 ) / 120 = .2333 ± .0757; LCL = .1576, UCL = .3090

12.110 p̂  z  / 2 p̂(1  p̂) / n = .600 ±1.96 .600 (1  .600 ) / 1508 = .600 ± .025; LCL = .575, UCL = .625 Total number of Canadians who prefer artificial Christmas trees: LCL = 6 million (.575) = 3.452 million, UCL = 6 million (.625) = 3.749 million

12.111a. p̂  z  / 2 p̂(1  p̂) / n = .7840 ±1.96 .7840 (1  .7840 ) / 426 = .7840 ± .0391; LCL = .7449, UCL = .8231

12.112 H0: p = .50 H1: p > .50

z

p̂  p p(1  p) / n

=

.57  .50 .50 (1  .50 ) / 100

= 1.40, p-value = P(Z > 1.40) = 1 – .9192 =.0808. There is not enough

evidence to conclude that more than 50% of all business students would rate the book as excellent.

12.113 Codes 1, 2, and 3 have been recoded to 5. H0: p = .90 H1: p > .90

z

p̂  p p(1  p) / n

=

.96  .90 .90 (1  .90 ) / 100

= 2.00, p-value = P(Z > 2.00) = 1 – .9772 =.0228. There is enough evidence

to conclude that more than 90% of all business students would rate the book as at least adequate.

12.114 p̂  z  / 2 p̂(1  p̂) / n = .0360 ±1.96 .0360 (1  .0360 ) / 5000 = .0360 ± .0052; LCL = .0308, UCL = .0412

270


Proportion: LCL = .0308, UCL = .0412. Number: LCL = 126.54million (.0308) = 3.89 million , UCL = 126.54 million (.0412) = 5.21 million 12.115 H0: p = .2155 H1: p ≠ .2155

z

p̂  p p(1  p) / n

=

.2442  .2155 .2155 (1  .2155 ) / 1040

= 2.25, p-value = 2P(Z > 2.25) = 2(1 – .9878) =.0244. There is enough

evidence to conclude that the proportion of 4-4-3-2 hands differs from the theoretical probability. This deviation is likely caused by insufficient shuffling.

12.116a. p̂  z  / 2 p̂(1  p̂) / n = .2031 ±1.96 .2031 (1  .2031 ) / 650 = .2031 ± .0309; LCL = .1722, UCL = .2340 Number: LCL = 5 million (.1722) = .861 million, UCL = 5 million (.2340) = 1.17 million

12.117 H0: p = .795 H1: p < .795

z = -6.28, p-value = 0. There is enough evidence to infer that the proportion of White Americans has decreased.

12.118 H0: p = .269 H1: p ≠ .269

z = -.30, p-value = .7659. There is not enough evidence to conclude that the proportion has changed.

12.119 H0: p = .658 H1: p ≠ .658

271


p-value = .0137. There is enough evidence to infer that the proportion of home ownership has changed.

12.120 H0: p = .129 H1: p ≠ .129

p-value = .8777. There is not enough evidence to conclude that the proportion of Americans who did not finish high school has changed.

12.121 H0: p = .104 H1: p > .104

p-value = 0. There is enough evidence to conclude that the proportion has increased.

12.122 H0: p = .102 H1: p > .102

272


p-value = .6904. There is not enough evidence to conclude that the proportion is larger than 10.2%.

12.123 H0: p = .14 H1: p > .14

p-value = .3088. There is not enough evidence to conclude that the proportion has increased.

12.124 H0: p = .129 H1: p < .129

p-value = .1329. There is not enough evidence

12.125 H0: p = .658 H1: p > .658

273


p-value = .1318. There is not enough evidence

12.126 H0: p = .168 H1: p > .168

p-value = .2508. There is not enough evidence

12.127 H0: p = .124 H1: p < .124

p-value = .1583. There is not enough evidence

12.128 H0: p = .158 H1: p < .158

274


p-value = .0223. There is enough evidence

12.129 H0: p = .105 H1: p > .105

p-value = 0. There is enough evidence

12.130 Codes 3 and 4 were changed to 5

p̂  z  / 2 p̂(1  p̂) / n = .7305  1.96 .7305 (1  .7305 ) / 475 = .7305  .0399; LCL = .6906, UCL = .7704; Market segment size: LCL = 41,580,000 (.6906) = 28,715,148, UCL = 41,580,000 (.7704) = 32,033,232

12.131 Code 2 was changed to 3.

p̂  z  / 2 p̂(1  p̂) / n = .5313 ±1.96 .5313 (1  .5313 ) / 320 = .5313 ± .0547; LCL = .4766, UCL = .5860; Market segment size: LCL = 16,015,493 (.4766) = 7,632,984 , UCL = 16,015,493 (.5860) = 9,385,079

12.132a. p̂  z  / 2 p̂(1  p̂) / n = .2919 ±1.96 .2919 (1  .2919 ) / 1836 = .2919 ± .0208; LCL = .2711, UCL = .3127 b LCL = 124,723,003 (.2711) = 33,812,406, UCL = 124,723,003 (.3127) = 39,000,883

12.133 p̂  z  / 2 p̂(1  p̂) / n = .1077 ±1.96 .1077 (1  .1077 ) / 325 = .1077 ± .0337; LCL = .0740, UCL = .1414; Market segment size: LCL = 44,679,192(.0740) = 3,306,260, UCL = 44,679,192(.1414) = 6,317,638

275


12.134 p̂  z  / 2 p̂(1  p̂) / n = .1748 ±1.645 .1748 (1  .1748 ) / 412 = .1748 ± .0308; LCL = .1440, UCL = .2056; Number: LCL = 211,306,936(.1440) = 30,428.199, UCL = 211,306,936(.2056) = 43,444,706

12.135 p̂  z  / 2 p̂(1  p̂) / n = .1500 ±1.96 .1500 (1  .1500 ) / 340 = .1500 ± .0380; LCL = .1120, UCL = .1880; Number: LCL = 211,306,936(.1120) = 23,666,377, UCL = 211,306,936(.1880) = 39,725,704

12.136

Proportion: LCL = .0861, UCL = .1605. Amount: LCL =$8,354, UCL = $15,573

12.137

LCL = 701.26, UCL = 760.02

12.138a. H0: µ = 30 H1: µ > 30

276


t = 3.04, p-value = .0015; there is enough evidence to infer that the candidate is correct. b

LCL = $30.68, UCL = $33.23 c The costs are required to be normally distributed.

12.139a.

LCL = .0892, UCL = .1461 b.

277


LCL = .1105, UCL = .1483

12.40

H0: σ2 = 17 H1: σ2 > 17

 2 = 30.71, p-value = .0435. There is enough evidence to infer that problems are likely.

12.141

LCL = 1193.39, UCL = 1258.43 Total: 354,520(1193.39) = 423,080,623, UCL = 354,520(1258.43) = 446,138,604

12.142 278


LCL = .3718, UCL = .4428

12.143

LCL = 10.51, UCL = 11.33

12.144

LCL = 9.92, UCL = 10.66

12.145

LCL = .3044, UCL = .4119 279


12.146a

LCL = 69.03, UCL = 74.73

b

H0: µ = 68 H1: µ > 68

t = 2.74, p-value = .0043; there is enough evidence to infer that students with a calculus background would perform better in statistics than students with no calculus.

12.147

LCL = .0945, UCL = .1492

280


12.148 H0: p = .75 H1: p > .75

z = .64, p-value = .2611; there is not enough evidence to infer that more than 75% of workers drove alone to work.

12.149

LCL = .0765, UCL = .1271

12.150

LCL = 26.44, UCL = 27.55

281


12.151a.

LCL = 6.84, UCL = 6.98 b. The histogram is bell shaped. c.

H 0: µ = 7 H 1: µ < 7

t = –3.48, p-value = .0004; there is enough evidence to infer that postal workers are spending less than seven hours doing their jobs.

12.152

LCL = .5818, UCL = .6822

282


12.153

LCL = 5.11, UCL = 6.47

12.154

Proportion: LCL = .2977, UCL = .3680. Number: LCL = 69,833,339, UCL = 86,316,357 12.155 H0: σ2 = 4 H1: σ2 > 4

 2 = 161.25, p-value = .0001; there is enough evidence to conclude that the number of springs requiring reworking is unacceptably large.

283


12.156 H0: p = .90 H1: p < .90

z = –1.33, p-value = .0912; there is enough evidence to infer that less than 90% of the springs are the correct length. 12.157

LCL = .937, UCL = 1.263

12.158a. H0: µ = 9.8 H1: µ < 9.8

284


t = –2.97, p-value = .0018; there is enough evidence to infer that enclosure of preaddressed envelopes improves the average speed of payments. b.

H0: σ2 = (3.2)2 = 10.24 H1: σ2 < 10.24

 2 = 101.58, p-value = .0011; there is enough evidence to infer that the variability in payment speeds decreases when a preaddressed envelope is sent.

2

2

z  2.575 .5(1  .5)  p̂(1  p̂)   = 4144 12.159 n =   / 2 =      . 02 B    

285


12.160

Proportion: LCL = .1245, UCL = .1822. Total: LCL = 400,000(.1245) = 49,800 UCL = 400,000(.1822) = 72,880

12.161 Number of cars: H0: µ = 125 H1: µ > 125

t = .459, p-value = .3351; there is not enough evidence to infer that the employee is stealing by lying about the number of cars.

Amount of time H0: µ = 3.5 H1: µ > 3.5

286


t = 7.00, p-value = 0; there is enough evidence to infer that the employee is stealing by lying about the amount of time.

12.162a.

LCL = -5.54 UCL = 29.61 b.

H0: µ = 16 H1: µ < 16

287


t = -.473, p-value = .3210; there is not enough evidence to infer that Mr. Cramer does less well than the S&P 500 index.

12.163

LCL = .0508, UCL = .0740

12.164 H0: µ = 14.35 H1: µ > 14.35

t = .908, p-value = .1823; There is not enough evidence to that the debt ratio increased since 2006.

12.165 H0: µ = 17.62 H1: µ > 17.62

288


t = 4.23, p-value = 0; there is enough evidence to infer that financial obligations increased since 2006.

12.166 H0: µ = 25.97 H1: µ > 25.97

t = .959, p-value = .1693; there is not enough evidence to infer that financial obligations for renters increased since 2006.

12.167

Proportion: LCL = .7267, UCL = .7768. Number: LCL = 57,289,123, UCL = 61,236,017

289


12.168 H0: µ = 65.8 H1: µ > 65.8

t = 2.44, p-value = .0083; there is enough evidence to infer that the percentage of total compensation increased between 2007 and 2008.

12.169

Proportion: LCL = .1439, UCL = .1868. Number: LCL = 33,763,728, UCL = 43,815,015

12.170a.

LCL = .6135, UCL = .7107 b.

290


LCL = 26,823, UCL = 37,144 c.

LCL = 4.67, UCL = 5.79

d.

LCL = 1.41, UCL = 1.84

291


12.171

Total: LCL = 244,137,873(.1260) = 30,761,372, UCL = 244,137,873(.1530) = 37,353,095

12.172

Total: LCL = 244,137,873(.1786) = 43,603,024, UCL = 244,137,873(.2102) = 51,317,781

12.173

Total: LCL = 244,137,873(.1080) = 26,366,891, UCL = 244,137,873(.1338) = 32,665,647

12.174a. H0: µ = 3.0 H1: µ < 3.0

p-value = .3762. There is not enough evidence to support the claim. b. 292


12.175 a H0:p = .5 H1: p >.5

p-value = 0. There is enough evidence to infer that there are more Democrats than Republicans.

12.176 H0: p = .5 H1: p >.5

p-value = .0002. There is enough evidence to infer that there are more conservatives than liberals.

12.177 There are more Democrats than Republicans but there are more conservatives than liberals.

12.178 a H0: µ = 2.08 H1: µ < 2.08

p-value = 0. There is enough evidence to infer that the mean number of children is less than 2.08.

12.179 H0: p = .5 H1: p ≠.5

293


p-value = 0. There is enough evidence to infer that there is no gender equality.

12.180

Total: LCL = 242,823,652(.7475) = 181,510,680, UCL = 242,823,652(.7691) = 186,755,671

12.181

Total: LCL = 123,460,000(.7147) = 88,236,862, UCL = 123,460,000(.7373) = 91,027,058

12.182

Total: LCL = 123,460,000(.1256) = 15,506,576, UCL = 123,460,000(.1400) = 17,284,400

12.183

294


Total: LCL = 123,460,000(.2919) = 36,037,974, UCL = 123,460,000(.3226) = 39,828,196

12.184

12.185a. H0: µ = 60 H1: µ > 60

p-value = .1049. There is not enough evidence to infer that the mean age is greater than 60. b. AGE is required to be normally distributed. The histogram is approximately bell shaped.

12.186

12.187

295


12.188 a.

b. DEBT is required to be normally distributed. The histogram is extremely positively skewed.

12.189a.

b. The required condition is normality. The histogram is extremely positively skewed.

12.190 H0: µ = 5250 H1: µ > 5250

p-value = 0. There is enough evidence to infer that wealthy households spend more than $5250.

296


Case 12.1 95% confidence interval estimate of mean weekly consumption per student:

1 2 3 4 5 6 7

A B t-Estimate: Mean

Mean Standard Deviation LCL UCL

C

D

Cans 1.316 1.115 1.218 1.414

Estimated Mean Number of Cans per Student

Profit

Current Profit

Net

LCL = 1.218

$652,600

$616,000

$ 36,600

UCL = 1.414

$789,800

$616,050

$173,800

Pepsi should sign the exclusivity agreement. They would increase profit by between $36,600 and $173,800.

297


Case 12.2 Estimated total number of soft drink sales per year LCL = 1.218 × 40 × 50,000 = 2,436,000. Pepsi sells 22,000 × 40 = 880,000 Total sales by Coke = 2,436,000- 880,000 = 1,556,000 cans per year Coke’s current estimated profit = 1,556,000 × .70 = $1,089,200 Profit with exclusivity agreement: $652,600

UCL = 1.414 × 40 × 50,000 = 2,828,000. Pepsi sells 22,000 × 40 = 880,000 Total sales by Coke = 2,828,000- 880,000 = 1,948,000 cans per year Coke’s current estimated profit = 1,948,000 × .70 = $1,363,600 Profit with exclusivity agreement: $789,800

Coke would not sign the exclusivity agreement. Coke is expected to lose from the exclusivity agreement because they currently have a much larger share of the market and would not gain by paying for exclusivity.

Case 12.3 a 95% confidence interval estimate of the mean medical costs for each of the four age categories:

1 2 3 4 5 6 7

1 2 3 4 5 6 7

A B t-Estimate: Mean

C

Age:45-64 1830 749 1784 1877

Mean Standard Deviation LCL UCL

A B t-Estimate: Mean

Mean Standard Deviation LCL UCL

D

C

D

Age:65-74 4494 1820 4381 4607

298


1 2 3 4 5 6 7

1 2 3 4 5 6 7

A B t-Estimate: Mean

C

Age:75-84 8074 3186 7876 8272

Mean Standard Deviation LCL UCL

A B t-Estimate: Mean

D

C

D

Age:85+ 15957 6207 15572 16342

Mean Standard Deviation LCL UCL

Estimated annual costs for 2016 Estimate of mean

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

45 to 64

10,094

1,784

1,877

18,007,696

18,946,438

65 to 74

3,375

4,381

4,607

14,785,875

15,548,625

75 to 84

1,745

7,876

8,272

13,743,620

14,434,640

15,572 16,342

12,333,024

12,942,864

58,870,215

61,872,567

85 and over Total

792 16,006

Estimated annual costs for 2021 Estimate of mean

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

45 to 64

10,054

1,784

1,877

17,936,336

18,871,358

65 to 74

4,023

4,381

4,607

17,624,763

18,533,961

75 to 84

2,087

7,876

8,272

16,437,212

17,263,664

15,572 16,342

13,578,784

14,250,224

65,577,095

68,919,207

85 and over Total

872 17,036

Estimated annual costs for 2026 Estimate of mean

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

45 to 64

9,892

1,784

1,877

17,647,238

18,567,284

65 to 74

4,538

4,381

4,607

19,880,978

20,906,566

299


75 to 84 85 and over Total

2,666 973

7,876

8,272

20,997,416

22,053,152

15,572 16,342

15,151,556

15,900,766

73,677,2785

77,427,768

18,069

Estimated annual costs for 2031 Estimate of mean

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

45 to 64

9,823

1,784

1,877

17,524,232

18,437,771

65 to 74

4,874

4,381

4,607

21,352,994

22,454,518

75 to 84

3,209

7,876

8,272

25,274,084

26,544,848

85 and over

1,193

15,572 16,342

18,577,396

19,496,006

Total

19,099

82, 728,706

86, 933,143

Estimated annual costs for 2036 Estimate of mean

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

45 to 64

10,150

1,784

1,877

18,107,600

19,051,550

65 to 74

4,675

4,381

4,607

20,481,175

21,537,725

75 to 84

3,669

7,876

8,272

28,897,044

30,349,968

85 and over

1,543

15,572 16,342

24,027,596

25,215,706

Total

20,037

91,513,415

96,154,949

Case 12.4 a 95% confidence interval estimate of the proportion of Americans with Alzheimer’s disease for each of the three age categories:

300


Estimated number of Americans with Alzheimer’s disease for 2015 Estimate of proportion

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

65 to 74

26,967

.0009

.0025

24.270

67.418

75 to 84

13,578

.0077

.0115

104.551

156.147

6,292

.0074

.0112

46.561

70.470

175.382

294.035

85 and over Total

46,837

Estimated number of Americans with Alzheimer’s disease for 2020 Estimate of proportion

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

65 to 74

32,312

.0009

.0025

29.081

80.780

75 to 84

15,895

.0077

.0115

122.392

182.793

6,597

.0074

.0112

48.818

73.886

200.290

337.459

85 and over Total

54,804

Estimated number of Americans with Alzheimer’s disease for 2025 Estimate of proportion

Estimate of total (1,000s)

Age category Number (1,000s)

LCL

UCL

LCL

UCL

65 to 74

36,356

.0009

.0025

32.720

90.890

75 to 84

20,312

.0077

.0115

156.402

233.588

301


85 and over Total

7,239

.0074

.0112

63,907

53.569

81.077

242.691

405.555

Case 12.5 a. If there is no middle bias the proportion of bets on 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, and 35 is 12/36, or .3333. b.

H0: p = .3333 H1: p > .3333

The number of bets on 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, and 35 are 450, 636, 633, 783, 649, 1079, 983, 746, 703, 878, 925, and 627, respectively. The total number of bets on the middle numbers is 9092. Thus the sample proportion is p̂ = 9092/21,731 = .4184. The test statistic is

z

p̂  p p(1  p) / n

=

.4184  ..3333 .3333 (1  .3333 ) / 21,731

 26 .61, p-value = P(Z > 26.61) = 0. There is overwhelming

evidence of middle bias. c. If there is no middle bias the proportion of bets on 17 and 20 is 2/36 = .0556. d.

H0: p = .0556 H1: p > .0556

The number of bets on 17 and 20 are 1079 and 983, respectively. The total number of bets on the middle numbers is 2062. Thus the sample proportion is p̂ = 2062/21,731 = .0949. The test statistic is

z

p̂  p p(1  p) / n

=

.0949  ..0556 .0556 (1  .0556 ) / 21,731

 25 .28, p-value = P(Z > 25.28) = 0. There is overwhelming

evidence of the middle of the middle bias.

302


Chapter 13 13.1a Equal-variances estimator

1 1  ( x1  x 2 )  t  / 2 s 2p    = (524 – 469)  2.009  n1 n 2 

 (25  1)129 2  (25  1)131 2  1   1    25 25   25  25  2  

= 55  73.87 b Equal-variances estimator

1 1  ( x1  x 2 )  t  / 2 s 2p    = (524 – 469)  2.009  n1 n 2 

 (25  1)255 2  (25  1)260 2  1    1    25 25  25  25  2  

= 55  146.33 c The interval widens. d Equal-variances estimator

1 1  ( x1  x 2 )  t  / 2 s 2p    = (524 – 469)  1.972 n n 2  1

 (100  1)129 2  (100  1)131 2  1 1        100  100  2   100 100 

= 55  36.26 e The interval narrows.

13.2

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

a Equal-variances test statistic Rejection region: t   t  / 2,   t .025, 22  –2.074 or t  t  / 2,  t .025, 22  2.074 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(74  71)  0  (12  1)18 2  (12  1)16 2  1 1       12 12   12  12  2  

= .43, p-value = .6703. There is not enough

evidence to infer that the population means differ.

b Equal-variances test statistic Rejection region: t   t  / 2,   t .025, 22  –2.074 or t  t  / 2,  t .025, 22  2.074 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(74  71)  0  (12  1)210 2  (12  1)198 2  1 1       12 12   12  12  2  

evidence to infer that the population means differ. c The value of the test statistic decreases and the p-value increases. 281

= .04, p-value = .9716. There is not enough


d Equal-variances test statistic Rejection region: t   t  / 2,   t .025, 298  –1.960 or t  t  / 2,  t .025, 298  1.960 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

(74  71)  0

=

 (150  1)18 2  (150  1)16 2  1 1        150  150  2  150 150  

= 1.53, p-value = .1282. There is not

enough evidence to infer that the population means differ. e The value of the test statistic increases and the p-value decreases. f Rejection region: t   t  / 2,   t .025, 22  –2.074 or t  t  / 2,  t .025, 22  2.074 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

(76  71)  0

=

 (12  1)18 2  (12  1)16 2  1 1       12 12   12  12  2  

= .72, p-value = .4796. There is not enough

evidence to infer that the population means differ. g The value of the test statistic increases and the p-value decreases.

13.3 a Unequal-variances estimator



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 64.8 (rounded to 65, approximated by   70 )

 s2 s2  (x1  x 2 )  t  / 2  1  2  = (63 – 60)  1.667 n n  2  1

 18 2 7 2   = 3  4.59    50 45   

b Unequal-variances estimator



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 63.1 (rounded to 63, approximated by   60 )

 s2 s2  (x1  x 2 )  t  / 2  1  2  = (63 – 60)  1.671 n n  2  1

 41 2 15 2    = 3  10.38   50 45  

c The interval widens. d Unequal-variances estimator



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 131 (approximated by   140 )

 s2 s2  (x1  x 2 )  t  / 2  1  2  = (63 – 60)  1.656 n n  2  1

 18 2 7 2   = 3  3.22    100 90   

282


e The interval narrows.

13.4

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

a Unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 200.4 (rounded to 200)

Rejection region: t  t ,  t.05, 200  1.653

t

( x 1  x 2 )  ( 1   2 )  s 12 

 n1 

s 22 

=

n 2 

(412  405 )  0  128 2 54 2      150 150   

= .62, p-value = .2689. There is not enough evidence to infer that 1

is greater than  2 . b Unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 223.1 (rounded to 223)

Rejection region: t  t ,  t .05, 223  1.645 t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(412  405 )  0  31 2 16 2      150 150   

= 2.46, p-value = .0074. There is enough evidence to infer that 1 is

greater than  2 . c The value of the test statistic increases and the p-value decreases.

d Unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 25.6 (rounded to 26)

Rejection region: t  t ,  t.05,26  1.706

t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(412  405 )  0  128 2 54 2      20 20  

= .23, p-value = .4118. There is not enough evidence to infer that 1

is greater than  2 . e The value of the test statistic decreases and the p-value increases.

283


f Unequal-variances test statistic Rejection region: t  t ,  t.05, 200  1.653

t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(409  405 )  0  128 2 54 2      150 150   

= .35, p-value = .3624. There is not enough evidence to infer that 1

is greater than  2 . g The value of the test statistic decreases and the p-value increases.

13.5 a Equal-variances t-test degrees of freedom = 28, Unequal-variances t-test degrees of freedom =26.4 b Equal-variances t-test degrees of freedom = 24, Unequal-variances t-test degrees of freedom = 10.7 c Equal-variances t-test degrees of freedom = 98 , Unequal-variances t-test degrees of freedom = 91.2 d Equal-variances t-test degrees of freedom = 103, Unequal-variances t-test degrees of freedom = 78.5

13.6 a In all cases the equal-variances t-test degrees of freedom is greater than the unequal-variances t-test degrees of freedom.

13.7

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = 1.85, p-value = .4352; use equal-variances test statistic Rejection region: t < -tα/2,υ = -t.05,14 = -1.761 or t > tα/2,υ = t.05,14 = 1.761 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(6.25  5.88)  0  (8  1)18 .21  (8  1)9.84  1 1      882  8 8  

 .20, p-value = .8442. There is not enough

evidence of a difference.

13.8

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = 2.87, p-value = .2608; use equal-variances test statistic Rejection region: t < -tα/2,υ = -t.05,13 = -1.771 or t > tα/2,υ = t.05,13 = 1.771 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(8.44  6.50 )  0  (9  1)15 .78  (6  1)5.50  1 1      962   9 6 

evidence of a difference.

13.9

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 284

 1.07 , p-value = .3028. There is not enough


Two-tail F test: F = .08, p-value = .0694; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n 2 1 n1 1

= 3.49 (rounded to 3)

Rejection region: : t < -tα/2,υ = -t.05,3 = -2.353 or t > tα/2,υ = t.05,3 = 2.353 t

( x 1  x 2 )  ( 1   2 )  s 12 

 n1 

s 22 

=

(107 .75  116 .75)  0  7.58 92 .25     4   4

n 2 

= -1.80, p-value = .1694. There is not enough evidence to infer

that the batteries differ.

13.10

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 2.55, p-value = .3272; use equal-variances test statistic Rejection region: t > tα,υ = t.10,10 = 1.372 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(36 .67  34 .50 )  0  (6  1)5.87  (6  1)2.30  1 1      662   6 6 

 1.86, p-value = .0465. There is enough evidence

to infer that new putter is better.

13.11

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = 1.02, p-value = .9856; use equal-variances test statistic Rejection region: t   t ,   t .10,10  1.372 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(361 .50  381 .83)  0  (6  1)6767 .5  (6  1)6653 .4  1 1      662   6 6 

 .43, p-value = .3382. The manager should

choose to use cameras.

13.12

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = 1.02, p-value = .9823; use equal-variances test statistic Rejection region: t   t ,   t .10,18  1.330

285


t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(5.10  7.30 )  0  (10  1)5.88  (10  1)5.79  1 1      10  10  2   10 10 

 2.04, p-value = .0283. There is enough

evidence to infer that there are fewer errors when the yellow ball is used.

13.13

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = 1.13, p-value = .8430; use equal-variances test statistic Rejection region: t   t ,   t .10,19  1.328 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(12 .62  14 .20 )  0 1  (10  1)2.95  (11  1)2.61  1     10  11  2  10 11  

 2.17 , p-value = .0213. There is enough

evidence to infer that users of the new device leave larger tips.

13.14

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2)  0

Two-tail F test: F = 1.04, p-value = .9873; use equal-variances test statistic Rejection region: t   t  / 2,   t .05,13  1.771 or t  t  / 2,  t .05,13  1.771 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(3,372  4,093 )  0  (9  1)755 ,196  (6  1)725 ,778  1 1      962   9 6 

 1.59, p-value = .1368. There is not

enough evidence to infer a difference between the two types of vacation expenses.

13.15

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .69, p-value = .5486; use equal-variances test statistic Rejection region: t   t ,   t .10,22  1.321 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(172 .0  176 .8)  0 1  (12  1)157 .1  (12  1)227 .7  1     12  12  2  12 12  

 .84 , p-value = .2053. There is not

enough evidence to conclude that games take longer to complete than 5 years ago.

13.16

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2)  0

Two-tail F test: F = 1.67, p-value = .4060; use equal-variances test statistic

286


Rejection region: t   t  / 2,   t .05,22  1.717 or t  t  / 2,  t .05,22  1.717 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(33 .33  31 .50 )  0 1  (12  1)20 .24  (12  1)12 .09  1     12 12 2 12 12     

 1.12 , p-value = .2761. There is not

enough evidence to infer a difference in speeds.

13.17a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.59, p-value = .3050; use equal-variances test statistic Rejection region: t  t ,  t.05,38  1.684 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(36 .93  31 .36 )  0  (15  1)4.23 2  (25  1)3.35 2  1   1    15 25   15  25  2  

= 4.61, p-value = 0. There is enough

evidence to infer that Tastee is superior.

1 1   = (36.93 – 31.36)  2.021 b ( x1  x 2 )  t  / 2 s 2p    n1 n 2 

 (15  1)4.23 2  (25  1)3.35 2  1   1    15 25   15  25  2  

= 5.57  2.43; LCL = 3.14, UCL = 8.00 c The histograms are somewhat bell shaped. The weight gains may be normally distributed.

13.18

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = .99, p-value = .9571; use equal-variances test statistic Rejection region: t   t  / 2,   t .025, 238  1.960 or t  t  / 2,  t .025, 238  1.960 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(10 .01  9.12 )  0  (120  1)4.43 2  (120  1)4.45 2  1 1        120  120  2  120 120  

 1.55, p-value = .1204. There is not

enough evidence to infer that oat bran is different from other cereals in terms of cholesterol reduction?

13.19 a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = .50, p-value = 0; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 449

Rejection region: t   t  / 2,   t .025, 449  1.960 or t  t  / 2,  t .025, 449  1.960 287


t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(58 .99  52 .96 )  0  30 .77 2 43 .32 2      250 250  

= 1.79, p-value = .0734. There is not enough evidence to

conclude that a difference in mean listening times exist between the two populations.

 s2 s2  b ( x 1  x 2 )  t  / 2  1  2  = (58.99 –52.96)  1.960  n1 n 2   

 30 .77 2 43 .32 2    = 6.03  6.59;   250 250  

LCL = –.56, UCL = 12.62 c The histograms are bell shaped.

13.20 a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.01, p-value = .9619; use equal-variances test statistic Rejection region: t  t ,  t .05, 282  1.645 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(59 .81  57 .40 )  0  (125  1)7.02 2  (159  1)6.99 2  1 1      125 159   125 159 2    

 2.88, p-value = .0021. There is

enough evidence to infer that the cruise ships are attracting younger customers.

1 1   = (59.81 – 57.40) b ( x1  x 2 )  t  / 2 s 2p    n1 n 2 

 (125  1)7.02 2  (159  1)6.99 2  1 1     2.576     125  159  2   125 159 

= 2.41  2.16; LCL = .25, UCL = 4.57

13.21a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = .98, p-value = .9254; use equal-variances test statistic Rejection region: t   t  / 2,   t.025,198  1.972 or t  t  / 2,  t.025,198  1.972

t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(10 .23  9.66 )  0  (100  1)2.87 2  (100  1)2.90 2  1 1      100 100   100  100  2  

= 1.40, p-value = .1640. There is not

enough evidence to infer that the distance males and females drive differs.

1 1   = (10.23 – 9.66)  1.972 b ( x1  x 2 )  t  / 2 s 2p   n n 2  1

 (100  1)2.87 2  (100  1)2.90 2  1 1      100 100   100  100  2  

= .57  .80; LCL = –.23, UCL = 1.37 288


c The histograms are bell shaped.

13.22

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .98, p-value = .9520; use equal-variances test statistic Rejection region: t  t ,  t .05,58  1.671 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(115 .50  110 .20 )  0  (30  1)21 .69 2  (30  1)21 .93 2  1   1    30 30   30  30  2  

 .94 , p-value = .1753. There is not

enough evidence to retain supplier A - switch to supplier B.

13.23

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = .92, p-value = .5000; use equal-variances test statistic Rejection region: t   t  / 2,   t .025,594  1.960 or t  t  / 2,  t .025,594  1.960 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(5.56  5.49 )  0  (306  1)5.36 2  (290  1)5.58 2  1 1      306 290   306 290 2    

 .16 , p-value = .8759. There is no

evidence of a difference in job tenures between men and women.

13.24a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = 5.18, p-value = .0019; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 33.9 (rounded to 34)

Rejection region: t   t  / 2,   t .005,34  2.724 or t  t  / 2,  t .005,34  2.724 t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(70 .42  56 .44 )  0  20 .54 2 9.03 2      24 16  

 2.94 , p-value = .0060. There is enough evidence to conclude

that the two packages differ in the amount of time needed to learn how to use them.

 s2 s2  b ( x1  x 2 )  t  / 2  1  2  = (70.42 –56.44)  2.030 n n  2  1

 20 .54 2 9.03 2    = 13.98  9.67;   24 16  

LCL = 4.31, UCL = 23.65 c The amount of time is required to be normally distributed. 289


d The histograms are somewhat bell shaped.

13.25a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = 5.18, p-value = .0019; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 276.5 (rounded to 277)

Rejection region: t   t ,   t .01, 277  2.326

t

( x 1  x 2 )  ( 1   2 )  s 12 

 n1 

s 22 

=

n 2 

(5.02  7.80 )  0  1.39 2 3.09 2      200 200  

= –11.60, p-value = 0. There is enough evidence to infer that the

amount of time wasted in unsuccessful firms exceeds that of successful firms.

 s2 s2  b ( x1  x 2 )  t  / 2  1  2  = (5.02 – 7.80)  1.960 n n  2  1

 1.39 2 3.09 2    = –2.78  .47;   200  200  

LCL = –3.25, UCL = –2.31. Workers in unsuccessful companies waste on average between 2.31 and 3.25 hours per week more than workers in successful companies.

13.26

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .73, p-value = .0699; use equal-variances test statistic Rejection region: t  t ,  t .05, 268  1.645 t

( x 1  x 2 )  (1   2 )  1 1    s 2p   n1 n 2 

=

(.646  .601)  0  (125  1).045 2  (145  1).053 2  1 1        125  145  2  125 145  

= 7.54, p-value = 0. There is enough

evidence to conclude that the reaction time of drivers using cell phones is slower that for non-cell phone users.

13.27

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = 1.10, p-value = .6406; use equal-variances test statistic Rejection region: t   t  / 2,   t .025,183  1.973 or t  t  / 2,  t .025,183  1.973 t

( x1  x 2 )  (1   2 )  1 1   s 2p   n n 2   1

=

(.654  .662 )  0  (95  1).048 2  (90  1).045 2  1   1    95 90   95  90  2  

= –1.17, p-value = .2444. There is not

enough evidence to infer that the type of discussion affects reaction times. 290


13.28

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .97, p-value = .9054; use equal-variances test statistic Rejection region: t  t ,  t.05,143  1.656 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(6.18  5.94 )  0  (64  1)1.59 2  (81  1)1.61 2  1   1    64 81   64  81  2  

= .90, p-value = .1858. There is not enough

evidence to infer that people spend more time researching for a financial planner than they do for a stock broker.

13.29

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .74, p-value = .0446; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 373

Rejection region: t   t ,   t.05,373  1.645

t

( x 1  x 2 )  ( 1   2 )  s 12 s 22      n1 n 2   

=

(63 .71  66 .80 )  0  5.90 2 6.85 2      173 202  

= –4.69, p-value = 0. There is enough evidence to infer that

students without textbooks outperform those with textbooks.

13.30

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = .85, p-value = .2494; use equal-variances test statistic Rejection region: t   t  / 2,   t .025, 413  1.960 or t  t  / 2,  t .025, 413  1.960 t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(149 .85  154 .43)  0  (213  1)21 .82 2  (202  1)23 .64 2  1 1        213  202  2  213 202  

= –2.05, p-value = .0412. There is

enough evidence to conclude that there are differences in service times between the two chains.

13.31

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = .80, p-value = .1818; use equal-variances test statistic Rejection region: t   t  / 2,   t .025,309  1.960 or t  t  / 2,  t .025,309  1.960 291


t

( x 1  x 2 )  (1   2 )  1 1    s 2p  n n 2   1

=

(488 .4  498 .1)  0  (124  1)19 .62 2  (187  1)21 .93 2  1 1      124  187  124 187 2    

= –3.985, p-value = 0. There is

enough evidence to conclude that there are differences in the amount of sleep between men and women.

13.32 a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = 1.51, p-value = .0402; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 190

Rejection region: t   t  / 2,   t.025,190  1.973 or t  t  / 2,  t.025,190  1.973

t

( x1  x 2 )  (1   2 )  s12 

n  1

s 22 

=

n 2 

(130 .93  126 .14 )  0  31 .99 2 26 .00 2      100 100  

= 1.16, p-value = .2467. There is not enough evidence to infer

that differences exist between the two types of customers.

13.33

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.07, p-value = .8792; use equal-variances test statistic Rejection region: t  t ,  t.05,38  1.684

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(73 .60  69 .20 )  0  (20  1)15 .60 2  (20  1)15 .06 2  1   1    20 20   20  20  2  

= .91, p-value = .1849. There is not

enough evidence to infer that the new design tire lasts longer than the existing design tire.

13.34

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .95, p-value = .8252; use equal-variances test statistic Rejection region: t   t ,   t .05,178  1.653

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

60,245  63,563 )  0  (90  1)10,506 2  (90  1)10,755 2  1 1       90  90  2  90 90  

= −2.09, p-value = .0189. There is

enough evidence to conclude that commission salespeople outperform fixed-salary salespersons

292


13.35

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = .88, p-value = .4709; use equal-variances test statistic Rejection region: t  t  / 2,  t.025,429  1.645 and t   t  / 2,   t .025,429  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

633 .97  661 .86 )  0  (93  1)49 .45 2  (338  1)52 .69 2  1   1    93 338   93  338  2  

= −4.58, p-value = 0. There is enough

evidence to conclude there is a difference in scores between those who have and those who do not have accidents in a three-year period.

13.36

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .41, p-value = 0; use unequal-variances test statistic



(s12 / n 1  s 22 / n 2 ) 2 (s12 / n 1 ) 2 (s 22 / n 2 ) 2  n1 1 n 2 1

= 222

Rejection region: t  t ,  t .05,222  1.645

t

( x1  x 2 )  (1   2 )

(14 .20  11 .27 )  0

= 6.28, p-value = 0. There is enough evidence to conclude that  s12 s 22   2.84 2 4.42 2        n n   130 130  2  1  bottles of wine with metal caps are perceived to be cheaper.

13.37

=

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = 1.14, p-value = .2430; use equal-variances test statistic Rejection region: t   t ,   t .05,641  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(496 .9  511 .3)  0  (355  1)173 .78 2  (288  1)69 .07 2  1 1        355  288  2  355 288  

= −2.54, p-value = .0057. There is

enough evidence to conclude that SAT scores increased after the change in school start time.

13.38

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 3.14, p-value = 0; use unequal-variances test statistic

293




(s12 / n1  s 22 / n 2 ) 2 (s12 / n1 ) 2 (s 22 / n 2 ) 2  n1  1 n2 1

 74

Rejection region: t  t ,  t .05,74  1.665

t

( x1  x 2 )  (1   2 )  s12 

n  1

s 22 

=

n 2 

(93 .81  61 .25)  0  15 .89 2 8.96 2     48  48   

= 12.37, p-value = 0. There is enough evidence to conclude that the

larger the bucket the more people will eat?

13.39a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .48, p-value = .0856; use equal-variances test statistic Rejection region: t  t ,  t .05,46  1.679

t

( x1  x 2 )  (1   2 ) 1 1  s 2p    n n 2  1

=

(97 .71  94 .58 )  0  (24  1)5.10 2  (24  1)7.36 2  1 1       24  24  2   24 24 

= 1.71, p-value = .0471. There is enough

evidence to conclude that diners who believe they are drinking a fine wine (California wine) eat more than diners who believe they are drinking an inferior wine?

b

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.07, p-value = .8796; use equal-variances test statistic Rejection region: t  t ,  t .05,46  1.679

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(64 .00  57 .33)  0  (24  1)9.44 2  (24  1)9.14 2  1 1       24 24  24  24  2  

= 2.49, p-value = .0083. There is enough

evidence to conclude that diners who believe they are drinking a fine wine (California wine) spend more time in the restaurant than diners who believe they are drinking an inferior wine?

13.40

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.57, p-value = .2486; use equal-variances test statistic Rejection region: t  t ,  t .05,54  1.673

294


t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(10 .04  8.64 )  0  (28  1)2.32 2  (28  1)1.85 2  1 1      28 28   28  28  2  

= 2.49, p-value = .0080. There is enough

evidence to conclude that people eat more when they are not aware of how much they have already eaten?

3.41

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 5.99, p-value = .0001; use unequal-variances test statistic



(s12 / n1  s 22 / n 2 ) 2 (s12 / n1 ) 2 (s 22 / n 2 ) 2  n1  1 n2 1

 25

Rejection region: t  t  /   t .05,25  1.708

t

( x1  x 2 )  (1   2 )  s12 

n  1

s 22 

=

n 2 

(136 .80  72 .80 )  0  124 .16 2 9.87 2     20  20   

= 10.97, p-value = 0. There is enough evidence to conclude that

people will eat more M&M’s when given a full pound.

13.42

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .92, p-value = .1565; use equal-variances test statistic Rejection region: t < -tα,υ = -t.05,∞ = -1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(4.69  4.86 )  0 1   (756  1)8.58  (7643  1)9.30  1     756  7643  2   756 7643 

enough to conclude that the adage is true.

13.43

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 2.12, p-value = 0; use unequal-variances test statistic



( s12 / n1  s 22 / n 2 ) 2 ( s12 / n1 ) 2 ( s 22 / n 2 ) 2  n1  1 n2  1

= 926

Rejection region: t  t /  t.05,926  1.645

295

= -1.48, p-value = .0692. There is not


t

( x1  x 2 )  (1   2 )  s12 s 22     n n  2  1

=

34 .97  31 .63)  0  97 .02 45 .83     483   521

= 6.31, p-value = 0. There is enough evidence to conclude that

guesses this year are higher than they were 5 years ago.

13.44

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.27, p-value = .2120 use equal-variances test statistic Rejection region: t  t /  t.05, 234  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(57 ,030  56,055 )  0 1   (129  1)24,910 ,081  (97  1)19,545 ,241  1     129  97  2 129 97   

= 1.53, p-value = .0643.There is

not enough evidence to conclude that electrical engineers receive higher starting salaries than do mechanical engineers.

13.45

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 2.39, p-value = .0220; use unequal-variances test statistic



( s12 / n1  s 22 / n 2 ) 2 ( s12 / n1 ) 2 ( s 22 / n 2 ) 2  n1  1 n2  1

= 50

Rejection region: t  t /  t.05,50  1.676

t

( x1  x 2 )  (1   2 )  s12 

n  1

s 22 

=

n 2 

166 .8  116 .7)  0  522 .6 218 .7     30   30

= 10.08, p-value = 0. There is enough evidence to conclude that

chocolate helps improve cognitive memory.

13.46

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .67, p-value = .0081; use unequal-variances test statistic



( s12 / n1  s 22 / n 2 ) 2 ( s12 / n1 ) 2 ( s 22 / n 2 ) 2  n1  1 n2  1

= 339

Rejection region: t  t /  t.05,339  1.645

296


t

( x1  x 2 )  (1   2 )  s12 s 22     n n  2  1

=

7.31  6.97 )  0  31 .18 46 .53     177   178

= .51, p-value = .3053. There is not enough evidence to conclude

that Americans have more missing teeth.

13.47

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.04, p-value = .9378 use equal-variances test statistic Rejection region: t  t /  t.05,162  1.654

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(20 .15  14 .82 )  0 1   (134  1)14 .67  (30  1)14 .06  1     134  30  2 134 30   

= 6.92, p-value = 0.There is enough

evidence that White American adults get more slow-wave sleep than Black American adults.

13.48

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = 1.02, p-value = .9310 use equal-variances test statistic Rejection region: t  t /  t.05,182  1.653

t

( x1  x 2 )  (1   2 ) 1 1  s 2p    n n 2  1

=

(74 .85  67 .81)  0 1   (86  1)113 .85  (98  1)111 .72  1     86  98  2 86 98   

= 4.49, p-value = 0.There is enough

evidence to conclude that the exercise improves cognitive impairment.

13.49

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .97, p-value = .6983 use equal-variances test statistic Rejection region: t  t /  t .05,  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p    n n 2  1

=

(1452 .56  1246 .61)  0 1   (438  1)130 ,588  (571  1)135 ,277  1     438  571  2 438 571   

= 8.88, p-value = 0.There is enough

evidence to infer that low-income smokers spend more than high-income smokers.

13.50

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .2.70, p-value = 0; use unequal-variances test statistic

297




( s12 / n1  s 22 / n 2 ) 2 ( s12 / n1 ) 2 ( s 22 / n 2 ) 2  n1  1 n2  1

= 498

Rejection region: t  t /  t.05, 498  1.645

t

( x1  x 2 )  (1   2 )  s12 

n  1

s 22 

=

n 2 

1157 .77  1091 .71)  0  157 ,220 58,327     441   325

= 2.66, p-value = .0040. There is enough evidence to conclude

that low-income alcoholic drinkers spend more than middle-income.

13.51

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

Two-tail F test: F = 1.00, p-value = .9644 use equal-variances test statistic Rejection region: t < -tα/2,υ = -t.025,∞ = -1.96 or t > tα/2,υ = t.025,∞ = 1.96

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

448 .28  443 .03)  0 1   (552  1)9799  (577  1)9837  1     552  577  2  552 577  

= .89, p-value = .3741.There is not

enough to infer a difference between the two sexes.

13.52

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .91, p-value = .2915 use equal-variances test statistic Rejection region: t  t /  t.05,  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(6.44  10 .53)  0 1   (524  1)15 .15  (409  1)16 .71  1     524  409  2  524 409  

= -15.57, p-value = 0.There is enough

evidence to infer that white-collar public servants take more sick days than do white-collar private sector workers.

13.53

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .93, p-value = .4742 use equal-variances test statistic Rejection region: t  t /  t .05,  1.645

t

( x1  x 2 )  (1   2 ) 1 1  s 2p     n1 n 2 

=

(10 .53  10 .32 )  0 1   (409  1)16 .71  (397  1)17 .94  1     409  397  2  409 397  

= .68, p-value = .2469.There is not

enough evidence to conclude that white-collar public sector workers take more sick days than they did 5 years ago.

298


13.54a. Two-tail F test: F = 2.24, p-value = 0. Use unequal-variances estimator.

LCL = 15,826, UCL = 23,645 b. Incomes are required to be normally distributed. c. The required condition is not satisfied. However, the sample sizes are large enough to overcome this deficiency.

13.55 Two-tail F test: F = 1.16, p-value = .1318. Use the equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 7.36, p-value = 0. There is enough evidence to conclude that men work longer hours. b The numbers of hours appear to be bell shaped.

13.56a. Two tail F test: F = .415, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 4.99, p-value = 0. There is enough evidence to infer that American born residents are more educated than people born outside the United States. b. The required condition of normality is satisfied. c. A nonparametric test, the Wilcoxon rank sum test can be used.

13.57a Two tail F test: F = 1.03, p-value = .7934. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 299


t = 1.04, p-value = .1490. There is not enough evidence to infer that American born residents have higher incomes than people born outside the United States. b. Incomes are required to be normally distributed. c. The required condition is not satisfied. However, the sample sizes are large enough to overcome this deficiency.

13.58a. Two tail F test: F = .999, p-value = .9879. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.73, p-value = .0420. There is enough evidence to infer that Republicans are older than Democrats. b. The required condition of normality is satisfied. c. Wilcoxon rank sum test.

13.59a. Two tail F test: F = .756, p-value = .0054. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -3.37, p-value = .0004. There is enough evidence to infer that Republicans earn more income than Democrats. b.

300


LCL = -16,041, UCL = -4,223 c. Incomes are required to be normally distributed. d The required condition is not satisfied. However, the sample sizes are large enough to overcome this deficiency.

13.60a. Two tail F test: F = .813, p-value = .0417. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.46, p-value = .0730. There is not enough evidence to infer that Republicans do not work harder than do Democrats. b. The required condition is satisfied. c. Wilcoxon rank sum test

13.61 Two tail F test: F = 1.30, p-value = .0010. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.32, p-value = .0931. There is not enough evidence to infer that Republicans are more educated than Democrats. b.

LCL = -,537, UCL = .104

301


13.62 Two tail F test: F = 1.41, p-value = .0002. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.77, p-value = .0389. There is enough evidence to infer that Republicans are older than Democrats when their first child is born. b. Ages are required to be normally distributed. c. We can use the Wilcoxon rank sum test.

13.63a. Two tail F test: F = .961, p-value = .6748. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -2.08, p-value = .0189. There is enough evidence to infer that conservatives have higher incomes than liberals. b. Incomes are required to be normally distributed. This requirement is not satisfied. c. Wilcoxon rank sum test.

13.64 Two tail F test: F = .955, p-value = .5403. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

302


t = 4.65, p-value = 1.00. The p-value tells us that in fact there is enough evidence that Liberals are more educated than conservatives.

13.65a. Two tail F test: F = .865, p-value = .1316. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.65, p-value = .0493. There is enough evidence to infer that conservatives work longer hours than do liberals. b. Hours of work are required to be normally distributed. c. Wilcoxon rank sum test.

13.66a. Two tail F test: F = .942, p-value = .4214. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 9.25, p-value = 0. There is enough evidence to conclude that government workers have more education than do private sector employees. b. The required condition is satisfied. c. Wilcoxon rank sum test

13.67a. Two tail F test: F = .534, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

303


t = 1.45, p-value = .0744. There is not enough evidence to infer that government workers have higher incomes than do private sector workers. b. Incomes are not normally distributed. c. Wilcoxon rank sum test

13.68a. Two tail F test: F = 2.36, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 3.09, p-value = .0011. There is enough evidence to infer that self-employed individuals have higher incomes. b. Incomes are not normally distributed. c. Wilcoxon rank sum test

13.69 Two tail F test: F = .900, p-value = .4602. Use equal variances estimator.

LCL = -35,341, UCL = 19,248.

13.70a. Two tail F test: F = .541, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

304


t = 1.02, p-value = .1541. There is not enough evidence to infer that people who work for someone else have higher incomes than self-employed people. b

LCL = -6,513, UCL = 20,433.

13.71 Two tail F test: F = .548, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -3.66, p-value = .0001. There is enough evidence to infer that heads of households who completed high school have more debt than those who did not finish high school.

13.72 Two tail F test: F = .502, p-value = 0. Use unequal variances estimator.

LCL = -23,715, UCL = -10,707.

13.73 Two tail F test: F = 1.031, p-value = .8148. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

305


t = 1.20, p-value = .2321. There is not enough evidence to conclude that a difference exists in unrealized capital gains between those who finished high school and those who did not.

13.74 Two tail F test: F = .927, p-value = .5336. Use equal variances estimator.

LCL = -24,385, UCL = -3,092.

13.75 Two tail F test: F = .534, p-value = 0. Use unequal variances estimator

LCL = -31,633, UCL = -16,571.

13.76 Two tail F test: F = .680, p-value = .0015. Use unequal variances test statistic H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = .34, p-value = .3680. There is not enough evidence to infer that heads of households with college graduates have smaller unrealized capital gains than do heads of households with only some college.

13.77a. Two tail F test: F = 2.07, p-value = 0. Use unequal variances test statistic H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

306


t = 10.67, p-value = 0. There is sufficient evidence to infer that male heads of households have higher incomes than female heads of households. b.

LCL = 23,059, UCL = 33,463. c. Incomes are not normally distributed.

13.78 Two tail F test: F = 1.009, p-value = .9274. Use equal variances test statistic H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -.085, p-value = .4660. The is not enough evidence to infer that heads of households who finished high school have greater net worth than those who did not finish high school.

13.79 Two tail F test: F = .519, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

307


t = -6.24, p-value = 0. here is enough evidence to infer that heads of households with college degrees have more assets than those with only some college.

13.80 Two tail F test: F = .404, p-value = 0. Use unequal variances test statistic. H0: (µ 4- µ 3) = 0 H1: (µ 4 - µ 3) < 0

t = -6.03, p-value = 0. There is enough evidence to infer that heads of households with some college have less debts than those with college degrees.

t =6.03, p-value = 0. There is enough evidence to conclude that households whose heads have some college have less debt than households whose heads completed a college degree.

13.81 The data are observational. Experimental data could be produced by randomly assigning babies to either Tastee or the competitor’s product.

13.82 Assuming that the volunteers were randomly assigned to eat either oat bran or another grain cereal the data are experimental.

13.83a. Since 40 students were selected randomly, but were given the choice as to which software package to use, the data must be observational. b. Experimental data could have been derived by selecting the 40 students at random and assigning either of the software packages at random.

308


c. Students may choose the software package to which they have prior experience and greater proficiency. The differences in the amount of time needed to learn how to use each software package may be a function of the popularity of one software package over the other.

13.84a Let students select the section they wish to attend and compare test results. b Randomly assign students to either section and compare test results.

13.85 Randomly assign patients with the disease to receive either the new drug or a placebo.

13.86a Randomly select finance and marketing MBA graduates and determine their starting salaries. b Randomly assign some MBA students to major in finance and others to major in marketing. Compare starting salaries after they graduate. c Better students may be attracted to finance and better students draw higher starting salaries.

13.87a The data are observational because to obtain experimental data would entail randomly assigning some people to smoke and others not to smoke. b It is possible that some people smoke because of a genetic defect (Genetics have been associated with alcoholism.), which may also be linked to lung cancer. c In our society the experiment described in part a is impossible.

H0 : D = 0

13.88

H1 :  D < 0 Rejection region: t   t ,   t .05,7  1.895 t

x D  D sD / n D

 4.75  0 4.17 / 8

 3.22 , p-value = .0073. There is enough evidence to infer that the Brand A is better than

Brand B.

13.89

H0: µ D = 0 H1: µ D < 0

Rejection region: t   t ,   t .05,7  1.895

t

xD D sD / nD

13.90

 .175  0 .225 / 8

 2.20, p-value = .0320. There is enough evidence to infer that ABS is better.

H0: µ D = 0 H1: µ D > 0

309


Rejection region: t  t ,  t .05,6  1.943

t

xD  D

1.86  0

sD / nD

 1.98, p-value = .0473. There is enough evidence to infer that the camera reduces the

2.48 / 7

number of red-light runners.

13.91a H0: µ D = 0 H1: µ D < 0 Rejection region: t   t ,   t.05,11  1.796

t

x D  D

 1.00  0

sD / nD

3.02 / 12

 –1.15, p-value = .1375. There is not enough evidence to infer that the new fertilizer is

better. b xD  t / 2

sD nD

=  1.00  2.201

3.02 12

 1.00  1.92 ; LCL = –2.92, UCL = .92

c The differences are required to be normally distributed d No, the histogram is bimodal. e The data are experimental. f The experimental design should be independent samples.

13.92 a H0: µ D = 0 H 1: µ D > 0 Rejection region: t  t ,  t.05,11  1.796

t

x D  D

3.08  0

sD / nD

 1.82, p-value = .0484. There is enough evidence to infer that companies with exercise

5.88 / 12

programs have lower medical expenses. b xD  t / 2

sD nD

= 3.08  2.201

5.88 12

 3.08  3.74 ; LCL = –.66, UCL = 6.82

c Yes because medical expenses will vary by the month of the year.

13.93

H 0: µ D = 0 H 1: µ D > 0

Rejection region: t  t ,  t.05,149  1.656 t

x D  D sD / n D

12 .4  0 99 .1 / 150

 1.53, p-value = .0638. There is not enough evidence to infer that mortgage payments

have increases in the past 5 years.

310


13.94

H 0: µ D = 0 H 1: µ D ≠ 0

Rejection region: t   t  / 2,   t .025, 49  2.009 or t  t  / 2,  t .025, 49  2.009 t

x D  D sD / n D

 1.16  0

2.22 / 50

 –3.70, p-value = .0005. There is enough evidence to infer that waiters and waitresses

earn different amounts in tips.

13.95 a x D  t  / 2 b

sD nD

= 19 .75  1.684

30 .63 40

 19 .75  8.16 ; LCL = 11.59, UCL = 27.91

H 0: µ D = 0 H 1: µ D > 0

Rejection region: t  t ,  t.05,39  1.684 t

x D  D sD / n D

19 .75  0 30 .63 / 40

 4.08, p-value = .0001. There is enough evidence to conclude that companies that

advertise in the Yellow Pages have higher sales than companies that do not. c The histogram of the differences is bell shaped. d No, because we expect a great deal of variation between stores.

13.96a H0: µ D = 0 H1: µ D < 0 Rejection region: t   t ,   t .10,14  1.345 t

xD  D sD / n D

 57 .40  0 13 .14 / 15

 16 .92 , p-value = 0. There is enough evidence to conclude that heating costs for

insulated homes is less than that for uninsulated homes. b xD  t / 2

sD nD

=  57 .40  2.145

13 .14 15

 57 .40  7.28 ; LCL = -64.68, UCL = -50.12; mean savings lies

between $50.12 and $64.80 c Differences are required to be normally distributed.

13.97

H 0: µ D = 0 H 1: µ D ≠ 0

Rejection region: t   t  / 2,   t .025, 44  2.014 or t  t  / 2,  t .025, 44  2.014

311


t

x D  D sD / n D

 42 .94  0

 –.91, p-value = .3687. There is not enough evidence to infer men and women spend

317 .16 / 45

different amounts on health care.

13.98

H0: µ D = 0 H1: µ D < 0

Rejection region: t   t ,   t.05,169  1.654 t

x D  D sD / n D

 183 .35  0 1568 .94 / 170

 –1.52, p-value = .0647. There is not enough to infer stock holdings have

decreased.

13.99

H0: µ D = 0 H1: µ D > 0

Rejection region: t  t ,  t.05,37  1.690

t

xD  D

sD / nD

.0422  0

 1.59, p-value = .0597. There is not enough evidence to conclude that ratios are higher

.1634 / 38

this year.

13.100 H0: µ D = 0 H1: µ D > 0 Rejection region: t  t ,  t.05,54  1.676

t

x D  D

sD / nD

520 .85  0

 2.08, p-value = .0210. There is enough evidence to infer that company 1’s

1854 .92 / 55

calculated tax payable is higher than company 2’s.

13.101 H0: µ D = 0 H1: µ D > 0 Rejection region: t  t ,  t.05,19  1.729 t

x D  D sD / n D

4.55  0 7.22 / 20

 2.82, p-value = .0055. There is enough evidence to that the new design tire lasts longer

than the existing design.

13.102 The matched pairs experiment reduced the variation caused by different drivers.

312


13.103 H0: µ D = 0 H 1: µ D > 0 Rejection region: t  t ,  t.05,24  1.711 t

x D  D sD / n D

4587  0 22 ,851 / 25

 1.00, p-value = .1628. There is not enough evidence to infer that finance majors

attract higher salary offers than do marketing majors.

13.104 Salary offers and undergraduate GPA are not as strongly linked as are salary offers and MBA GPA.

13.105 a H0: µ D = 0 H1: µ D < 0 Rejection region: t   t ,   t.05, 41  1.684

t

x D  D

sD / nD

 .10  0 1.95 / 42

 –.33, p-value = .3704. There is not enough evidence to infer that for companies where

an offspring takes the helm there is a decrease in operating income. b

H0: µ D = 0 H1: µ D > 0

Rejection region: t  t ,  t .05,97  1.660

t

x D  D sD / nD

1.24  0

 4.34, p-value = 0. There is enough evidence to infer that when an outsider becomes

2.83 / 98

CEO the operating income increases.

13.106 H0: µ D = 0 H1: µ D > 0

t = 24.93, p-value = 0. There is enough evidence to conclude that Americans are more educated than their fathers.

13.107 H0: µ D = 0 H1: µ D > 0

313


t = 28.38, p-value = 0. There is enough evidence to conclude that Americans are more educated than their mothers.

13.108

LCL = 23,515, UCL = 26,929.

13.109 H0: µ D = 0 H 1: µ D ≠ 0

t = .64, p-value = 5200. There is not enough evidence of a difference between spouses’ hours or work.

13.110 H0: µ D = 0 H1: µ D < 0

t = -3.28, p-value = .0005. There is enough evidence to infer that this year is worse than normal.

314


13.111

LCL = 3581, UCL = 6767.

13.112 a H 0 : 12 /  22  1

H1 : 12 /  22  1 Rejection region: F  F / 2,1 ,2  F.05,29,29  1.88 or F  F1 / 2,1 ,2  1 / F / 2,2 ,1  1 / F.05,29,29  1 / 1.88  .53 F = s12 / s 22 = 350/700 =.50, p-value = .0669. There is enough evidence to conclude that the population variances differ. b Rejection region: F  F / 2,1 ,2  F.025,14,14  2.98 or F  F1 / 2,1 ,2  1 / F / 2,2 ,1  1 / F.025,14,14  1 / 2.98  .34 F = s12 / s 22 = 350/700 =.50, p-value = .2071. There is not enough evidence to conclude that the population variances differ. c The value of the test statistic is unchanged and in this exercise the conclusion changed as well.  s2   s2  1  28   28  1 13.113 a LCL =  1  =   = .366, UCL =  1 F / 2, 2 ,1 =  4.03 = 5.939 2    s 2  F / 2, , 19 4 . 03  19     s2   2 1 2  s2   s2  1  28   28  1 =   = .649, UCL =  1 F / 2, 2 ,1 =  2.27 = 3.345 b LCL =  1  2    s 2  F / 2, ,  19   19  2.27  s2   2 1 2

c The interval narrows.

13.114

H 0 : 12 /  22  1 H1 : 12 /  22  1

Rejection region: F  F / 2,1 ,2  F.025,9,10  3.78 or F  F1 / 2,1 ,2  1 / F / 2,2 ,1  1 / F.025,10,9  1 / 3.96  .25 F = s12 / s 22 = .0000057/.0000114 =.50, p-value = .3179. There is not enough evidence to conclude that the two machines differ in their consistency of fills.

13.115

H 0 : 12 /  22  1

H1 : 12 /  22  1 Rejection region: F  F1,1 ,2  1 / F,2 ,1  1 / F.05,9,9  1 / 3.18  .314

315


F = s12 / s 22 = .1854/.1893 =.98, p-value = .4879. There is not enough evidence to infer that the second method is more consistent than the first method.

H 0 : 12 /  22  1

13.1116

H1 : 12 /  22  1 Rejection region: F  F / 2,1 ,2  F.025,10,10  3.72 or F  F1 / 2,1 ,2  1 / F / 2,2 ,1  1 / F.025,10,10  1 / 3.72  .269 F = s12 / s 22 = 193.67/60.00 = 3.23, p-value = .0784. There is not enough evidence to infer that the variances of the marks differ between the two sections.

13.117

H 0 : 12 /  22  1

H1 : 12 /  22  1 Rejection region: F  F,1 ,2  F.05,99,99  1.39 F = s12 / s 22 = 19.38/12.70 = 1.53, p-value = .0183. There is enough evidence to infer that limiting the minimum and maximum speeds reduces the variation in speeds.

13.118 H 0 : 12 /  22  1

H1 : 12 /  22  1 Rejection region: F  F / 2,1 ,2  F.025,99,99  1.48 or F  F1 / 2,1 ,2  1 / F / 2,2 ,1  1 / F.025,99,99  1 / 1.48  .68 F = s12 / s 22 = 41,309/19,850 = 2.08, p-value = .0003. There is enough evidence to conclude that the variances differ.

13.119

H 0 : 12 /  22 = 1 H1 : 12 /  22 < 1

Rejection region: F  F1,1 ,2  1 / F,2 ,1  1 / F.05,51,51  1 / 1.60  .63 F = s12 / s 22 = .0261/.0875 = .298, p-value = 0. There is enough evidence to infer that portfolio 2 is riskier than portfolio 1.

13.120

H 0 : 12 /  22 = 1 H1 : 12 /  22  1

Rejection region: F  F / 2,1 ,2  F.05,99,99  1.39 or F  F1,1 ,2  1 / F,2 ,1  1 / F.05,99,99  1 / 1.39  .72

316


F = s12 / s 22 = 3.35/10.95 = .31, p-value = 0. There is enough evidence to conclude that the variance in service times differ.

13.121

H 0 : 12 /  22 = 1

H1 : 12 /  22 > 1

F = 2.36, p-value = 0. There is enough evidence to infer that the variance in income is greater for self-employed individuals. Note; That the version of XLSTAT that the author had showed that the p-value is 1.000, which is clearly incorrect. The author changed the printout to show the correct result.

13.122

H 0 : 12 /  22 = 1

H1 : 12 /  22 > 1

F = 2.12, p-value = 0. There is enough evidence to conclude that the hours worked by the self-employed vary more than do the hours of employees who work for someone else. Note; That the version of XLSTAT that the author had showed that the p-value is 1.000, which is clearly incorrect. The author changed the printout to show the correct result.

13.123

H 0 : 12 /  22 = 1

H1 : 12 /  22 < 1 317


F = .54, p-value = 0. There is enough evidence to infer that the variation in income is greater for self-employed individuals.

13.124

H 0 : 12 /  22 = 1

H1 : 12 /  22 < 1

F = .904, p-value = .2390. There is not enough evidence to infer that the variation in net worth is greater for selfemployed individuals.

13.125

H 0 : 12 /  22 = 1 H1 : 12 /  22 < 1

318


F = .90, p-value = .2301. There is not enough evidence to infer that the variation in debt is greater for self-employed individuals.

13.126

H 0 : 12 /  22 = 1

H1 : 12 /  22 < 1

F = .652, p-value = .0014. There is enough evidence to infer that the variation in total capital gains is greater for self-employed individuals. 13.127 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

a z

b z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2  (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

=

(.45  .40 ) 1   1 .425 (1  .425 )   100 100   (.45  .40 ) 1   1 .425 (1  .425 )   400 400  

= .72, p-value = 2P(Z > .72) = 2(1 – .7642) = .4716.

= 1.43, p-value = 2P(Z > 1.43) = 2(1 – .9236) = .1528.

c The p-value decreases. 13.128 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 a z

bz 

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2  (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

=

(.60  .55) 1   1  .575 (1  .575 )  225 225   (.95  .90 ) 1   1 .925 (1  .925 )   225 225  

 1.07 , p-value = 2P(Z > 1.07) = 2(1 – .8577) = .2846

 2.01, p-value = 2P(Z > 2.01) = 2(1 – .9778) = .0444.

c. The p-value decreases. 319


d z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.10  .05) 1   1 .075 (1  .075 )    225 225 

 2.01, p-value = 2P(Z > 2.01) = 2(1 – .94778) = .0444.

e. The p-value decreases.

13.129 a (p̂1  p̂ 2 )  z  / 2

p̂1 (1  p̂ 1 ) p̂ 2 (1  p̂ 2 ) .18(1  .18) .22 (1  .22 )  = (.18–.22)  1.645  n2 n1 100 100

= –.040  .0929 b (p̂1  p̂ 2 )  z  / 2

p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 ) .48(1  .48) .52 (1  .52 )  = (.48–.52)  1.645  n1 n2 100 100

= –.040  .1162 c The interval widens. 13.130 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.205  .140 )

 1.70, p-value = P(Z > 1.70) = 1 – .9554 = .0446.

1   1 .177 (1  .177 )   229 178  

There is enough evidence to conclude that those who paid the regular price are more likely to buy an extended warranty. 13.131 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.145  .234 ) 1   1 .209 (1  .209 )    83 209 

 1.69, p-value = 2P(Z < −1.69) = 2(.0455) = .0910.

There is not enough evidence to conclude that new and old accounts are different with respect to overdue accounts. 13.132 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(. 26  .07 ) 1   1  .165 (1  .165    100 100 

 3.62 , p-value = P(Z > 3.62) = 0.

There is enough evidence to conclude that obese Chinese food eaters are less likely to use chop sticks?

320


13.133a H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.56  .46 ) 1   1 .518 (1  .518 )    1100 800 

= 4.31, p-value = 0. There is enough evidence to infer that

the leader’s popularity has decreased. b

H0: (p1 – p2) = .05 H1: (p1 – p2) > .05

Rejection region: z  z   z .05 = 1.645

z

(p̂1  p̂ 2 )  (p1  p 2 ) p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 )  n2 n1

=

(.56  .46 )  .05

= 2.16, p-value = P(Z > 2.16) = 1 – .9846 = .0154.

.56 (1  .56 ) .46 (1  .46 )  1100 800

There is enough evidence to infer that the leader’s popularity has decreased by more than 5%. c (p̂1  p̂ 2 )  z  / 2

p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 ) .56 (1  .56 ) .46 (1  .46 )  = (.56  .46 )  1.96  n1 n2 1100 800

= .10  .045 13.134 H0: (p1 – p2) = -.08 H1: (p1 – p2) < -.08 Rejection region: z  z  z.01 = –2.33

z

(p̂1  p̂ 2 )  (p1  p 2 ) p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 )  n1 n2

=

(.11  .28)  (.08) .11(1  .11) .28(1  .28)  300 300

= –2.85, p-value =P(Z < –2.85) = 1 – .9978 = .0022.

There is enough evidence to conclude that management should adopt process 1. 13.135 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.071  .064 ) 1   1  .068 (1  .068 )   1604 1109 

= .71, p-value = P(Z > .71) = 1 – .7611= .2389.

There is not enough evidence to infer that the claim is false. 13.136 a H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 321


Rejection region: z  z   z .05 = –1.645

z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.093  .115 ) 1   1 .104 (1  .104 )   6281 6281  

= –4.04, p-value = 0. There is enough evidence to infer

that Plavix is effective. 13.137 a H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .01 = –2.33 p̂1 

z

104  189 104 189  .0095 p̂ 2   .0172 p̂   .01335 11,000 11,000 22,000

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.0095  .0172 ) 1   1 .01335 (1  .01335 )    11,000 11,000 

= –4.98, p-value = 0. There is enough evidence

to infer that aspirin is effective in reducing the incidence of heart attacks. 13.138 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05  1.645 p̂1 

z

1,084 1,084  997 997  .0906 p̂   .0946  .0985 p̂ 2  11,000 22,000 11,000

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.0985  .0906 ) 1   1 .0946 (1  .0946 )    11,000 11,000 

 2.00 , p-value = P(Z > 2.00) = 1 – .9772 = .0228.

There is enough evidence to infer that aspirin leads to more cataracts. 13.139 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .01 = –2.33 p̂1 

z

132 75 75  132  .0399  .0289 p̂ 2   .0509 p̂  2,594 2,594 2,594  2,594

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.0289  .0509 ) 1   1 .0399 (1  .0399 )    2,594 2,594 

There is enough evidence to infer that Letrozole works.

322

 4.04 , p-value = 0.


13.140 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .05  1.645

p̂1  z

88  105 88 105  .2409  .2228 p̂ 2   .2586 p̂  395  406 395 406 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.2228  .2586 )

 1.19, p-value = P(Z < –1.19) = .1170. There is not

1   1 .2409 (1  .2409 )   395 406  

enough evidence to infer that exercise training reduces mortality. 13.141 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05  1.645

p̂1  z

27 18  27 18  .1585  .1895 p̂ 2   .1429 p̂  95  189 95 189 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(. 1895  .1429 ) 1   1 .1585 (1  .1585 )   95 189  

 1.02 , p-value = P(Z > 1.02) = 1-.8461 = .1539.

There is not enough evidence to infer that women drivers in a following car are more likely to honk when the lead car driver is using a cell phone. 13.142 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05  1.645

p̂1  z

35  34 34 35  .2536 p̂ 2   .2018  .1667 p̂  138 138  204 204 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(. 2536  .1667 ) 1   1 .2018 (1  .2018 )    138 204 

 1.97 , p-value = P(Z > 1.97) = 1-.9756 = .0244.

There is enough evidence to infer that men drivers in a following car are more likely to honk when the lead car driver is using a cell phone. 13.143 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .05  1.645

323


p̂1  z

33  49 49 33  .2129 p̂ 2   .2547  .2934 p̂  155  167 155 167 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(. 2129  .2934 ) 1   1 .2547 (1  .2547 )    155 167 

 1.66, p-value = P(Z < –1.66) = .0485

There is enough evidence to allow us to conclude that drivers behind expensive cars are less likely to honk. 13.144 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05  1.645

p̂1  z

64  47 47 64  .3000  .3422 p̂ 2   .2568 p̂  187  183 187 183 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(. 3422  .2568 ) 1   1  .3000 (1  .3000 )   187 183 

 1.79 , p-value = P(Z > 1.79) = 1- .9633 = .0367

There is enough evidence to allow us to conclude that drivers behind a male driver are more likely to honk. 13.145 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .05  1.645

p̂1  z

68  105 105 68  .3223 p̂ 2   .3836  .4375 p̂  211 211  240 240 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(. 3223  .4375 ) 1   1  .3836 (1  .3836 )   211 240 

 2.51, p-value = P(Z < –2.51) = .0060

There is enough evidence to allow us to conclude that women are less likely to honk. 13.146 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .05  1.645

p̂1  z

86 21  86 21  .2616  .1858 p̂ 2   .2905 p̂  113  296 113 296 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(. 1858  .2905 ) 1   1  .2616 (1  .2616 )   113 296 

 2.15, p-value = P(Z < –2.15) = .0158

There is enough evidence to allow us to draw the inference that drivers of convertibles are less likely to honk.

324


13.147 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 (p̂1  p̂ 2 )

a z

 1 1   p̂(1  p̂)   n1 n 2 

=

(.9057  .8878 1   1 .8975 (1  .8975 )    350 294 

 .75,

p-value = 2P(Z > .75) = 2(1 – .7734)= .4532. There is not enough evidence to infer that the two populations of car owners differ in their satisfaction levels. 13.148a H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .10 = 1.28 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.2632  .0741 ) 1   1 .11(1  .11)   38 162  

= 3.35, p-value = 0. There is enough evidence to conclude that

smokers have a higher incidence of heart diseases than nonsmokers. b (p̂1  p̂ 2 )  z  / 2

p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 )  n2 n1

= (.2632–0741)  1.645

.2632 (1  .2632 ) .0741 (1  .0741 ) =.1891  .1223; LCL = .0668, UCL = .3114  38 162

13.149 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.3585  .3420 ) 1   1 .3504 (1  .3504 )    477 462 

= .53, p-value = P(Z > .53) = 1 – .7019 = .2981.

There is not enough evidence to infer that the use of illicit drugs in the United States has increased in the past decade. 13.150 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 Rejection region: z  z  / 2  1.96, z  z  / 2 , z .025  1.96 1 = Success

z

( p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.5169  .375 ) 1   1 .4463 (1  .4463 )   445 440   325

= 4.24, p-value = 2P(Z > 4.24) = 0


There is enough evidence to infer that Canadians and Americans differ in their responses to the survey question. 13.151 H0: (p1 – p2) = -.02 H1: (p1 – p2) < -.02 Rejection region: z  z   z .05 = –1.645

z

(p̂1  p̂ 2 )  (p1  p 2 ) p̂1 (1  p̂1 ) p̂ 2 (1  p̂ 2 )  n1 n2

=

(.055  .11)  (.02 ) .055 (1  .055 ) .11(1  .11)  200 200

= –1.28,

p-value = P(Z < –1.28) = .1003. There is not enough evidence to choose machine A. 13.152 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645 2 = Success

z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(. 8608  .7875 ) 1   1   .8394 (1  .8394 ) 80 194  

= 1.50, p-value = .0668.

There is not enough evidence to infer that those with more education are less likely to work 11 hours or more per day. 13.153 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 Rejection region: z  z   z .05 = -1.645 z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(. 6058  .6480 ) 1   1  .6288 (1  .6288 )   104 125 

= -0.66, p-value = .2551

There is not enough evidence to infer that Americans are more dissatisfied with their jobs in 2011 than they were in 2008. 13.154 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645

326


z = 3.64, p-value = .0001. There is enough evidence to infer that the proportion of American adults who believe that global warming is a very serious problem decreased between 2012 and 2013. 13.155 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 Rejection region: z  z  / 2  1.96, z  z  / 2 , z .025  1.96

z = 1.56, p-value = .1198. There is not enough evidence to infer that men and women differ in their support of the pipeline. 13.156 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

327


z = 1.47, p-value = .0702. There is not enough evidence to infer that including a reference to religious activity reduces the probability of a callback. 13.157 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -.65, p-value = .2588. There is not enough evidence to infer that the Wallonians had a higher callback frequency than did main stream religions. 13.158 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 6.21, p-value = 0. There is enough evidence to infer that females are more likely to be in ideal cardiovascular health than males. 13.159 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = -3.97, p-value = .0001. There is sufficient evidence that there is a difference in the error rate between the two umpires.

328


13.160 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = -.29, p-value = .7708. There is not enough evidence that there is a difference in the error rate between the two umpires. 13.16a. H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = 1.90, p-value = .0570. There is not enough evidence of a difference.

b.

H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = 4.74, p-value = 0. There is enough evidence of a difference. 13.162 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

329


z = 3.84, p-value = .0001. There is enough evidence to infer that government workers are thriving compared to other workers. 13.163 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -4.81, p-value = .0702. There is enough evidence to infer that American adults are less satisfied than five years ago. 13.164 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645

z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.1385  .0905 ) 1   1 .1035 (1  .1035 )    231 619 

z = 2.04, p-value = P(Z > 2.04) = 1 – .9793 = .0207.

There is enough evidence to conclude that health conscious adults are more likely to buy Special X. 13.165a H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 Rejection region: z  z   z .05 = 1.645

p̂1 

68 29 68  29  .3579  .4172 p̂ 2   .2685 p̂  163 108 163  108

330


z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.4172  .2685 ) 1   1 .3579 (1  .3579 )    163 108 

 2.50,

p-value = P(Z > 2.50) = 1 – .9938 = .0062. There is enough evidence to conclude that members of segment 1 are more likely to use the service than members of segment 4. b

H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

Rejection region: z  z  / 2  z .05 = –1.96 or z  z  / 2  z .05 = 1.96

p̂ 1  z

20  10 20 10  .3896  .3704 p̂ 2   .4348 p̂  54  23 54 23 (p̂1  p̂ 2 )  1 1   p̂(1  p̂)  n n 2   1

=

(.3704  .4348 1   1 .3896 (1  .3896 )    54 23 

 .53,

p-value = 2P(Z < –.53) = 2(.2981) = .5962. There is not enough evidence to infer that retired persons and spouses that work in the home differ in their use of services such as Quik Lube. 13.166 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 Rejection region: z  z  / 2  z .025 = –1.96 or z  z  / 2  z .025 = 1.96

z

(p̂1  p̂ 2 )  1 1   p̂(1  p̂)   n1 n 2 

=

(.0995  .1297 ) 1   1 .1132 (1  .1132 )    382 316 

= –1.25, p-value = 2P(Z < –1.25) = 2(.1056) = .2112.

There is not enough evidence to infer differences between the two sources.

13.167 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = .98, p-value = .2393. There is not enough evidence to infer that a difference exists.

331


13.168 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = 3.56, p-value = .0004. There is enough evidence to conclude that a difference exists. 13.169 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -2.54, p-value = .0055. There is enough evidence to infer that foreign-born people are more likely to have a graduate degree than native-born Americans. 13.170 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -2.76, p-value = .0029. There is enough evidence to conclude that women are more likely to work for the government. 13.171 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

332


z = 1.53, p-value = .0624. There is not enough evidence to conclude that Democrats are more likely to work for the government than Republicans. 13.172 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -1.02, p-value = .1528. There is not enough evidence to conclude that Republicans are more likely than Democrats to be working full time. 13.173 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -3.92, p-value = 0. There is enough evidence to conclude that Republicans are more likely than Democrats to be working for themselves. 13.174 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

333


z = .95, p-value = .3439. There is not enough evidence to infer that married people are more likely to be working full time than never married people. 13.175 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 4.89, p-value = 0. There is enough evidence to infer that married people are more likely to be working for themselves than never married people. 13.176 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = 7.31, p-value = 0. There is enough evidence to infer that there is a difference in the proportion of men and women working full time. 13.177 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

334


z = 5.85, p-value = 0. There is enough evidence to infer that there is a difference between married and never married people with respect to completion of a graduate degree. 13.178 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 1.64, p-value = .0502. There is not enough evidence to infer that married people are more likely to be working for the government than never married people. 13.179 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 3.25, p-value = .0006. There is enough evidence to conclude that native-born Americans are more likely to work for the government. 13.180 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

335


z = -.33, p-value = .3693. There is not enough evidence to infer that foreign-born people are more likely than nativeborn Americans to work for themselves. 13.181 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = .49, p-value = .6260. There is not enough evidence to infer that the proportion of men and women who answer correctly differ. 13.182 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = .42, p-value = .6773. There is not enough evidence to infer that the proportion of men and women who answer correctly differ. 13.183 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

336


z = 4.34, p-value = 0. There is enough evidence to infer that the proportion of men and women who answer correctly differ. 13.184 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

z = 6.21, p-value = 0. There is enough evidence to infer that the proportion of men and women who answer correctly differ. 13.185 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 10.45, p-value = 0. There is enough evidence to infer that male heads of households are more likely to have a college degree than female heads of households. 13.186 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

337


z = 11.34, p-value = 0. There is enough evidence to infer that male heads of households are more likely to have a higher employment rate than female heads of households. 13.187 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 16.14, p-value = 0. There is enough evidence to infer that male heads of households are more likely to own the home they live in than female heads of households. 13.188 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 53.36, p-value = 0. There is overwhelming evidence to infer that male heads of households are more likely to be married or living with partner than female heads of households.

13.89 It appears that female heads of households are likely to be never married or divorced and have less education, rent their homes, and less likely to be working. 13.190 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0 338


z = 11.82, p-value = 0. There is enough evidence to infer that married people are more likely to have a college degree. 13.191 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

z = -12.02, p-value = 0. There is enough evidence to infer that married people are less likely to be unemployed. 13.192 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 14.67, p-value = 0. There is enough evidence to infer that married people are more likely to be self-employed. 13.193 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

339


z = -.10, p-value = .4589. There is not enough evidence to infer that married people are less likely to have declared bankruptcy in the last five years. 13.194a. H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 2.49, p-value = .0065. There is enough evidence to conclude that there has been a decrease in participation among boys over the past 10 years. b.

H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = .893, p-value = .1859. There is not enough evidence to conclude that there has been a decrease in participation among girls over the past 10 years. c.

H0: (p1 – p2) = 0 340


H1: (p1 – p2) > 0

z = 2.61, p-value = .0045. There is enough evidence to infer girls are less likely to participate in sports than boys in 2011.

13.195 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .859, p-value = .7850; use equal-variances t-test

t = 2.65, p-value = .0059. There is enough evidence to conclude that the campaign is successful.

13.196 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 2.57, p-value = 0; use unequal-variances t-test

341


t = 15.82, p-value = 0. There is sufficient evidence to infer that government employees work less than private-sector workers.

13.197 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = .551, p-value = .0054; use unequal-variances t-test

t = –8.73, p-value = 0. There is enough evidence to conclude that men and women who suffer heart attacks vacation less than those who do not suffer heart attacks.

13.198

H0 : D = 0 H1 :  D  0

342


t = –6.09, p-value = 0. There is enough evidence to infer that the new drug is effective. 13.199 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = .958, p-value = .7726; use equal-variances t-test

t = –1.29, p-value = .1993. There is not enough evidence of a difference in reading time between the two groups. 13.200 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

343


z = –2.30, p-value = .0106. There is enough evidence to infer an increase in seatbelt use. 13.201a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = .811, p-value = .1404; use equal-variances t-test

t = –2.02, p-value = .0218. There is enough evidence to infer that housing cost a percentage of total income has increased. b The histograms are be bell shaped.

13.202a. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = 1.10, p-value = .8430; use equal-variances t-test

344


A B C 1 t-Test: Two-Sample Assuming Equal Variances 2 3 Male Female 4 Mean 39.75 49.00 5 Variance 803.88 733.16 6 Observations 20 20 7 Pooled Variance 768.52 8 Hypothesized Mean Difference 0 9 df 38 10 t Stat -1.06 11 P(T<=t) one-tail 0.1490 12 t Critical one-tail 1.3042 13 P(T<=t) two-tail 0.2980 14 t Critical two-tail 1.6860 t = –1.06, p-value = .2980. There is not enough evidence to conclude that men and women differ in the amount of time spent reading magazines. b

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

Two-tail F test: F = .266, p-value = .0053; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 Low High 3 33.10 56.84 4 Mean 278.69 1047.81 5 Variance 6 Observations 21.00 19 0.00 7 Hypothesized Mean Difference 26.00 8 df -2.87 9 t Stat 0.0040 10 P(T<=t) one-tail 1.3150 11 t Critical one-tail 0.0080 12 P(T<=t) two-tail 13 t Critical two-tail 1.7056 t = –2.87, p-value = .0040. There is enough evidence to conclude that high-income individuals devote more time to reading magazines than do low-income individuals. 13.203a H 0 :  D = 0

H1 :  D  0

345


A B 1 t-Test: Paired Two Sample for Means 2 3 Female 4 Mean 55.68 5 Variance 105.64 6 Observations 25 7 Pearson Correlation 0.96 8 Hypothesized Mean Difference 0 9 df 24 10 t Stat -1.13 11 P(T<=t) one-tail 0.1355 12 t Critical one-tail 1.3178 13 P(T<=t) two-tail 0.2710 14 t Critical two-tail 1.7109

C

Male 56.40 116.75 25

t = –1.13, p-value = .2710. There is no evidence to infer that gender is a factor. b A large variation within each gender group was expected. c The histogram of the differences is somewhat bell shaped. 13.204 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

This Year 3 Years Ago 0.4351 0.3558 393 385 0 2.26 0.0119 1.2816 0.0238 1.6449

z = 2.26, p-value = .0119. There is enough evidence to infer that Americans have become more distrustful of television and newspaper reporting this year than they were three years ago.

13.205 H0: (µ 1 - µ 2) = 25 H1: (µ 1 - µ 2) > 25 Two-tail F test: F = 1.35, p-value = .4809; use equal-variances t-test

346


t = 1.75, p-value = .0433. There is enough evidence to conclude that machine A should be purchased. 13.206 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 The totals in columns A through D are 5788, 265, 5154, and 332, respectively.

1 2 3 4 5 6 7

A B C D z-Test of the Difference Between Two Proportions (Case 1)

Sample proportion Sample size Alpha

Sample 1 Sample 2 z Stat 0.045800 0.064400 P(Z<=z) one-tail 5788 5154 z Critical one-tail 0.05 P(Z<=z) two-tail z Critical two-tail

E

-4.28 0.0000 1.6449 0.0000 1.9600

z = –4.28, p-value = 0. There is enough evidence to infer that the defective rate differs between the two machines.

13.207

H0 : D = 0 H1 :  D  0

Dry Cleaner

347


A B C 1 t-Test: Paired Two Sample for Means 2 3 Dry C Before Dry C After 4 Mean 168.00 165.50 5 Variance 351.38 321.96 6 Observations 14 14 7 Pearson Correlation 0.86 8 Hypothesized Mean Difference 0 9 df 13 10 t Stat 0.96 11 P(T<=t) one-tail 0.1780 12 t Critical one-tail 1.7709 13 P(T<=t) two-tail 0.3559 14 t Critical two-tail 2.1604 t = .96, p-value = .1780. There is not enough evidence to conclude that the dry cleaner sales have decreased. Doughnut shop

A B 1 t-Test: Paired Two Sample for Means 2 3 Donut Before 4 Mean 308.14 5 Variance 809.67 6 Observations 14 7 Pearson Correlation 0.86 8 Hypothesized Mean Difference 0 9 df 13 10 t Stat 3.24 11 P(T<=t) one-tail 0.0032 12 t Critical one-tail 1.7709 13 P(T<=t) two-tail 0.0065 14 t Critical two-tail 2.1604

C

Donut After 295.29 812.07 14

t = 3.24, p-value = .0032. There is enough evidence to conclude that the doughnut shop sales have decreased. Convenience store

348


A B C 1 t-Test: Paired Two Sample for Means 2 3 Convenience Before Convenience After 4 Mean 374.64 348.14 5 Variance 2270.40 2941.82 6 Observations 14 14 7 Pearson Correlation 0.97 8 Hypothesized Mean Difference 0 9 df 13 10 t Stat 7.34 11 P(T<=t) one-tail 0.0000 12 t Critical one-tail 1.7709 13 P(T<=t) two-tail 0.0000 14 t Critical two-tail 2.1604 t = 7.34, p-value = 0. There is enough evidence to conclude that the convenience store sales have decreased.

13.208 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.77, p-value = .0084; use unequal-variances test statistic

z = -4.53, p-value = 0. There is enough evidence to conclude that Facebook users maintain lower GPA scores. 13.209a H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

349


A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

Depressed Not Depressed 0.2879 0.2004 132 1058 0 2.33 0.0100 2.3263 0.0200 2.5758

z = 2.33, p-value = .0100. There is enough evidence to infer that men who are clinically depressed are more likely to die from heart diseases. b No, we cannot establish a causal relationship.

13.210 a H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .438, p-value = .0482; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 Exercise Drug 3 13.52 9.92 4 Mean 5.76 13.16 5 Variance 25 25 6 Observations 0 7 Hypothesized Mean Difference 42 8 df 4.14 9 t Stat 0.0001 10 P(T<=t) one-tail 2.4185 11 t Critical one-tail 12 P(T<=t) two-tail 0.0002 13 t Critical two-tail 2.6981 t = 4.14, p-value = .0001. There is enough evidence that exercise is more effective than medication in reducing hypertension. b 1 2 3 4 5 6 7 8

A B C D t-Estimate of the Difference Between Two Means (Unequal-Variances)

Mean Variance Sample size Degrees of freedom Confidence level

E

Sample 1 Sample 2 Confidence Interval Estimate 13.52 9.92 3.60  5.76 13.16 Lower confidence limit 25 25 Upper confidence limit 41.63 0.95

LCL = 1.84, UCL = 5.36 350

F

1.76 1.84 5.36


c The histograms are bell shaped.

13.211 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = 1.93, p-value = .0232; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Group 1 Group 2 4 Mean 7.46 8.46 5 Variance 25.06 12.98 6 Observations 50 50 7 Hypothesized Mean Difference 0 8 df 89 9 t Stat -1.14 10 P(T<=t) one-tail 0.1288 11 t Critical one-tail 1.6622 12 P(T<=t) two-tail 0.2575 13 t Critical two-tail 1.9870 t = –1.14, p-value = .1288. There is not enough evidence to conclude that people who exercise moderately more frequently lose weight faster

13.212

H0 : D = 0 H1 :  D  0

A B C 1 t-Test: Paired Two Sample for Means 2 3 Group 1 Group 2 4 Mean 7.53 8.57 5 Variance 29.77 43.37 6 Observations 50 50 7 Pearson Correlation 0.89 8 Hypothesized Mean Difference 0 9 df 49 10 t Stat -2.40 11 P(T<=t) one-tail 0.0100 12 t Critical one-tail 1.6766 13 P(T<=t) two-tail 0.0201 14 t Critical two-tail 2.0096 t = –2.40, p-value = .0100. There is enough evidence to conclude that people who exercise moderately more frequently lose weight faster

13.213 Two tail F test: F = 1.20, p-value = .0375. Use unequal variances t test

351


H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 7.89, p-value = 0. There is enough evidence to infer consumers are optimistic about the economy.

13.214 Two tail F test: F = 1.69, p-value = .0156. Use unequal variances t test H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 6.08, p-value = 0. There is enough evidence to conclude that kidneys from living donors last longer than kidneys from deceased donors.

13.215 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 352


Two-tail F test: F = .363, p-value = .0161; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Small space Large space 4 Mean 1245.7 1915.8 5 Variance 23812 65566 6 Observations 25 25 7 Hypothesized Mean Difference 0 8 df 39 9 t Stat -11.21 10 P(T<=t) one-tail 0.0000 11 t Critical one-tail 1.6849 12 P(T<=t) two-tail 0.0000 13 t Critical two-tail 2.0227 t = –11.21, p-value = 0. There is enough evidence to infer that students write in such a way as to fill the allotted space.

13.216 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .624, p-value = .0431; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Computer No Computer 4 Mean 69,933 48,246 5 Variance 63,359,040 101,588,525 6 Observations 89 61 7 Hypothesized Mean Difference 0 8 df 109 9 t Stat 14.07 10 P(T<=t) one-tail 0.0000 11 t Critical one-tail 1.2894 12 P(T<=t) two-tail 0.0000 13 t Critical two-tail 1.6590 t = 14.07, p-value = 0. There is enough evidence to conclude that single-person businesses that use a PC earn more. 13.217 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

353


A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

New 0.948 250 0 1.26 0.1037 1.2816 0.2074 1.6449

Older 0.920 250

z = 1.26, p-value = .1037. There is not enough evidence to conclude that the new company is better.

13.218 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = 1.62, p-value = .1008; use equal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Equal Variances 2 3 Supplement Placebo 4 Mean 19.02 21.85 5 Variance 41.34 25.49 6 Observations 48 48 7 Pooled Variance 33.41 8 Hypothesized Mean Difference 0 9 df 94 10 t Stat -2.40 11 P(T<=t) one-tail 0.0092 12 t Critical one-tail 1.6612 13 P(T<=t) two-tail 0.0183 14 t Critical two-tail 1.9855 t = –2.40, p-value = .0092. There is enough evidence to infer that taking vitamin and mineral supplements daily increases the body's immune system? 13.219 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

354


A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

$100-Limit $3000-Limit 0.5234 0.5551 491 490 0 -1.00 0.1598 1.2816 0.3196 1.6449

z = –1.00, p-value = .1598. There is not enough evidence to infer that the dealer at the more expensive table is cheating.

13.220 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = 1.43, p-value = 0; use unequal-variances test statistic

t = 0.71, p-value = .4763. There is not enough evidence to conclude that math test scores differ between males and females. 13.221 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.04, p-value = .6531; use equal-variances test statistic

355


t = 2.59, p-value = .0049. There is sufficient evidence to conclude that American cars this year are on average older than cars last year. 13.222 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 1.57, p-value = .0578. There is not enough evidence to infer that Americans are more generous.

13.223 Time:

H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

Two-tail F test: F = .257, p-value = .0009; use unequal-variances test statistic

356


t = 4.96, p-value = 0. There is sufficient evidence to conclude that when the soft relaxing music was played diners spent more time in the restaurant.

Bar bill: H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 2.05, p-value = .0723; use equal-variances test statistic

t = 2.68, p-value = .0049. There is sufficient evidence to conclude that when the soft relaxing music was played diners spent more money on drinks.

13.224 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.61, p-value = .0429; use unequal-variances test statistic

357


t = 12.67, p-value = 0. There is sufficient evidence to infer that when the restaurant features bright lights and loud music customers spend less time in the restaurant. 13.225 H0: (p1 – p2) = 0 H1: (p1 – p2) > 0

z = 3.32, p-value = .0005. There is enough evidence to conclude that when the restaurant features bright lights and loud music customers are less likely to return.

13.226 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .869, p-value = .6699; use equal-variances test statistic

358


t = 16.47, p-value = 0. There is enough evidence to conclude that the amounts in the short wide glass are greater than the amounts in the tall skinny glasses.

13.227 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0 Two-tail F test: F = .901, p-value = .6052; use equal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Equal Variances 2 3 Female Male 4 Mean 7.27 7.02 5 Variance 2.57 2.85 6 Observations 103 97 7 Pooled Variance 2.71 8 Hypothesized Mean Difference 0 9 df 198 10 t Stat 1.08 11 P(T<=t) one-tail 0.1410 12 t Critical one-tail 1.6526 13 P(T<=t) two-tail 0.2820 14 t Critical two-tail 1.9720 t = 1.08, p-value = .2820. There is not enough evidence to conclude that female and male high school students differ in the amount of time spent at part-time jobs.

13.228 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 1.41, p-value = .1433; use equal-variances t-test

359


A B C 1 t-Test: Two-Sample Assuming Equal Variances 2 City Suburb 3 4 Mean 2.42 1.97 5 Variance 1.08 0.77 6 Observations 70 78 7 Pooled Variance 0.92 0 8 Hypothesized Mean Difference 146 9 df 2.85 10 t Stat 11 P(T<=t) one-tail 0.0025 12 t Critical one-tail 1.6554 13 P(T<=t) two-tail 0.0050 14 t Critical two-tail 1.9763

t = 2.85, p-value = .0025. There is enough evidence to infer that city households discard more newspaper than do suburban households.

13.229 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = 2.73, p-value = 0; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Teenagers 20-to-30 4 Mean 18.18 14.30 5 Variance 357.32 130.79 6 Observations 176 154 7 Hypothesized Mean Difference 0 8 df 293 9 t Stat 2.28 10 P(T<=t) one-tail 0.0115 11 t Critical one-tail 1.6501 12 P(T<=t) two-tail 0.0230 13 t Critical two-tail 1.9681 t = 2.28, p-value = .0115.There is enough evidence to infer that teenagers see more movies than do twenty to thirty year olds. 13.230 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0

360


A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

No HS HS 0.127 0.358 63 257 0 -3.54 0.0002 1.6449 0.0004 1.96

z = –3.54, p-value = .0002. There is enough evidence to conclude that Californians who did not complete high school are less likely to take a course in the university’s evening program.

13.231 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0 Two-tail F test: F = .433, p-value = 0; use unequal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Group 1 Groups 2-4 4 Mean 11.58 10.60 5 Variance 9.28 21.41 6 Observations 269 981 7 Hypothesized Mean Difference 0 8 df 644 9 t Stat 4.15 10 P(T<=t) one-tail 0.0000 11 t Critical one-tail 1.6472 12 P(T<=t) two-tail 0.0000 13 t Critical two-tail 1.9637 t = 4.15, p-value = 0. There is enough evidence to conclude that on average the market segment concerned about eating healthy food (group 1) outspends the other market segments.

13.232 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = .501, p-value = 0; use unequal-variances t-test

361


A B C 1 t-Test: Two-Sample Assuming Unequal Variances 2 3 Sale-CDs Sale-fax 4 Mean 59.04 65.57 5 Variance 425.4 849.7 6 Observations 122 144 7 Hypothesized Mean Difference 0 8 df 256 9 t Stat -2.13 10 P(T<=t) one-tail 0.0171 11 t Critical one-tail 1.6508 12 P(T<=t) two-tail 0.0341 13 t Critical two-tail 1.9693 t = –2.13, p-value = .0171. There is enough evidence to conclude that those who buy the fax/copier outspend those who buy the package of CD-ROMS.

13.233 H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0 Two-tail F test: F = 1.15, p-value = .5373; use equal-variances t-test

A B C 1 t-Test: Two-Sample Assuming Equal Variances 2 3 No Yes 4 Mean 91,467 97,836 5 Variance 461,917,705 401,930,840 6 Observations 466 55 7 Pooled Variance 455,676,297 8 Hypothesized Mean Difference 0 9 df 519 10 t Stat -2.09 11 P(T<=t) one-tail 0.0184 12 t Critical one-tail 1.6478 13 P(T<=t) two-tail 0.0369 14 t Critical two-tail 1.9645 t = –2.09, p-value = .0184. There is enough evidence to infer that professors aged 55 to 64 who plan to retire early have higher salaries than those who don’t plan to retire early. 13.234 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0 The data must first be unstacked. Success = 2

362


A 1 z-Test: Two Proportions 2 3 4 Sample Proportions 5 Observations 6 Hypothesized Difference 7 z Stat 8 P(Z<=z) one tail 9 z Critical one-tail 10 P(Z<=z) two-tail 11 z Critical two-tail

B

C

D

Female Male 0.5945 0.6059 762 746 0 -0.45 0.3256 1.6449 0.6512 1.96

z = −.45, p-value = .6512. There is not enough evidence to conclude that men and women differ in their choices of Christmas trees.

13.235a. Two tail F test: F = .978, p-value = .6929. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

t = .29, p-value = .7735. There is not enough evidence of a difference in years of education between men and women. b. Years of education are required to be normally distributed. c. The histograms are bell shaped. d. The Wilcoxon rank sum test. 13.236 H0: (p1 – p2) = 0 H1: (p1 – p2) < 0 363


z = -1.29, p-value = .1000. There is not enough evidence to conclude that the fraction of Americans working for the government increased between 2008 and 2014.

13.237 Two tail F test: F = .974, p-value = .7060. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = .71, p-value = .2378. There is not enough evidence to conclude that men watch more television than women.

13.238 Two tail F test: F = 1.036, p-value = .3994. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

364


t = -1.84, p-value = .0326. There is enough evidence to infer that Americans are more educated in 2014 than in 2012.

13.239 Two tail F test: F = 1.22, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 1.06, p-value = .1473. There is not enough evidence to infer that Americans are watching less televison on 2014 than in 2012. 13.240 H0: (p1 – p2) = 0 H1: (p1 – p2) ≠ 0

365


z = -.74, p-value = .4604. There is not enough evidence to conclude that the percentage of Americans with graduate degrees changed between 2012 and 2014.

13.241 Two tail F test: F = 1.01, p-value = .9109. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -.96, p-value = .1683. There is not enough evidence to conclude that Americans were working longer in 2014 than in 2010.

13.242 Two tail F test: F = 1.02, p-value = .7914. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

366


t = -1.55, p-value = .0612. There is not enough evidence to infer that immigrants have more children than nativeborn Americans.

13.243 Two tail F test: F = .873, p-value = .0147. Use unequal variances estimator.

LCL = -7843, UCL = -1952.

13.244 Two tail F test: F = 1.09, p-value = .4279. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

367


t = .78, p-value = .2170. There is not enough evidence to infer that native-born Americans work longer hours than do immigrants.

13.245 Two tail F test: F = 2.02, p-value = 0. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 3.67, p-value = .0001. There is enough evidence to infer that Democrats watch more television than do Republicans.

13.246 Two tail F test: F = 1.01, p-value = .8610. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 368


H1: (µ 1 - µ 2) < 0

t = -4.97, p-value = 0. There is enough evidence to infer that conservatives are older than liberals.

13.247

LCL = -.0003, UCL = .0373.

13.248 Two tail F test: F = .621, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

369


t = .006, p-value = .5023. There is no evidence that government workers work fewer hours than do private sector workers.

13.249 Two tail F test: F = 1.026, p-value = .7124. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 4.24, p-value = 0. There is enough evidence to conclude that government workers are older than private sector workers. 370


13.250a. Two tail F test: F = 2.12, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = -1.48, p-value = .9296. The p-value indicates that there is some evidence that self-employed people work less hours than people who work for someone else. b. The required condition of normality is satisfied. c. The Wilcoxon rank sum test.

13.251a. Two tail F test: F = .869, p-value = .1233. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

371


t = 2.67, p-value = .0038. There is enough evidence to infer that self-employed people have more education than other workers. b. The required condition is satisfied.

13.252a. Two tail F test: F = .931, p-value = .0666. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -6.50, p-value = 0. There is enough evidence to infer that the United States has grown older between 2004 and 2014.

372


13.253 Two tail F test: F = .743, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = 15.64, p-value = 0. There is enough evidence to conclude that the number of children per family decreased between 2004 and 2014.

13.254 Two tail F test: F = .789, p-value = 0. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -1.82, p-value = .0345. There is not enough evidence to conclude that the age at which families have their first child has increased between 2004 and 2014.

373


13.255 Two tail F test: F = .885, p-value = .0016. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = 0, p-value = .4986. There is no evidence to infer that the United States is more educated in 2014 than in 2004.

13.256 Two tail F test: F = 1.096, p-value = .0665. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) > 0

t = .69, p-value = .2452. There is not enough evidence to infer that Americans are working less in 2014 than in 2004.

13.257 Two tail F test: F = 1.02, p-value = .6926. Use equal variances test statistic.

374


H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) ≠ 0

t = -1.09, p-value = .2780. There is not enough evidence to conclude that the amount of television differs between 2004 and 2014.

13.258 Two tail F test: F = .904, p-value = .4779. Use equal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

375


t = -1.61, p-value = .0537. There is not enough evidence to conclude that self-employed heads of households have less net worth than heads of households who work for someone else.

13.259 Two tail F test: F = .652, p-value = .0028. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

t = -2.15, p-value = .0168. There is enough evidence to conclude that heads of households who work for someone else have larger unrealized capital gains than heads of households who are self-employed.

13.260a. Two tail F test: F = .706, p-value = .0251. Use unequal variances test statistic. H0: (µ 1 - µ 2) = 0 H1: (µ 1 - µ 2) < 0

376


t = -2.70, p-value = .0038. There is enough evidence to infer that heads of households with a high school diploma have more household assets than do heads of households who did not finish high school.

Case 13.1 2000 to 2008 Two tail F test of RINCOME: F = .4779, p-value = 0.

377


Chapter 14 14.1a x 

5(10 )  5(15 )  5(20 ) = 15 555

 n (x  x)  5(10 – 15) + 5(15 – 15) + 5(20 – 15) = 250 SSE =  (n  1)s  (5 –1)(50) + (5 – 1)(50) + (5 – 1)(50) = 600 SST =

2

j

j

2

2

2

2 j

j

ANOVA Table Source

Degrees of Freedom Sum of Squares

Treatments

k 1  2

SST = 250

Error

n  k  12

SSE = 600

Mean Squares SST 250 = 125  k 1 2 SSE 600  = 50 n  k 12

F MST 125  = 2.50 MSE 50

__________________________________________ SS(Total) = 850 Total n  k  14 bx 

10 (10 )  10 (15)  10 (20 ) = 15 555

 n (x  x)  10(10 – 15) + 10(15 – 15) + 10(20 – 15) = 500 SSE =  (n  1)s  (10 –1)(50) + (10 – 1)(50) + (10 – 1)(50) = 1350 SST =

2

j

j

2

2

2

2 j

j

ANOVA Table Source

Degrees of Freedom Sum of Squares

Treatments

k 1  2

SST = 500

Error

n  k  27

SSE = 1350

Mean Squares

F

SST 500  = 250 k 1 2 SSE 1350   50 nk 27

MST 250  = 5.00 MSE 50

__________________________________________ n  k  29 Total SS(Total) = 1850 c The F statistic increases.

14.2 a x 

4(20 )  4(22 )  4(25) = 22.33 444

 n (x  x)  4(20 – 22.33) + 4(22 – 22.33) + 4(25 – 22.33) = 50.67 SSE =  (n  1)s  (4 –1)(10) + (4 – 1)(10) + (4 – 1)(10) = 90 SST =

2

j

j

j

2

2

2 j

349

2

.


ANOVA Table Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Treatments

k 1  2

SST = 50.67

SST 50 .67   25 .33 k 1 2

MST 25 .33  = 2.53 MSE 10

Error

nk 9

SSE = 90

SSE 90  = 10 n  k1 9

b SSE =

 (n  1)s  (4 –1)(25) + (4 – 1)(25) + (4 – 1)(25) = 225 2 j

j

ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 25 .33  = 1.01 MSE 25 .0

Treatments

k 1  2

SST = 50.67

SST 50 .67   25 .33 k 1 2

Error

nk 9

SSE = 225

SSE 225   25 .0 nk 9

c The F statistic decreases.

14.3 a x 

10 (30 )  14 (35)  11(33)  18(40 )  35 .34 10  14  11  18

 n (x  x)  10(30– 35.34) +14(35 – 35.34) + 11(33 –35.34) + 18(40 –35.34) = 737.9 SSE =  (n  1)s  (10 –1)(10) + (14 – 1)(10) + (11 – 1)(10) + (18−1)(10) = 490.0 2

SST =

j

j

2

2

2

2

2 j

j

ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 246 .0   24 .60 MSE 10 .00

Treatments

k 1  3

SST = 737.9

SST 737 .9   246 .0 k 1 3

Error

n  k  49

SSE = 490.0

SSE 490 .0   10 .00 nk 49

b x

10 (130 )  14 (135 )  11(133 )  18(140 )  135 .34 10  14  11  18

SST =

 n (x  x) j

2

j

= 10(130– 135.34) 2 +14(135 – 135.34) 2 + 11(133 –135.34) 2 + 18(140 –135.34) 2 = 737.9 SSE =

 (n  1)s  (10 –1)(10) + (14 – 1)(10) + (11 – 1)(10) + (18−1)(10) = 610.0 j

2 j

350


ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 246 .0   24 .60 MSE 10 .00

Treatments

k 1  3

SST = 737.9

SST 737 .9   246 .0 k 1 3

Error

n  k  49

SSE = 490.0

SSE 490 .0   10 .00 nk 49

c No change 14.4

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F,k 1,n  k  F.05, 2,9  4.26 Finance

Marketing

Management

Mean

2.25

3.25

5.75

Variance

2.25

2.92

2.92

Grand mean = 3.75

 n (x  x) = 4(2.25–3.75) + 4(3.25 – 3.75) + 4(5.75 –3.75) = 26.00 SSE =  (n  1)s  (4 –1)(2.25) + (4 – 1)(2.92) + (4 – 1)(2.92) = 24.25 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 13 .00   4.82 MSE 2.69

Treatments

k 1  2

SST = 26.00

SST 26 .00   13 .00 k 1 2

Error

nk 9

SSE = 24.25

SSE 24 .25   2.69 nk 9

.

F = 4.82, p-value = .0377. There is enough evidence to conclude that there are differences in the number of job offers between the three MBA majors. 14.5

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1,n  k  F.01, 2,15  6.36 Brand 1

Brand 2

Brand 3

Mean

1.33

2.50

2.67

Variance

1.87

2.30

1.47

Grand mean = 2.17

351


 n (x  x) = 6(1.33 – 2.17) + 6(2.50 – 2.17) + 6(2.67 – 2.17) = 6.33 SSE =  (n  1)s  (6 –1)(1.87) + (6 – 1)(2.30) + (6 – 1)(1.47) = 28.17 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 3.17   1.69 MSE 1.88

Treatments

k 1  2

SST = 6.33

SST 6.33   3.17 k 1 2

Error

n  k  15

SSE = 28.17

SSE 28 .17   1.88 nk 15

.

F = 1.69, p-value = .2185. There is not enough evidence to conclude that differences exist between the three brands. 14.6

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1,n  k  F.05, 2,12  3.89 BA

BSc

BBA

Mean

3.94

4.78

5.76

Variance

1.26

.92

1.00

Grand mean = 4.83

 n (x  x) = 5(3.94 – 4.83) + 5(4.78 – 4.83) + 5(5.76 – 4.83) = 8.30 SSE =  (n  1)s  (5 –1)(1.26) + (5 – 1)(.92) + (5 – 1)(1.00) = 12.73 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 4.15   3.91 MSE 1.06

Treatments

k 1  2

SST = 8.30

SST 8.30   4.15 k 1 2

Error

n  k  12

SSE = 12.73

SSE 12 .73   1.06 nk 12

.

F = 3.91, p-value = .0493. There is enough evidence to conclude that students in different degree program differ in their summer earnings. 14.7

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

. Rejection region: F  F,k 1,n  k  F.025, 2,12  5.10

Mean

Professors

Administrators

Students

6.4

10.4

10.8 352


Variance

48.3

16.3

37.7

Grand mean = 9.2

 n (x  x) = 5(6.4 – 9.2) + 5(10.4 – 9.2) + 5(10.8 – 9.2) = 59.2 SSE =  (n  1)s  (5 –1)(48.3) + (5 – 1)(16.3) + (5 – 1)(37.7) = 409.2 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 29 .6   .87 MSE 34 .1

Treatments

k 1  2

SST = 59.2

SST 59 .2   29 .6 k 1 2

Error

n  k  12

SSE = 409.2

SSE 409 .2   34 .1 nk 12

.

F = .87, p-value = .4445. There is not enough evidence to conclude that the differing university communities differ in the amount of spam they receive in their emails. 14.8

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1,n  k  F.05,3,8  4.07 IBM

Dell

HP

Other

Mean

13.33

11.00

9.67

17.00

Variance

12.33

79.00

22.33

39.00

Grand mean = 12.75

 n (x  x) = 3(13.33 – 12.75) + 3(11.00 – 12.75) + 3(9.67 – 12.75) + 3(17.00 – 12.75) = 92.92 SSE =  (n  1)s  (3 –1)(12.33) + (3 – 1)(79.00) + (3 – 1)(22.33) + (3 – 1)(39.00) = 305.33 2

SST =

j

2

j

2

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 30 .97   .81 MSE 38 .17

Treatments

k 1  3

SST = 92.92

SST 92 .92   30 .97 k 1 3

Error

nk 8

SSE = 305.33

SSE 305 .33   38 .17 nk 8

.

F = .81, p-value = .5224. There is not enough evidence to conclude that there are differences in age between the computer brands. 14.9a

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

Rejection region: F  F,k 1,n  k  F.05,3,77  2.76 Grand mean = 65.30 353


SST =

 n (x  x) j

2

j

= 20(68.83 – 65.30) 2 + 26(65.08 – 65.30) 2 + 16(62.01 – 65.30) 2 + 19(64.64 – 65.30) 2 = 430.95 SSE =

 (n  1)s  (20 –1)(52.28) + (26 – 1)(37.38) + (16 – 1)(63.46) + (19 – 1)(56.88) = 3903.57 2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

Treatments

k 1  3

SST = 430.95

SST 430 .95   143 .65 k 1 3

Error

n  k  77

SSE = 3903.57

SSE 3903 .57   50 .70 nk 77

F

.

MST 143 .65   2.83 MSE 50 .70

F = 2.83, p-value = .0437. There is enough evidence to infer that there are differences in grading standards between the four high schools. b The grades of the students at each high school are required to be normally distributed with the same variance. c The histograms are approximately bell-shaped and the sample variances are similar. 14.10a H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ Rejection region: F  F,k 1, n  k  F.05,3,116  2.68 Grand mean = 101.0 SST =

 n (x  x) j

2

j

= 30(90.17 – 101.0) 2 + 30(95.77 – 101.0) 2 + 30(106.8 – 101.0) 2 + 30(111.17 – 101.0) 2 = 8,464 SSE =

 (n  1)s j

2 j

= (30 –1)(991.52) + (30 – 1)(900.87) + (30 – 1)(928.70) + (30 – 1)(1,023.04) = 111,480 ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 2,821   2.94 MSE 961 .0

Treatments

k 1  3

SST = 8,464

SST 8,464   2,821 k 1 3

Error

n  k  116

SSE = 111,480

SSE 111,480   961 .0 nk 116

.

F = 2.94, p-value = .0363. There is enough evidence to infer that there are differences between the completion times of the four income tax forms. b The times for each form must be normally distributed with the same variance. c The histograms are approximately bell-shaped and the sample variances are similar.

354


14.11

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

Rejection region: F  F,k 1,n  k  F.05,3, 275  2.61 Grand mean = 218.0 SST =

 n (x  x) = 41(196.83 – 218.0) + 73(207.78 – 218.0) + 86(223.38 – 218.0) 2

j

2

j

2

2

+ 79(232.67 – 218.0) 2 = 45,496 SSE =

 (n  1)s = (41 –1)(914.05) + (73 – 1)(861.12) + (86 – 1)(1,195.44) 2 j

j

+ (79 – 1)(1,079.81) = 284,400 ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

Treatments

k 1  3

SST = 45,496

SST 45,496   15,165 k 1 3

Error

n  k  275

SSE = 284,400

SSE 284 ,400   1,034 nk 275

F

.

MST 15,165   14 .66 MSE 1,034

F = 14.66, p-value = 0. There is enough evidence to infer that there are differences in test scores between children whose parents have different educational levels. 14.12

H0:µ 1= µ 2 = µ 3 = µ 4= µ 5 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.01, 4,120  3.48 Grand mean = 173.3 SST =

 n (x  x) = 25(164.6 – 173.3) + 25(185.6 – 173.3) + 25(154.8 – 173.3) 2

j

2

j

2

2

+ 25(182.6 – 173.3) 2 + 25(178.9 – 173.3) 2 = 17,251 SSE =

 (n  1)s = (25 –1)(1,164) + (25 – 1)(1,720) + (25 – 1)(1,114) + (25 – 1)(1,658) j

2 j

+ (25 – 1)(841.8) = 155,941 ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

.

Treatments

k 1  4

SST = 17,251

SST 17 ,251 MST 4,312 .6   4,312 .6   3.32 k 1 4 MSE 1,299 .5

Error

n  k  120

SSE = 155,941

SSE 155 ,941   1,299 .5 nk 120

F = 3.32, p-value = .0129. There is not enough evidence to allow the manufacturer to conclude that differences exist between the five lacquers. b The times until first sign of corrosion for each lacquer must be normally distributed with a common variance. c The histograms are approximately bell-shaped with similar sample variances. 355


14.13

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

Rejection region: F  F,k 1,n  k  F.05,3, 297  2.61 Grand mean = 16.11 SST =

 n (x  x) = 39(22.21 – 16.11) + 114(18.46 – 16.11) + 81(15.49 – 16.11) 2

j

2

j

2

2

+ 67(9.31 – 16.11) 2 = 5,202 SSE =

 (n  1)s = (39 –1)(121.64) + (114 – 1)(90.39) + (81 – 1)(85.25) + (67 – 1)(65.40) = 25,973 2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

Treatments

k 1  3

SST = 5,202

SST 5,202   1,734 k 1 3

Error

n  k  297

SSE = 25,973

SSE 25,973   87 .45 nk 297

F

.

MST 1,734   19 .83 MSE 87 .45

F = 19.83, p-value = 0. There is enough evidence to infer that there are differences exist between the four groups. 14.14

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.05, 2,57  3.15 Grand mean = 562.6

 n (x  x) = 20(551.50 – 562.6) + 20(576.75 – 562.6) + 20(559.45 – 562.6) = 6,667 SSE =  (n  1)s = (20 –1)(2,741.95) + (20 – 1)(2,641.14) + (20 – 1)(3,129.31) = 161,736 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

Treatments

k 1  2

SST = 6,667

SST 6,667   3,334 k 1 2

Error

n  k  57

SSE = 161,736

SSE 161,736   2,837 nk 57

F

.

MST 3,334   1.17 MSE 2,837

F = 1.17, p-value = .3162. There is not enough evidence of a difference between fertilizers in terms of crop yields. 14.15

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.05, 2, 297  3.07 Grand mean = 5.48 356


 n (x  x) = 100(5.81 – 5.48) + 100(5.30 – 5.48) + 100(5.33 – 5.48) = 16.38 SSE =  (n  1)s = (100 –1)(6.22) + (100 – 1)(4.05) + (100 – 1)(3.90) = 1,402.5 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 8.19   1.73 MSE 4.72

Treatments

k 1  2

SST = 16.38

SST 16 .38   8.19 k 1 2

Error

n  k  297

SSE = 1,402.5

SSE 1,402 .5   4.72 nk 297

.

F = 1.73, p-value = .1783. There is not enough evidence of a difference between the three departments. 14.16

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.05,3,116  2.68 Grand mean = 77.39 SST =

 n (x  x) = 30(74.10 – 77.39) + 30(75.67 – 77.39) + 30(78.50 – 77.39) 2

j

2

j

2

2

+ 30(81.30 – 77.39) 2 = 909.42 SSE =

 (n  1)s = (30 –1)(249.96) + (30 – 1)(184.23) + (30 – 1)(233.36) + (30 – 1)(242.91) = 26,403 2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 303 .1   1.33 MSE 227 .6

Treatments

k 1  3

SST = 909.4

SST 909 .4   303 .1 k 1 3

Error

n  k  116

SSE = 26,403

SSE 26 ,403   227 .6 nk 116

.

F = 1.33, p-value = .2675. There is not enough evidence of a difference between the four groups of companies. 14.17

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

a Leaf size: Rejection region: F  F,k 1,n  k  F.05, 2,147  3.06 Grand mean = 21.49

 n (x  x) = 50(24.97 – 21.49) + 50(21.65 – 21.49) + 50(17.84 – 21.49) = 1,270 SSE =  (n  1)s = (50 –1)(48.23) + (50 – 1)(54.54) + (50 – 1)(33.85) = 6,695

SST =

2

j

j

j

2

2

2

2 j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares 357

F

.


Treatments

k 1  2

SST = 1,270

SST 1,270   635 .0 k 1 2

Error

n  k  147

SSE = 6,695

SSE 6,695   45 .54 nk 147

MST 635 .0   13 .95 MSE 45 .54

F = 13.95, p-value = 0. There is sufficient evidence to conclude that the leaf sizes differ between the 3 groups. b Nicotine: Rejection region: F  F,k 1,n  k  F.05, 2,147  3.06 Grand mean = 13.00

 n (x  x) = 50(15.52 – 13.00) + 50(13.39 – 13.00) + 50(10.08 – 13.00) = 753.2 SSE =  (n  1)s = (50 –1)(3.72) + (50 – 1)(3.59) + (50 – 1)(3.83) = 545.54 2

SST =

j

2

j

2

2

2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

k 1  2

SST = 753.2

MST 376 .6 SST 753 .2   101 .47   376 .6 MSE 3.71 k 1 2

Error

n  k  147

SSE = 545.54

SSE 545 .54   3.71 nk 147

.

F = 101.47, p-value = 0. There is sufficient evidence to infer that the amounts of nicotine differ between the 3 groups. 14.18

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

a Ages: Rejection region: F  F,k 1, n  k  F.05,3, 291  2.61 Grand mean = 36.23 SST =

 n (x  x) = 63(31.30 – 36.23) + 81(34.42 – 36.23) + 40(37.38 – 36.23) 2

j

2

j

2

2

+ 111(39.93 – 36.23) 2 = 3,366 SSE =

 (n  1)s = (63 –1)(28.34) + (81 – 1)(23.20) + (40 – 1)(31.16) + (111 – 1)(72.03) = 12,752 j

2 j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 1,122   25 .60 MSE 43 .82

Treatments

k 1  3

SST =3,366

SST 3,366   1,122 k 1 3

Error

n  k  291

SSE = 12,752

SSE 12 ,752   43 .82 nk 291

.

F = 25.60, p-value = 0. There is sufficient evidence to infer that the ages of the four groups of cereal buyers differ. b Incomes: Rejection region: F  F,k 1, n  k  F.05,3, 291  2.61 Grand mean = 39.97

358


SST =

 n (x  x) = 63(37.22 – 39.97) + 81(38.91 – 39.97) + 40(41.48 – 39.97) 2

j

2

j

2

2

+ 111(41.75 – 39.97 ) 2 = 1,008 SSE =

 (n  1)s = (63 –1)(39.82) + (81 – 1)(40.85) + (40 – 1)(61.38) + (111 – 1)(46.59) = 13,256 2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 336 .0   7.37 MSE 45 .55

Treatments

k 1  3

SST =1,008

SST 1,008   336 .0 k 1 3

Error

n  k  291

SSE = 13,256

SSE 13,256   45 .55 nk 291

.

F = 7.37, p-value = .0001. There is sufficient evidence to conclude that incomes differ between the four groups of cereal buyers. c Education: Rejection region: F  F,k 1, n  k  F.05,3, 291  2.61 Grand mean = 11.98 SST =

 n (x  x) = 63(11.75 – 11.98) + 81(12.41 – 11.98) + 40(11.73 – 11.98) 2

j

2

j

2

2

+ 111(11.89 – 11.98) 2 = 21.71 SSE =

 (n  1)s = (63 –1)(3.93) + (81 – 1)(3.39) + (40 – 1)(4.26) + (111 – 1)(4.30) = 1,154 2 j

j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 7.24   1.82 MSE 3.97

Treatments

k 1  3

SST =21.71

SST 21 .71   7.24 k 1 3

Error

n  k  291

SSE = 1,154

SSE 1,154   3.97 nk 291

.

F = 1.82, p-value = .1428. There is not enough evidence to infer that education differs between the four groups of cereal buyers. d Using the F-tests and the descriptive statistics we see that the mean ages and mean household incomes are in ascending order. For example, Sugar Smacks buyers are younger and earn less than the buyers of the other three cereals. Cheerio purchasers are older and earn the most. 14.19

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.05, 2,57  3.15 Grand mean = 146.1

 n (x  x) = 20(153.60 – 146.1) + 20(151.50 – 146.1) + 20(133.25 – 146.1) = 5,011 SSE =  (n  1)s = (20 –1)(654.25) + (20 – 1)(924.05) + (20 – 1)(626.83) = 41,898 SST =

2

j

j

j

2

2

2 j

359

2


ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 2,506   3.41 MSE 735 .0

Treatments

k 1  2

SST = 5,011

SST 5,011   2,506 k 1 2

Error

n  k  57

SSE = 41,898

SSE 41,898   735 .0 nk 57

.

F = 3.41, p-value = .0400. There is enough evidence to infer that sales will vary according to price. 14.20

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

Rejection region: F  F, k 1, n  k  F.05, 2, 232  3.07 Grand mean = 19.50

 n (x  x) = 61(18.54 – 19.50) + 83(19.34 – 19.50) + 91(20.29 – 19.50) = 114.5 SSE =  (n  1)s = (61 –1)(177.95) + (83 – 1)(171.42) + (91 – 1)(297.50) = 51,508 2

SST =

j

2

j

j

2

2

2 j

ANOVA table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 57 .3   .26 MSE 222 .0

Treatments

k 1  2

SST = 114.5

SST 114 .5   57 .3 k 1 2

Error

n  k  232

SSE = 51,508

SSE 51,508   222 .0 nk 232

F = .26, p-value = .7730. There is not enough evidence of a difference between the three segments. 14.21

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

360

.


F = 36.7, p-value = 0. There is enough evidence to conclude that at least two means differ. 14.22 Reading H0:µ 1= µ 2 = µ 3 H1: At least two means differ

F = 182.43, p-value = 0. There is enough evidence to conclude that at least two means differ. Mathematics H0:µ 1= µ 2 = µ 3 H1: At least two means differ

361


F = 353.45, p-value = 0. There is enough evidence to conclude that at least two means differ. Science H0:µ 1= µ 2 = µ 3 H1: At least two means differ

F = 121.0, p-value = 0. There is enough evidence to conclude that at least two means differ. 14.23

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

362


F = 9.19, p-value = 0. 14.24

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

F = 6.09, p-value = 0. 14.25

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

F = 2.38, p-value = .0201. 14.26

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

F = 56.97, p-value = 0. 14.27

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

363


F = 44.02, p-value = 0. 14.28

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

F = 52.41, p-value = 0. 14.29

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 = µ 8 H1: At least two means differ

F = 62.35, p-value = 0. 14.30

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = 12.84, p-value = 0. 14.31

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

364


F = 4.68, p-value = 0. 14.32

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = .87, p-value = .5189. 14.33

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = 7.22, p-value = 0. 14.34

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = 41.85, p-value = 0. 14.35

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

365


F = 27.61, p-value = 0. 14.36

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = 41.91, p-value = 0. 14.37

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ

F = 40.96, p-value = 0. 14.38

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ

F = 83.86, p-value = 0. 14.39

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

366


F = 23.72, p-value = 0. 14.40

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ

F = 2.50, p-value = .0406. 14.41

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 8.57, p-value = 0. 14.42

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 18.23, p-value = 0. 14.43

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

367


F = 8.67, p-value = 0. 14.44

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 8.29, p-value = 0. 14.45

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ

F = 7.07, p-value = 0. 14.46

H0:µ 1= µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ

F = .745, p-value = .5610. 14.47

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

368


F = 32.45, p-value = 0. 14.48

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 39.25, p-value = 0. 14.49

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 20.24, p-value = 0. 14.50

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 2.52, p-value = .0565. 14.51

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 110.0, p-value = 0.

369


14.52

H0:µ 1= µ 2 = µ 3 = µ 4 H1: At least two means differ

F = 152.9, p-value = 0. 14.53

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

F = 12.25, p-value = 0. 14.54

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

F = 14.27, p-value = 0. 14.55

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

F = 3.07, p-value = .0466. 14.56

H0:µ 1= µ 2 = µ 3 H1: At least two means differ

370


F = 9.73, p-value = 0. 14.57 a  = .05: t  / 2, n  k  t .025, 27  2.052  1 1 1 1  = 2.052 700    = 24.28 LSD = t  / 2,n  k MSE    ni n j   10 10   

Treatment

Means

Difference

i = 1, j = 2

128.7

101.4

27.3

i = 1, j = 3

128.7

133.7

−5.0

i = 2, j = 3

101.4

133.7

−32.3

Conclusion:  2 differs from 1 and  3 . b C = 3(2)/2 = 3,  E = .05,    E / C  .0167: t  / 2,n  k  t .0083, 27  2.552 (from Excel)  1 1 1  1 LSD = t  / 2,n  k MSE   = 2.552 700    = 30.20  ni n j   10 10   

Treatment

Means

Difference

i = 1, j = 2

128.7

101.4

27.3

i = 1, j = 3

128.7

133.7

−5.0

i = 2, j = 3

101.4

133.7

−32.3

Conclusion:  2 and  3 differ. c q  (k, )  q .05 (3,27 )  3.53   q  (k , ) Treatment

Means

MSE 700 = 3.53 = 29.53 ng 10

Difference

i = 1, j = 2

128.7

101.4

27.3

i = 1, j = 3

128.7

133.7

−5.0

i = 2, j = 3

101.4

133.7

−32.3

Conclusion:  2 and  3 differ. 14.58 a  = .05: t  / 2, n  k  t.025, 20  2.086  1 1  1 1 LSD = t  / 2,n  k MSE   = 2.086 125    = 14.75  ni n j  5 5   Treatment Means Difference

i = 1, j = 2

227

205

22

i = 1, j = 3

227

219

8

i = 1, j = 4

227

248

−21

i = 1, j = 5

227

202

25

371


i = 2, j = 3

205

219

−14

i = 2, j = 4

205

248

−43

i = 2, j = 5

205

202

3

i = 3, j = 4

219

248

−29

i = 3, j = 5

219

202

17

i = 4, j = 5

248

202

46

Conclusion: The following pairs of means differ. 1 and  2 , 1 and  4 , 1 and  5 ,  2 and  4 ,  3 and  4 ,  3 and  5 , and  4 and  5 . b. C = 5(4)/2 = 10,  E = .05,    E / C  .005 t  / 2,n  k  t .0025, 20  3.153 (from Excel)  1 1  1 1 LSD = t  / 2,n  k MSE   = 3.153 125    = 22.30  ni n j  5 5  

Treatment

Means

Difference

i = 1, j = 2

227

205

22

i = 1, j = 3

227

219

8

i = 1, j = 4

227

248

−21

i = 1, j = 5

227

202

25

i = 2, j = 3

205

219

−14

i = 2, j = 4

205

248

−43

i = 2, j = 5

205

202

3

i = 3, j = 4

219

248

−29

i = 3, j = 5

219

202

17

i = 4, j = 5

248

202

46

Conclusion: The following pairs of means differ. 1 and  5 ,  2 and  4 ,  3 and  4 , and  4 and  5 . c q  (k, )  q .05 (5, 20) = 4.23   q  (k , ) Treatment

Means

MSE 125 = 4.23 = 21.15 ng 5

Difference

i = 1, j = 2

227

205

22

i = 1, j = 3

227

219

8

i = 1, j = 4

227

248

−21

i = 1, j = 5

227

202

25

i = 2, j = 3

205

219

−14

i = 2, j = 4

205

248

−43

i = 2, j = 5

205

202

3

i = 3, j = 4

219

248

−29

i = 3, j = 5

219

202

17

372


i = 4, j = 5

248

202

46

Conclusion: The following pairs of means differ. 1 and  2 , 1 and  5 ,  2 and  4 ,  3 and  4 , and  4 and  5 .

14.59 q  (k, )  q .05 (3, 15) = 3.67   q  (k , ) Treatment

Means

MSE 1.88 = 3.67 = 2.05 ng 6

Difference

i = 1, j = 2

1.33

2.50

−1.17

i = 1, j = 3

1.33

2.67

−1.34

i = 2, j = 3

2.50

2.67

−.17

There are no differences.

 1 1 1 1  = 1.782 1.06    = 1.16 14.60 a. LSD = t  / 2,n  k MSE    ni n j  5 5  

Treatment

Means

Difference

i = 1, j = 2

3.94

4.78

−.84

i = 1, j = 3

3.94

5.76

−1.82

i = 2, j = 3

4.78

5.76

−1.02

Means of BAs and BBAs differ. b. C = 3(2)/2 = 3,  E = .10,    E / C  .0333: t  / 2,n  k  t .0167,12  2.404 (from Excel) 1 1 1 1 LSD = t  / 2, n  k MSE   = 2.404 1.06    = 1.57  ni n j  5 5   Treatment Means Difference

i = 1, j = 2

3.94

4.78

−.84

i = 1, j = 3

3.94

5.76

−1.82

i = 2, j = 3

4.78

5.76

−1.02

Means of BAs and BBAs differ. 14.61 LSD method: C = 4(3)/2 = 6,  E = .05,    E / C  .0083 1 1 LSD = t  / 2, n  k MSE   (LSD must be calculated for each pair of treatments.)  ni n j   

Tukey’s method: n g 

4 k  19 .6 (rounded to 20) = 1 1 1 1 1 1 1      ...  20 26 16 19 n1 n 2 nk

q  (k, )  q .05 (4, 77)  3.74  = 3.74

50 .70 = 5.95 20

373


Treatment

Means

Difference

LSD

i = 1, j = 2

68.83

65.08

3.75

5.74

i = 1, j = 3

68.83

62.01

6.82

6.47

i = 1, j = 4

68.83

64.64

4.19

6.18

i = 2, j = 3

65.08

62.01

3.07

6.13

i = 2, j = 4

65.08

64.64

.44

5.82

i = 3, j = 4

62.01

64.64

−2.63

6.55

a The mean grades from high schools A and C differ. b The mean grades from high schools A and C differ.

14.62 Tukey’s method: q  (k, )  q .05 (4, 116)  3.68  = 3.68

961 .0 = 20.83 30

LSD method with the Bonferroni adjustment: C = 4(3)/2 = 6,  E = .05,    E / C  .0083  1 1   1 1  LSD = t  / 2,n  k MSE   = 2.64 961 .0   = 21.13  ni n j  30 30    

Treatment

Means

Difference

Tukey

LSD

i = 1, j = 2

90.17

95.77

−5.60

20.83

21.13

i = 1, j = 3

90.17

106.8

−16.67

20.83

21.13

i = 1, j = 4

90.17

111.17

−21.00

20.83

21.13

i = 2, j = 3

95.77

106.8

−11.07

20.83

21.13

i = 2, j = 4

95.77

111.17

−15.40

20.83

21.13

i = 3, j = 4

106.8

111.17

−4.33

20.83

21.13

a The means for Forms 1 and 4 differ. b No means differ.

14.63a

H 0 : 1   2   3 =  4 H 1 : At least two means differ.

x

10 (61 .60 )  10 (57 .30 )  10 (61 .80 )  10 (51 .80 )  58 .13 10  10  10  10

 n (x  x) = 10(61.6–58.13) +10(57.3–58.13) + 10(61.8–58.13) + 10(51.8–58.13) = 662.7 SSE =  (n  1)s  (10 –1)(80.49) + (10 – 1)(70.46) + (10 – 1)(22.18) + (10−1)(75.29) = 2,236

SST =

2

j

j

j

2

2

2 j

374

2

2


ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 220 .9   3.56 MSE 62 .11

Treatments

k 1  3

SST = 662.7

SST 662 .7   220 .9 k 1 3

Error

n  k  36

SSE = 2,236

SSE 2,236   62 .11 nk 36

F = 3.56, p-value = .0236. There is enough evidence to infer that differences exist between the flares with respect to burning times. LSD method: C = 4(3)/2 = 6,  E = .05,    E / C  .0083 t  / 2, n  k  t .0042,36  2.794 (from Excel) 1 1 1 1 LSD = t  / 2, n  k MSE   = 2.794 62 .11    9.85  ni n j   10 10   

Tukey’s method: q  (k, )  q .05 (4, 136)  3.79  = 3.79 Treatment

Means

62 .11 = 9.45 10

Difference

i = 1, j = 2

61.6

57.3

4.3

i = 1, j = 3

61.6

61.8

−.2

i = 1, j = 4

61.6

51.8

9.8

i = 2, j = 3

57.3

61.8

−4.5

i = 2, j = 4

57.3

51.8

5.5

i = 3, j = 4

61.8

51.8

10.0

b. The means of flares C and D differ. c The means of flares A and D, and C and D differ. 14.64 a LSD method: C = 5(4)/2 = 10,  E = .05,    E / C  .005 t  / 2,n  k  t .0025,120  2.860 (from Excel) 1 1   1 1 LSD = t  / 2, n  k MSE   = 2.860 1300     29 .17  ni n j  25 25    

b Tukey’s method: q  (k, )  q .05 (5,120) = 3.92  = 3.92 Treatment

Means

Difference

i = 1, j = 2

164.6

185.6

−21.0

i = 1, j = 3

164.6

154.8

9.8

i = 1, j = 4

164.6

182.6

−18.0

i = 1, j = 5

164.6

178.9

−14.3

i = 2, j = 3

185.6

154.8

30.8

i = 2, j = 4

185.6

182.6

3.0

i = 2, j = 5

185.6

178.9

6.7

375

1300 = 28.27 25


i = 3, j = 4

154.8

182.6

−27.8

i = 3, j = 5

154.8

178.9

−24.1

i = 4, j = 5

182.6

178.9

3.7

a The means of lacquers 2 and 3 differ b The means of lacquers 2 and 3 differ.

14.65a

H 0 : 1   2   3 H1 : At least two means differ.

x

30 (53 .17 )  30 (49 .37 )  30 (44 .33)  48 .96 30  30  30

 n (x  x) = 30(53.17–48.96) +30(49.37–48.96) + 30(44.33–48.96) = 1,178 SSE =  (n  1)s  (30 –1)(194.6) + (30 – 1)(152.6) + (30 – 1)(129.9) = 13,836 SST =

2

j

2

j

2

2

2 j

j

ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

MST 589 .0   3.70 MSE 159 .0

Treatments

k 1  2

SST = 1,178

SST 1,178   589 .0 k 1 2

Error

n  k  87

SSE = 13,836

SSE 13,836   159 .0 nk 87

F = 3.70, p-value = .0286. There is enough evidence to infer that speed of promotion varies between the three sizes of engineering firms. b q  (k, )  q .05 (3,87)  3.40  = 3.40

159 .0  7.83 30

Treatment

Difference

Means

i = 1, j = 2

53.17

49.37

3.80

i = 1, j = 3

53.17

44.44

8.83

i = 2, j = 3

49.37

44.33

5.04

The means of small and large firms differ. Answer (v) is correct. 14.66 Tukey’s method: q  (k, )  q .05 (3,57)  3.40  = 3.40

2,838  40 .50 20

LSD method: C = 3(2)/2 = 3,  E = .05,    E / C  .0167 t  / 2, n  k  t .0083,57  2.466 (from Excel) 1 1   1 1 LSD = t  / 2, n  k MSE   = 2.466 2,838     41 .54  ni n j   20 20   

376


Treatment

Means

Difference

i = 1, j = 2

551.5

576.8

−25.3

i = 1, j = 3

551.5

559.5

−8.0

i = 2, j = 3

576.8

559.5

17.3

a There are no differences. b There are no differences. 14.67a XLSTAT

PARTYID 3 differs from all the others. b. Excel Workbook PARTYID

Means

| Difference |

LSD

i = 0, j = 1

13.986 13.670

.316

.507

i = 0, j = 2

13.986 13.846

.140

.532

i = 1, j = 2

13.670 13.846

.176

.536

None of the pairs differ. c. Excel Workbook

377


PARTYID

Means

| Difference |

LSD

i = 4, j = 5

13.928 14.055

.127

.652

i = 4, j = 6

13.928 14.037

.109

.655

i = 5, j = 6

14.055 14.037

.018

.630

None of the three pairs differ. 14.68

PARTYID

Means

| Difference |

LSD

i = 2, j = 3

45,501 34,423

11,078

8599

i = 2, j = 4

45,501 49,374

3873

9912

i= 3, j = 4

34,423 49,374

14,951

8973

PARTYID 3 differs from 2 and 4. 14.69a. XLSTAT

PARTYID 4 differs from 0 and 3 only. At least two means differ. b. The analysis of variance F test produced the same result; at least two means differ. 14.70a. XLSTAT

378


b. Excel Workbook PARTYID

Means

| Difference |

LSD

i = 2, j = 3

3.196

3.433

.237

.377

i = 2, j = 4

3.196

4.768

1.572

.450

i= 3, j = 4

3.433

4.768

1.355

.417

PARTYID 4 differs from 2 and 3. 4.71a. XLSTAT

379


b. Excel Workbook PARTYID

Means

| Difference |

LSD

i = 2, j = 3

2.793

2.696

.097

.230

i = 2, j = 4

2.793

3.549

.756

.276

i= 3, j = 4

2.696

3.549

.853

.255

PARTYID 4 differs from 2 and 3 4.72a XLSTAT

F = 14.23, p-value = 0. b.

380


Whites differ from Blacks and Others. 14.73 a XLSTAT

F = .772, p-value = .4624. b.

There are no differences. c. Tukey and ANOVA produced the same result. 14.74a XLSTAT

b.

Independents differ from Democrats and Republicans. 14.75 Excel Workbook PARTYID

Means

| Difference |

LSD

i = 1, j = 3

2.265

2.490

.225

.229

i = 1, j = 5

2.265

3.259

.994

.264

i= 3, j = 5

2.490

3.259

.769

.252

PARTYID 5 differs from 1 and 3.

381


14.76 a. Bonferroni with C = 3. b. Excel Workbook POLVIEWS

Means

| Difference |

LSD

i = 1, j = 2

14.681 14.635

.046

.842

i = 1, j = 3

14.681 14.297

.384

.858

i= 2, j = 3

14.635 14.297

.338

.601

There are no differences. 14.77 XLSTAT

The mean of 4 differs from the means of 5, 6, and 7. 14.78 a. XLSTAT

382


F = 33.00, p-value = 0. b.

All three pairs differ. 14.79 XLSTAT

Blacks differ from Whites and others. 14.80 Excel Workbook POLVIEWS

Means

| Difference |

LSD

i = 1, j = 2

2.094

2.786

.692

.662

i = 1, j = 3

2.094

3.288

1.194

.671

i= 2, j = 3

2.786

3.288

.502

.485

All three differ. 14.81 XLSTAT

383


The mean of 4 differs from the means of 1, 2, and 3. 14.82 XLSTAT

All pairs of means differ except the means of 1 and 2. 14.83 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

100

50.00

24.04

Blocks

6

50

8.33

4.00

Error

12

25

2.08

Total

20

175

.

a Rejection region: F  F,k 1, n  k  b 1  F.05, 2,12 = 3.89 Conclusion: F = 24.04, p-value = .0001. There is enough evidence to conclude that the treatment means differ. b Rejection region: F  F , b 1, n  k  b 1  F.05,6,12 = 3.00 F = 4.00, p-value = .0197. There is enough evidence to conclude that the block means differ. 14.84 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

4

1,500

375.0

16.50

Blocks

11

1,000

90.91

4.00

Error

44

1,000

22.73

Total

59

3,500

a Rejection region: F  F,k 1, n  k  b 1  F.01, 4, 44  3.83 F = 16.50, p-value = 0. There is enough evidence to conclude that the treatment means differ. b Rejection region: F  F,b 1,n k b 1  F.01,11, 44  2.80 Conclusion: F = 4.00, p-value = .0005. There is enough evidence to conclude that the block means differ.

384


14.85 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

3

275

91.67

7.99

Blocks

9

625

69.44

6.05

Error

27

310

11.48

Total

39

1,210

a Rejection region: F  F,k 1, n  k  b 1  F.01,3, 27 = 4.60 Conclusion: F = 7.99, p-value = .0006. There is enough evidence to conclude that the treatment means differ. b Rejection region: F  F , b 1, n  k  b 1  F.01,9, 27 = 3.15 Conclusion: F = 6.05, p-value = .0001. There is enough evidence to conclude that the block means differ. 14.86 Rejection region: F  F,k 1, n  k  b 1  F.05, 2,14 = 3.74 a. ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

1,500

750.0

7.00

Blocks

7

500

71.43

.67

Error

14

1,500

107.1

Total

23

3,500

Conclusion: F = 7.00, p-value = .0078. There is enough evidence to conclude that the treatment means differ. b. ANOVA Table Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Treatments

2

1,500

750.0

10.50

Blocks

7

1,000

142.86

2.00

Error

14

1,000

71.43

Total

23

3,500

Conclusion: F = 10.50, p-value = .0016. There is enough evidence to conclude that the treatment means differ c. ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

1,500

750

21.00

Blocks

7

1,500

214.3

6.00

Error

14

500

35.71

Total

23

3,500

Conclusion: F = 21.00, p-value = .0001. There is enough evidence to conclude that the treatment means differ d. The test statistic increases.

385


14.87 a. k = 3, b = 5, Grand mean = 10.4 b

k

  (x  x)  (7  10.4)  (10  10.4)  (12  10.4)  (9  10.4)  (12  10.4) 2

2

SS(Total) =

2

2

2

2

ij

j1

i 1

 (12 10.4) 2  (8 10.4) 2  (16 10.4) 2  (13 10.4) 2  (10 10.4) 2  (8 10.4) 2  (9 10.4) 2  (13 10.4) 2  (6 10.4) 2  (1110.4) 2 = 99.6 k

SST =

 b(x[T]  x)  5[(10  10.4)  (11.8  10.4)  (9.4  10.4) ]  15.6 2

2

2

2

j

j1 b

SSB =

 k(x[B]  x)  3[(9  10.4)  (9  10.4)  (13.7  10.4)  (9.3  10.4)  (11  10.4) ]  48.3 2

2

2

2

2

2

i

i 1

SSE = SS(Total) – SST – SSB = 99.6− 15.6 – 48.3 = 35.7 k

b SS(Total) =

b

  (x  x)  (7  10.4)  (10  10.4)  (12  10.4)  (9  10.4)  (12  10.4) 2

2

2

2

2

2

ij

j1

i 1

 (12 10.4) 2  (8 10.4) 2  (16 10.4) 2  (13 10.4) 2  (10 10.4) 2

 (8 10.4) 2  (9 10.4) 2  (13 10.4) 2  (6 10.4) 2  (1110.4) 2 = 99.6 k

SST =

 n (x  x)  5(10 10.4)  5(11.8 10.4)  5(9.4  10.4)  15.6 2

2

j

2

2

j

j1

SSE = SS(Total) – SST = 99.6 – 15.6 = 84.0 c The variation between all the data is the same for both designs. d The variation between treatments is the same for both designs. e Because the randomized block design divides the sum of squares for error in the one-way analysis of variance into two parts. 14.88 a k = 4, b = 3, Grand mean = 5.6 k

SS(Total) =

b

  (x  x)  (6  5.6)  (8  5.6)  (7  5.6)  (5  5.6)  (5  5.6)  (6  5.6) 2

2

2

2

2

2

ij

j1

i 1

 (4  5.6) 2  (5  5.6) 2  (5  5.6) 2  (4  5.6) 2  (6  5.6) 2  (6  5.6) 2 = 14.9 k

SST =

 b(x[T]  x)  3[(7  5.6)  (5.3  5.6)  (4.7  5.6)  (5.3  5.6) ]  8.9 2

2

2

2

j

j1 b

SSB =

 k(x[B]  x)  4[(4.8  5.6)  (6  5.6)  (6  5.6)  4.2 2

2

2

2

i

i 1

SSE = SS(Total) – SST – SSB = 14.9− 8.9 – 4.2 = 1.8 386

2

2


b

k

b SS(Total) =

  (x  x)  (6  5.6)  (8  5.6)  (7  5.6)  (5  5.6)  (5  5.6)  (6  5.6) 2

2

2

2

2

2

2

ij

j1

i 1

 (4  5.6) 2  (5  5.6) 2  (5  5.6) 2  (4  5.6) 2  (6  5.6) 2  (6  5.6) 2 = 14.9 k

SST =

 b(x[T]  x)  3[(7  5.6)  (5.3  5.6)  (4.7  5.6)  (5.3  5.6) ]  8.9 2

2

2

2

2

j

j1

SSE = SS(Total) – SST = 14.9 – 8.9 = 6.0 c The variation between all the data is the same for both designs. d The variation between treatments is the same for both designs. e Because the randomized block design divides the sum of squares for error in the one-way analysis of variance into two parts.

H0 : 1  2   3

14.89

H 1 : At least two means differ. Rejection region: F  F,k 1, n  k  b 1  F.05, 2,6 = 5.14 k = 3, b = 4, Grand mean = 2.38 b

k

SS(Total) =

  (x  x)  (1.4  2.38)  (3.1  2.38)  (2.8  2.38)  (3.4  2.38) 2

2

2

2

2

ij

j1

i 1

 (1.5  2.38) 2  (2.6  2.38) 2  (2.1  2.38) 2  (3.6  2.38) 2

 (1.3  2.38) 2  (2.4  2.38) 2  (1.5  2.38) 2  (2.9  2.38) 2  7.30 k

SST =

 b(x[T]  x)  4[(2.68  2.38)  (2.45  2.38)  (2.03  2.38) ]  .87 2

2

2

2

j

j1 b

SSB =

 k(x[B]  x)  3[(1.4  2.38)  (2.7  2.38)  (2.1  2.38)  (3.3  2.38) ]  5.91 2

2

2

2

2

i

i 1

SSE = SS(Total) – SST – SSB = 7.30 − .87 – 5.91 = .52 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

.87

.44

5.06

Blocks

3

5.91

1.97

22.64

Error

6

.52

.087

Total

11

7.30

F = 5.08, p-value = .0512. There is not enough evidence to conclude that there are differences between the subjects being measured.

387


H 0 : 1   2   3   4

14.90

H1 : At least two means differ. Rejection region: F  F,k 1, n  k  b 1  F.01,3,12 = 5.95 k = 4, b = 5, Grand mean = 8.3 k

SS(Total) =

b

  (x  x)

2

ij

j1

i 1

 (5  8.3)2  (4  8.3)2  (6  8.3)2  (7  8.3)2  (9  8.3)2

 (2  8.3) 2  (7  8.3) 2  (12  8.3) 2  (11 8.3) 2  (8  8.3) 2  (6  8.3) 2  (8  8.3) 2  (9  8.3) 2  (16  8.3) 2  (15  8.3) 2  (8  8.3) 2  (10  8.3) 2  (2  8.3) 2  (7  8.3) 2  (14  8.3) 2 = 286.2 k

SST =

 b(x[T]  x)  5[(6.2  8.3)  (8.0  8.3)  (10.8  8.3)  (8.2  8.3) ]  53.8 2

2

2

2

2

j

j1

SSB 

b

 k(x[B]  x) i

2

 4[(5.25  8.3)2  (7.25  8.3)2  (7.25  8.3)2  (10.25  8.3)2  (11.5  8.3)2 ]  102.2

i 1

SSE = SS(Total) – SST – SSB = 286.2 – 53.8 – 102.2 = 130.2 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

3

53.8

17.93

1.65

Blocks

4

102.2

25.55

2.35

Error

12

130.2

10.85

Total

19

286.2

F = 1.65, p-value = .2296. There is not enough evidence to conclude there are differences between the four diets. 14.91 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

204.2

102.11

4.54

Blocks

11

1150.2

104.57

4.65

Error

22

495.1

22.51

a

H 0 : 1   2   3 H1 : At least two means differ.

F = 4.54, p-value = .0224. There is enough evidence to conclude that there are differences between the three couriers. 388


b

H 0 : 1   2  …  12 H1 : At least two means differ.

F = 4.65, p-value = .0011. The block means differ; the practitioner used the correct design. 14.92 ANOVA Table Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Treatments

2

7,131

3,566

123.36

Blocks

19

177,465

9,340

323.16

Error

38

1,098

28.90

a

H 0 : 1   2   3 H1 : At least two means differ.

Rejection region: F  F,k 1, n  k  b 1  F.05, 2,38  3.23 F = 123.36, p-value = 0. There is sufficient evidence to conclude that the three fertilizers differ with respect to crop yield. b F = 323.16, p-value = 0. There is sufficient evidence to indicate that there are differences between the plots. 14.93 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

2

10.26

5.13

.86

Blocks

19

3,020

159.0

26.64

Error

38

226.7

5.97

a

H 0 : 1   2   3 H1 : At least two means differ.

Rejection region: F  F,k 1, n  k  b 1  F.05, 2,38  3.23 F = .86, p-value = .4313. There is not enough evidence to conclude that there are differences in sales ability between the holders of the three degrees. b

H 0 : 1   2  …   20 H1 : At least two means differ.

F = 26.64, p-value = 0. There is sufficient evidence to indicate that there are differences between the blocks of students. The independent samples design would not be recommended. c The commissions for each type of degree are required to be normally distributed with the same variance. d The histograms are bell shaped and the sample variances are similar.

389


14.94 ANOVA Table Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Treatments

3

4,206

1,402

21.16

Blocks

29

126,843

4,374

66.02

Error

87

5,764

66.25

a

H 0 : 1   2   3 =  4 H1 : At least two means differ.

Rejection region: F  F,k 1, n  k  b 1  F.01,3,87  4.01 F = 21.16, p-value = 0. There is sufficient evidence to conclude that differences in completion times exist between the four forms. b

H 0 : 1   2  …   30 H1 : At least two means differ.

F = 66.02, p-value = 0. There is sufficient evidence to indicate that there are differences between the taxpayers, which tells us that this experimental design is recommended.

14.95

H 0 : 1   2   3 =  4 =  5 =  6 =  7 H1 : At least two means differ.

ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

6

28,674

4,779

11.91

Blocks

199

209,835

1,054

2.63

Error

1194

479,125

401.3

F = 11.91, p-value = 0. There is enough evidence to conclude that there are differences in time spent listening to music between the days of the week 14.96

ANOVA Table

Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Treatments

4

1,406.4

351.6

10.72

Blocks

35

7,309.7

208.9

6.36

Error

140

4,593.9

32.81

a

H 0 : 1   2   3 =  4 =  5 H1 : At least two means differ.

F = 10.72, p-value = 0. There is enough evidence to infer differences between medical specialties. b

H 0 : 1   2  …   36 H1 : At least two means differ.

390


F = 6.36, p-value = 0. There is sufficient evidence to indicate that there are differences between the physicians’ ages, which tells us that this experimental design is recommended.

14.97

H 0 : 1   2   3 =  4 H 1 : At least two means differ.

ANOVA Table Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Treatments

3

563.82

187.9

15.06

Blocks

20

1,327.33

66.37

5.32

Error

60

748.70

12.48

F = 15.06, p-value = 0. There is enough evidence to infer differences in grading standards between the four high schools.

14.98

H 0 : 1   2   3 H 1 : At least two means differ.

F = 476.7, p-value = 0.

14.99

H 0 : 1   2   3 H 1 : At least two means differ.

F = 355.0, p-value = 0. 14.100 ANOVA Table

391


Source

Degrees of Freedom

Sum of Squares

Mean Squares

F

Factor A

2

1,560

780

5.86

Factor B

3

2,880

960

7.18

Interaction

6

7,605

1268

9.53

Error

228

30,405

133

Total

239

42,450

Test for interaction: Rejection region: F  F,(a 1)( b 1),n ab  F.01,6, 228  2.80 F = 9.53. There is enough evidence to conclude that factors A and B interact. The F-tests in Parts b and c are irrelevant. 14.101 ANOVA Table Source

Degrees of Freedom Sum of Squares

Mean Squares

F

Factor A

3

203

67.67

.72

Factor B

2

859

429.5

4.60

Interaction

6

513

85.5

.92

Error

84

7845

93.39

Total

95

9420

a Rejection region: F  F,(a 1)( b 1),n ab  F.05,6,84  2.25 F = .92. There is not enough evidence to conclude that factors A and B interact. b Rejection region: F  F,a 1,n ab  F.05,3,84  2.76 F = .72. There is not enough evidence to conclude that differences exist between the levels of factor A. c Rejection region: F  F , b 1, n ab  F.05, 2,84  3.15 F = 4.60. There is enough evidence to conclude that differences exist between the levels of factor B. 14.102 ANOVA Table

A 23 ANOVA 24 Source of Variation 25 Sample 26 Columns 27 Interaction 28 Within 29 30 Total

B

C

SS 5.33 56.33 1.33 34.67

df

97.67

1 1 1 8

D

E

F

MS 5.33 56.33 1.33 4.33

F P-value 1.23 0.2995 13.00 0.0069 0.31 0.5943

G F crit 5.32 5.32 5.32

11

a F = .31, p-value = .5943. There is not enough evidence to conclude that factors A and B interact. b F = 1.23, p-value = .2995. There is not enough evidence to conclude that differences exist between the levels of factor A.

392


c F = 13.00, p-value = .0069. There is enough evidence to conclude that differences exist between the levels of factor B. 14.103 ANOVA Table

A 29 ANOVA 30 Source of Variation 31 Sample 32 Columns 33 Interaction 34 Within 35 36 Total

B

C

SS 177.25 0.38 9.25 159.75

df

346.63

2 1 2 18

D

E

MS 88.63 0.38 4.63 8.88

F P-value 9.99 0.0012 0.04 0.8394 0.52 0.6025

F

G F crit 3.55 4.41 3.55

23

a F = .52, p-value = .6025. There is not enough evidence to conclude that factors A and B interact. b F = 9.99, p-value = .0012. There is enough evidence to conclude that differences exist between the levels of factor A. c F = .04, p-value = .8394. There is not enough evidence to conclude that differences exist between the levels of factor B. 14.104 ANOVA Table A 35 ANOVA 36 Source of Variation 37 Sample 38 Columns 39 Interaction 40 Within 41 42 Total

B

C

SS 135.85 151.25 6.25 726.20

df 3 1 3 72

1019.55

D

E

F

G

MS 45.28 151.25 2.08 10.09

F

P-value 0.0060 0.0002 0.8915

F crit 2.7318 3.9739 2.7318

4.49 15.00 0.21

79

The test for interaction yields (F = .21, p-value = .8915) and the test for the differences between educational levels (F = 4.49, p-value = .0060) is the same as in Example 14.4. However, in this exercise there is evidence of a difference between men and women (F = 15.00, p-value = .0002). 14.105 ANOVA Table A 35 ANOVA 36 Source of Variation 37 Sample 38 Columns 39 Interaction 40 Within 41 42 Total

B

C

SS 345.85 61.25 72.25 726.20

df

1205.55

3 1 3 72

D

E

MS 115.28 61.25 24.08 10.09

F 11.43 6.07 2.39

79

393

F

G

P-value 3.25E-06 0.0161 0.0760

F crit 2.7318 3.9739 2.7318


Compared to Example 14.4, the test for interaction has the same conclusion, although the value of F is larger and the p-value is smaller. Moreover, the mean number of jobs differs between the educational levels (F = 11.43, p-value = 0) and between men and women (F = 6.07, p-value = .0161). 14.106a. There are 12 treatments. b. There are two factors, tax form and income group. c. There are a = 4 forms and b = 3 income groups.

A 28 29 ANOVA 30 Source of Variation 31 Sample 32 Columns 33 Interaction 34 Within 35 36 Total

B

C

D

SS 6719 6280 5102 88217

df

MS 3359.4 2093.3 850.3 816.8

2 3 6 108

106317

E

F

F P-value 4.11 0.0190 2.56 0.0586 1.04 0.4030

G

F crit 3.08 2.69 2.18

119

d. F = 1.04, p-value = .4030. There is not enough evidence to conclude that forms and income groups interact e. F = 2.56, p-value = .0586. There is not enough evidence to conclude that differences exist between the forms. f. F = 4.11, p-value = .0190. There is enough evidence to conclude that differences exist between the three income groups. 14.107a. Detergents and temperatures b. The response variable is the whiteness score. c. There are a = 5 detergents and b = 3 temperatures.

A 29 ANOVA 30 Source of Variation 31 Sample 32 Columns 33 Interaction 34 Within 35 36 Total

B

C

D

SS 3937 2967 2452 14910

df

MS 1968.5 741.9 306.5 110.4

24267

2 4 8 135

E

F

F P-value 17.82 0.0000 6.72 0.0001 2.78 0.0071

G F crit 3.06 2.44 2.01

149

d. Test for interaction: F = 2.78, p-value = .0071. There is sufficient evidence to conclude that detergents and temperatures interact. The F-tests in Parts e and f are irrelevant. 14.108a. Factor A is the drug mixture and factor B is the schedule. b. The response variable is the improvement index. c. There are a = 4 drug mixtures and b = 2 schedules.

394


A 23 ANOVA 24 Source of Variation 25 Sample 26 Columns 27 Interaction 28 Within 29 30 Total

B

C

SS 14.40 581.80 548.60 804.80

df

1949.60

D MS 1 14.40 3 193.93 3 182.87 32 25.15

E

F

G

F P-value 0.57 0.4548 7.71 0.0005 7.27 0.0007

F crit 4.15 2.90 2.90

39

d Test for interaction: F = 7.27, p-value = .0007. There is sufficient evidence to conclude that the schedules and drug mixtures interact. 14.109a. There are 2 factors--class configuration and time period. b. The response variable is the number of times students ask and answer questions. c. There are 2 levels of class configuration and 3 levels of time period.

ANOVA Source of Variation Sample Columns Interaction Within

SS 13.33 46.67 206.67 202.00

Total

468.67

df 1 2 2 24

MS 13.33 23.33 103.33 8.42

F

P-value 1.58 0.2203 2.77 0.0826 12.28 0.0002

F crit 4.26 3.40 3.40

29

d. Interaction: F = 12.28, p-value = .0002. There is sufficient evidence to conclude that the class configuration and time interact. The other two F-tests are invalid. 14.110

A 23 ANOVA 24 Source of Variation 25 Sample 26 Columns 27 Interaction 28 Within 29 30 Total

B

C

SS 16.04 6.77 0.025 39.17

df

62.00

1 1 1 36

D

E

F

MS 16.04 6.77 0.025 1.09

F P-value 14.74 0.0005 6.22 0.0173 0.023 0.8814

G F crit 4.11 4.11 4.11

39

The p-values for interaction, machines, and alloys are .8814, .0173, .0005, and, respectively. Both machines and alloys are sources of variation. 14.111

395


A 35 ANOVA 36 Source of Variation 37 Sample 38 Columns 39 Interaction 40 Within 41 42 Total

B

C

SS 0.000309 0.000515 0.000183 0.004953

df

0.005959

D

3 1 3 32

MS 0.000103 0.000515 0.000061 0.000155

E

F

F P-value 0.66 0.5798 3.33 0.0775 0.39 0.7584

G F crit 2.90 4.15 2.90

39

The p-values for interaction, devices, and alloys are .7584, .0775, and .5798, respectively. There are no sources of variation. 14.112

A 29 ANOVA 30 Source of Variation 31 Sample 32 Columns 33 Interaction 34 Within 35 36 Total

B

C

SS 211.78 0.59 0.13 211.42

df

D

2 1 2 42

423.91

MS 105.89 0.59 0.0640 5.03

E

F

F P-value 21.04 0.0000 0.12 0.7348 0.0127 0.9874

G F crit 3.22 4.07 3.22

47

The p-values for interaction, methods, and skills are .9874, .7348, and 0, respectively. The only source of variation is skill level. 14.113a. The factors are mental outlook (2 levels) and physical condition (3 levels).

A 29 ANOVA 30 Source of Variation 31 Sample 32 Columns 33 Interaction 34 Within 35 36 Total

B

C

SS 2118.4 166.7 20.0 2336.2

df

4641.3

D MS 2 1059.22 1 166.67 2 10.02 54 43.26

E

F

F P-value 24.48 0.0000 3.85 0.0548 0.23 0.7941

G F crit 3.17 4.02 3.17

59

Test for interaction: F = .23, p-value = .7941. There is not enough evidence to infer interaction. b. F = 3.85, p-value = .0548. There is not enough evidence to conclude that differences exist between optimists and pessimists. c. F = 24.48, p-value = 0. There is sufficient evidence to conclude that differences exist between the three levels of physical condition.

14.114

H 0 : 1   2   3 =  4 H1 : At least two means differ. 396


F = 7.63, p-value = 0. There is enough evidence to infer that there are differences between the four degrees. 14.115a H 0 : 1   2   3 =  4 =  5

H1 : At least two means differ.

A 12 ANOVA 13 Source of Variation 14 Between Groups 15 Within Groups 16 17 Total

B

C

D

SS 1747.4 23983.7

df

MS 436.86 97.89

25731.1

4 245

E

F

F P-value 4.46 0.0017

G F crit 2.41

249

F = 4.46, p-value = .0017. There is enough evidence to infer that there are differences in the effect of the new assessment system between the five boroughs. b

397


Multiple Comparisons LSD Omega Treatment Treatment Difference Alpha = 0.05 Alpha = 0.05 Borough A Borough B -7.42 3.90 5.40 Borough C -6.18 3.90 5.40 Borough D -2.42 3.90 5.40 Borough E -3.94 3.90 5.40 1.24 3.90 5.40 Borough B Borough C Borough D 5.00 3.90 5.40 Borough E 3.48 3.90 5.40 Borough C Borough D 3.76 3.90 5.40 Borough E 2.24 3.90 5.40 Borough D Borough E -1.52 3.90 5.40 The mean assessments in borough A differs from the means in boroughs B and C. c The assessments for each borough are required to be normally distributed with equal variances. d The histograms are approximately bell-shaped and the sample variances are similar.

14.116

H 0 : 1   2   3 =  4 H1 : At least two means differ.

A 31 ANOVA 32 Source of Variation 33 Rows 34 Columns 35 Error 36 37 Total

B

C

SS 43980 4438 6113

df

D MS 19 2314.72 3 1479.21 57 107.25

54530

E

F

F P-value 21.58 0.0000 13.79 0.0000

G F crit 1.77 2.77

79

F = 13.79, p-value = 0. There is sufficient evidence to conclude that the reading speeds differ between the four typefaces. The typeface that was read the fastest should be used.

14.117

H 0 : 1   2   3 H1 : At least two means differ.

A 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

SS 406.5 16445.8

df

MS 203.25 111.88

16852.3

2 147

E

F

F P-value 1.82 0.1662

G F crit 3.06

149

F = 1.82, p-value = .1662. There is not enough evidence to infer that differences in attention span exist between the three products.

398


14.118

H 0 : 1   2   3 H1 : At least two means differ

A 17 ANOVA 18 Source of Variation 19 Rows 20 Columns 21 Error 22 23 Total

B

C

SS 195.33 43.52 33.81

df 6 2 12

272.67

D

E

MS 32.56 21.76 2.82

F P-value 11.55 0.0002 7.72 0.0070

F

G F crit 3.00 3.89

20

F = 7.72, p-value = .0070. There is enough evidence to infer that differences in attention span exist between the three products. 14.119

A 23 ANOVA 24 Source of Variation 25 Sample 26 Columns 27 Interaction 28 Within 29 30 Total

B

C

SS 123553 3965110 30006 4856578

df

8975248

D

MS 1 123553 2 1982555 2 15003 144 33726

E

F

F P-value 3.66 0.0576 58.78 0.0000 0.44 0.6418

G F crit 3.91 3.06 3.06

149

Interaction: F = .44, p-value = .6418. There is not enough evidence to conclude that age and gender interact. Age: F = 58.78, p-value = 0. There is sufficient evidence to conclude that differences in offers exist between the three age groups. Gender: F = 3.66, p-value = .0576. There is not enough evidence to conclude that differences in offers exist between males and females 14.120a H 0 : 1   2   3

H1 : At least two means differ.

A 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

SS 1769.5 2409.8

df

MS 884.74 6.48

4179.3

2 372

E

F

F P-value 136.58 0.0000

G F crit 3.02

374

F = 136.58, p-value = 0. There is sufficient evidence to infer that differences exist between the effects of the three teaching approaches. b

399


A B C D E 1 Multiple Comparisons 2 LSD Omega 3 4 Treatment Treatment Difference Alpha = 0.0167 Alpha = 0.05 Whole LanguagEmbedded -0.856 0.774 0.754 5 Pure -4.976 0.774 0.754 6 7 Embedded Pure -4.120 0.774 0.754 All three means differ from one another. From the sample means we may infer that the pure method is best, followed by embedded, and by whole-language. 14.121a H 0 : 1   2   3

H1 : At least two means differ.

A 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

SS 1913 8727

df

10640

D

2 87

E

MS 956.70 100.31

F

G

F P-value 9.54 0.0002

F crit 3.10

89

F = 9.54, p-value = .0002. There is sufficient evidence to infer that there are differences between the three groups. b

A B C D E 1 Multiple Comparisons 2 LSD Omega 3 Treatment Difference Alpha = 0.0167 Alpha = 0.05 4 Treatment Mozart White noise -9.30 6.31 6.14 5 Glass -10.20 6.31 6.14 6 Glass -0.90 6.31 6.14 7 White noise The mean time of the Mozart group differs from the mean times of white noise and the Glass groups.

14.122

H 0 : 1   2   3 =  4 H1 : At least two means differ.

A 11 ANOVA 12 Source of Variation 13 Between Groups 14 Within Groups 15 16 Total

B

C

SS 5990284 40024172

df

46014456

D

MS 3 1996761 290 138014

E

F

F P-value 14.47 0.0000

G F crit 2.64

293

F = 14.47, p-value = 0. There is enough evidence to infer differences in debt levels between the four types of degrees. 400


14.123

H 0 : 1   2   3 =  4 H1 : At least two means differ.

A 11 ANOVA 12 Source of Variation 13 Between Groups 14 Within Groups 15 16 Total

B

C

D

SS 3263 29685

df

MS 1087.8 106.0

32948

3 280

E

F

F P-value 10.26 0.0000

G F crit 2.64

283

F = 10.26, p-value = 0. There is enough evidence of differences between the four groups of investors.

14.124

H 0 : 1   2   3 =  4 H1 : At least two means differ.

A 11 ANOVA 12 Source of Variation 13 Between Groups 14 Within Groups 15 16 Total

B

C

D

SS 3007 10576

df

MS 1002.3 72.4

13583

3 146

E

F

F P-value 13.84 0.0000

G F crit 2.67

149

F = 13.84, p-value = 0. There is enough evidence to infer that the length of time depends on the size of the party 14.125 H0: µ 1 = µ 2 = µ 3 = µ 4 H1: At least two means differ.

A 11 ANOVA 12 Source of Variation 13 Between Groups 14 Within Groups 15 16 Total

B

C

D

SS 2.12 7.99

df

MS 0.705 0.0769

10.11

3 104

E

F

F P-value 9.17 0.0000

G F crit 2.69

107

F = 9.17, p-value = 0. There is sufficient evidence to infer that there are differences in changes to the TSE depending on the loss the previous day. 14.126 H0: µ 1 = µ 2 = µ 3 H1: At least two means differ.

401


A 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

SS 1.57 46.98

df

48.55

2 97

D

E

F

MS 0.787 0.484

F P-value 1.62 0.202233

G F crit 3.09

99

F = 1.62, p-value = .2022. There is no evidence to infer that at least one buy indicator is useful. 14.127 H0: µ 1 = µ 2 = µ 3 H1: At least two means differ.

F = 25.98, p-value = 0. There is sufficient evidence to infer that the amount of sleep differs between commuting categories?

14.128 H0: µ 1 = µ 2 = µ 3 H1: At least two means differ.

402


F = 2.70, p-value = .0684. There is not enough evidence to infer a difference between the three groups of teenagers. 14.129 a.H0: µ 1 = µ 2 = µ 3 = µ 4 H1: At least two means differ.

F = 111.4, p-value = 0. b. The incomes are required to be normally distributed. c. The histograms are bell shaped. 14.130 H0: µ 1 = µ 2 = µ 3 H1: At least two means differ.

403


F = 21.63, p-value = 0. 14.131 H0: µ 1 = µ 2 = µ 3 = µ 4 H1: At least two means differ.

F = 113.4, p-value = 0. 14.132a. H0: µ 1 = µ 2 = µ 3 = µ 4 = µ 5 = µ 6 = µ 7 H1: At least two means differ.

404


F = 9.42, p-value = 0. b.

405


All six days differ from Thursday. 14.133 H0: µ 1 = µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ.

406


F = 445.5, p-value = 0. b.

All pairs differ except for 2011 and 2016. 14.134 H0: µ 1 = µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ.

407


F = 23.19, p-value = 0. b.

All pairs differ except Electrical and Mechanical and Chemical and Computer. 14.135 H0: µ 1 = µ 2 = µ 3 = µ 4 H1: At least two means differ.

408


F = 45.48, p-value = 0. 14.136 H0: µ 1 = µ 2 = µ 3 H1: At least two means differ.

F = 35.90, p-value = 0.

409


b. Excel Workbook Groups

Means

| Difference |

LSD

i = 1, j = 2

4958

5956

998

294.6

i = 1, j = 3

4958

5708

750

297.6

i= 2, j = 3

5956

5708

248

300.5

Group 1 differs from groups 2 and 3. 14.137 H0: µ 1 = µ 2 = µ 3 = µ 4 = µ 5 = µ 6 H1: At least two means differ.

F = 6.77, p-value = 0. b.

410


The only pairs that differ are Guelph and Windsor and Brantford and Windsor. 14.138 Two-factor ANOVA

Interaction: F = 15.63, p-value = 0. 14.139 H0: µ 1 = µ 2 = µ 3 = µ 4 = µ 5 H1: At least two means differ.

411


F = 14.76, p-value = 0. b.

Pairs that differ: 1&2, 1&3, 1&4, 1&5, 2&5, 3&4, 3&5 14.140a. H0: µ 1 = µ 2 = µ 3 = µ 4 H1: At least two means differ.

412


F = .451, p-value = .7166. There is not enough evidence to infer that differences exist. 14.141 H0: Factors A and B do not interact H1: Factors A and B do interact F = 3.82, p-value = .0543. There is not enough evidence of interaction H0: The means of the bowls are equal H1: The means of the bowls differ F = 26.51, p-value = 0. There is enough evidence to conclude that the mean amount of ice cream differs between the two sizes of bowls. H0: The means of the scoops are equal H1: The means of the scoops differ F = 9.51, p-value = .0028. There is enough evidence to conclude that the mean amount of ice cream differs between the two sizes of scoop.

413


14.142

H 0 : 1   2   3 H1 : At least two means differ.

F = 39.93, p-value = 0.There is enough evidence to conclude that the number of kisses differ by location.

414


14.143

H 0 : 1   2   3 H1 : At least two means differ.

F = 127.43, p-value = 0. There Is sufficient evidence to conclude that the prices differed according to the way the brownies were offered.

14.144

H 0 : 1   2   3 H1 : At least two means differ.

1 2 3 4 5 6

A ANOVA Source of Variation Between Groups Within Groups

B

C

D

E

SS 11374 229170

df

MS 5687 125.0

F P-value 45.49 0.0000

Total

240544

1835

2 1833

F

G F crit 3.00

F = 45.49, p-value = 0. There is enough evidence to infer that family incomes differ between the three market segments.

14.145

H 0 : 1   2   3 H1 : At least two means differ.

1 2 3 4 5 6

A ANOVA Source of Variation Between Groups Within Groups

B

C

SS 52.82 1607.82

df

Total

1660.6

2 197

D

E

MS 26.41 8.16

F

199

415

3.24

F

G

P-value 0.0414

F crit 3.04


F = 3.24, p-value = .0414. There is enough evidence to infer that the distances driven differ between drivers who have had 0, 1, or 2 accidents.

H 0 : 1   2   3 =  4

14.146

H1 : At least two means differ. 1 2 3 4 5 6

A ANOVA Source of Variation Between Groups Within Groups

B

C

SS 6636.1 3595.9

df

Total

10232.0

3 344

D

E

F

MS 2212.03 10.45

F P-value 211.61 0.0000

G F crit 2.63

347

F = 211.61, p-value = 0. There is enough evidence to conclude that there are differences in the age of the car between the four market segments. Case 14.1 One-way analysis of variance H0: All 91 means are equal H1: At least two means differ

F = .845, p-value = .8489. There is not enough evidence to infer that a difference exists. Case 14.2 Episodes A 1 Anova: Single Factor 2 3 SUMMARY 4 Groups 5 Surgery 6 Drug 7 Placebo 8 9 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

E

Count 60 60 60

Sum Average Variance 198 3.30 1.74 178 2.97 1.19 207 3.45 1.68

SS 7.34 271.4

df

278.7

2 177

MS 3.67 1.53

179

416

F 2.40

F

G

P-value 0.0941

F crit 3.0470


F = 2.40, p-value = .0941. There is not enough evidence to conclude that there are differences in the number of episodes between the three types of treatments. Visits A 1 Anova: Single Factor 2 3 SUMMARY 4 Groups 5 Surgery 6 Drug 7 Placebo 8 9 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

E

Count 60 60 60

Sum Average Variance 130 2.17 2.51 114 1.90 1.38 147 2.45 2.83

SS 9.08 396.6

df 2 177

405.7

MS 4.54 2.24

F 2.03

F

G

P-value 0.1349

F crit 3.0470

179

F = 2.03, p-value = .1349. There is not enough evidence to conclude that there are differences in the number of physician visits between the three types of treatments. Prescriptions A 1 Anova: Single Factor 2 3 SUMMARY 4 Groups 5 Surgery 6 Drug 7 Placebo 8 9 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

E

Count 60 60 60

Sum Average Variance 201 3.35 2.20 178 2.97 3.05 205 3.42 3.43

SS 7.08 512.2

df

519.2

2 177

MS 3.54 2.89

F 1.22

F

G

P-value 0.2968

F crit 3.0470

179

F = 1.22, p-value = .2968. There is not enough evidence to conclude that there are differences in the number of prescriptions between the three types of treatments. Days

417


A 1 Anova: Single Factor 2 3 SUMMARY 4 Groups 5 Surgery 6 Drug 7 Placebo 8 9 10 ANOVA 11 Source of Variation 12 Between Groups 13 Within Groups 14 15 Total

B

C

D

E

Count 60 60 60

Sum Average Variance 689 11.48 22.76 663 11.05 19.71 779 12.98 28.22

SS 123.5 4170.8

df

4294.3

2 177

MS 61.76 23.56

F 2.62

F

G

P-value 0.0756

F crit 3.0470

179

F = 2.62, p-value = .0756. There is not enough evidence to conclude that there are differences in the number of days with respiratory infections between the three types of treatments.

418


Chapter 15 15.1

H0: p1 = .1, p2 = .2, p3 = .3, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

28

0.2

20

8

3.20

2

17

0.2

20

-3

0.45

3

19

0.2

20

-1

0.05

4

17

0.2

20

-3

0.45

5

19

0.2

20

-1

0.05

Total

100

1

100

0

4.20

p-value

0.3796

Rejection region:  2 >  2 ,k 1   .201,4  13.3

 2 = 4.53, p-value = .3386. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.2

H0: p1 = .1, p2 = .2, p3 = .3, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

12

0.1

15

-3

0.60

2

32

0.2

30

2

0.13

3

42

0.3

45

-3

0.20

4

36

0.2

30

6

1.20

5

28

0.2

30

-2

0.13

Total

150

1

150

0

2.27

p-value

0.6868

Rejection region:  2 > 2 ,k 1  .201,4  13.3

 2 = 2.27, p-value = .6868. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.3

H0: p1 = .1, p2 = .2, p3 = .3, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

6

0.1

7.5

-1.5

0.30

2

16

0.2

15

1

0.07

3

21

0.3

22.5

-1.5

0.10

4

18

0.2

15

3

0.60

5

14

0.2

15

-1

0.07

Total

75

1

75

0

1.13

457


p-value

0.8889

Rejection region:  2 > 2 ,k 1  .201,4  13.3

 2 = 1.13, p-value = .8889. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.4 The  2 statistic decreases.

15.5

H0: p1 = .3, p2 = .3, p3 = .2, p4 = .2 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

38

0.3

45

-7

1.09

2

50

0.3

45

5

0.56

3

38

0.2

30

8

2.13

4

24

0.2

30

-6

1.20

Total

150

1

150

0

4.98

p-value

0.1734

Rejection region: 

> 2 ,k 1  .205,3  7.81

2

 2 = 4.98, p-value = .1734. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.6

H0: p1 = .3, p2 = .3, p3 = .2, p4 = .2 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

76

0.3

90

-14

2.18

2

100

0.3

90

10

1.11

3

76

0.2

60

16

4.27

4

48

0.2

60

-12

2.40

Total

300

1

300

0

9.96

p-value

0.0189

Rejection region:  2 >  2 ,k 1  .205,3  7.81

 2 = 9.96, p-value = .0189. There is enough evidence to infer that at least one p i is not equal to its specified value.

15.7

H0: p1 = .2, p2 = .2, p3 = .2, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value.

458


Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

28

0.2

20

8

3.20

2

17

0.2

20

-3

0.45

3

19

0.2

20

-1

0.05

4

17

0.2

20

-3

0.45

5

19

0.2

20

-1

0.05

Total

100

1

100

0

4.20

p-value

0.3796

Rejection region:  2 >  2 ,k 1   .210,4  7.78

 2 = 4.20, p-value = .3796. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.8

H0: p1 = .15, p2 = .40, p3 = .35, p4 = .10 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

41

0.15

34.95

6.05

1.05

2

107

0.4

93.2

13.8

2.04

3

66

0.35

81.55

-15.55

2.97

4

19

0.1

23.3

-4.3

0.79

Total

233

1

233

0

6.85

p-value

0.0769

Rejection region:  2 >  2 ,k 1  .205,3  7.81

 2 = 6.85, p-value = .0769. There is not enough evidence to infer that at least one p i is not equal to its specified value.

15.9

H0: p1 = 1/6, p2 = 1/6, p3 = 1/6, p4 = 1/6, p5 = 1/6, p6 = 1/6 H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

114

0.167

100

14

1.96

2

92

0.167

100

-8

0.64

3

84

0.167

100

-16

2.56

4

101

0.167

100

1

0.01

5

107

0.167

100

7

0.49

6

102

0.167

100

2

0.04

Total

600

1

600

459

0

5.70

p-value

0.3365


Rejection region:  2 >  2 ,k 1  .205,5  11.07

 2 = 5.70, p-value = .3365. There is not enough evidence to infer that the die is not fair.

15.10

H0: p1 = .05, p2 = .25, p3 = .40, p4 = .25, p5 = .05 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

1

11

0.05

2

32

0.25

3

62

0.4

4

29

0.25

5

16

Total

150

Expected value E

F-E

((F-E)^2)/E

7.5

3.5

1.63

37.5

-5.5

0.81

60

2

0.07

37.5

-8.5

1.93

0.05

7.5

8.5

9.63

1

150

0

14.07

p-value

0.0071

Rejection region:  2 >  2 ,k 1   .210,4  7.78

 2 = 14.07, p-value = .0071. There is enough evidence to infer that grades are distributed differently from grades in the past.

15.11

H0: p1 = .2, p2 = .2, p3 = .2, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

8

0.2

5

3

1.80

2

4

0.2

5

-1

0.20

3

8

0.2

5

3

1.80

4

3

0.2

5

-2

0.80

5

2

0.2

5

-3

1.80

Total

25

1

25

0

6.40

p-value

0.1712

Rejection region:  2 >  2 ,k 1   .205,4  9.49

 2 = 6.40, p-value = .1712. There is not enough evidence to infer that the professor does not randomly distribute the correct answer over the five choices.

15.12

H0: p1 = .72, p2 = .15, p3 = .10, p4 = .03 H1: At least one pi is not equal to its specified value.

460


Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

159

0.72

180

-21

2.45

2

28

0.15

37.5

-9.5

2.41

3

47

0.1

25

22

19.36

4

16

0.03

7.5

8.5

9.63

Total

250

1

250

0

33.85

p-value

0.0000

Rejection region:  2 >  2 ,k 1  .205,3  7.81

 2 =33.85, p-value = 0. There is enough evidence to infer that the aging schedule has changed.

15.13

H0: p1 = .15, p2 = .25, p3 = .40, p4 = .20 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

36

0.15

29.55

6.45

1.41

2

58

0.25

49.25

8.75

1.55

3

74

0.4

78.8

-4.8

0.29

4

29

0.2

39.4

-10.4

2.75

Total

197

1

197

0.00

6.00

p-value

0.1116

Rejection region:  2 >  2 ,k 1  .205,3  7.81

 2 = 6.00, p-value = .1116. There is not enough evidence to infer that certain sizes of cars are involved in a higher than expected percentage of accidents.

15.14

H0: p1 = .31, p2 = .51, p3 = .18 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

408

0.31

372

36

3.48

2

571

0.51

612

-41

2.75

3

221

0.18

216

5

0.12

Total

1200

1

1200

0

6.35

p-value

0.0419

Rejection region: 

2

>  2 ,k 1   .210,2  4.61

 2 = 6.35, p-value = .0419. There is enough evidence to infer that voter support has changed since the election.

15.15

H0: p1 = .0528, p2 = .3472, p3 = .51, p4 = .2, p5 = .09 H1: At least one pi is not equal to its specified value. 461


Cell

Frequency F

Hypothesized P

1

9

0.0528

2

123

0.3472

3

149

0.51

4

39

Total

320

Expected value E

F-E

((F-E)^2)/E

16.896

-7.896

3.69

111.104

11.896

1.27

163.2

-14.2

1.24

0.09

28.8

10.2

3.61

1

320

0.00

9.81

p-value

0.0202

Rejection region:  2 >  2 ,k 1  .205,3  7.81

 2 = 9.81, p-value = .0202. There is enough evidence to infer that this umpire differs.

15.16

H0: p1 = .40, p2 = .144, p3 = .178, p4 = .095, p5 = .183

H1 : At least one p i is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

36

0.4

46

-10

2.17

2

26

0.144

16.56

9.44

5.38

3

24

0.178

20.47

3.53

0.61

4

14

0.095

10.925

3.075

0.87

5

15

0.183

21.045

-6.045

1.74

Total

115

1

115

0

10.77

p-value

0.0293

Rejection region:  2 > 2 ,k 1   .205,4  9.49

 2 = 10.77, p-value = .0293. There is enough evidence to infer that this rookie differs.

15.17

H0: p1 = .52, p2 = .19, p3 = .15, p4 = .14 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

248

0.52

266.240

-18.24

1.25

2

108

0.19

97.280

10.72

1.18

3

47

0.15

76.800

-29.80

11.56

4

109

0.14

71.680

37.32

19.43

Total

512

1

512

0.00

33.42

p-value

0.0000

Rejection region:  2 >  2 ,k 1   .205,3  7.81

462


 2 = 33.42, p-value = 0. There is enough evidence to infer that the pitching distribution has changed.

15.18

H0: p1 = .17, p2 = .18, p3 = .17, p4 = .18, p5 = .21, P6 = .09 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

137

0.170

139.06

-2.06

0.03

2

133

0.180

147.24

-14.24

1.38

3

142

0.170

139.06

2.94

0.06

4

136

0.180

147.24

-11.24

0.86

5

171

0.210

171.78

-0.78

0.00

6

99

0.090

73.62

25.38

8.75

Total

818

1

818

0.00

11.08

p-value

0.0498

Rejection region:  2 >  2 ,k 1   .205,5  11.07

 2 = 11.08, p-value = .0498. There is enough evidence to infer that the distribution has changed.

15.19

H0: p1 = .024, p2 = .030, p3 = .067, p4 = .169, p5 = .212, P6 = .174, p7 = .080, p8 = .057, p9 = .059, p10 = .049, p11 = .049, p12 = .028 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

26

0.024

30.22

-4.22

0.59

2

2

0.030

37.77

-35.77

33.88

3

13

0.067

84.35

-71.35

60.36

4

185

0.169

212.77

-27.77

3.62

5

414

0.212

266.91

147.09

81.06

6

184

0.174

219.07

-35.07

5.61

7

122

0.080

100.72

21.28

4.50

8

44

0.057

71.76

-27.76

10.74

9

24

0.059

74.28

-50.28

34.04

10

57

0.049

61.69

-4.69

0.36

11

95

0.049

61.69

33.31

17.98

12

93

0.028

35.25

57.75

94.60

Total

1259

0.998

1256.482

2.52

347.33

p-value

0.0000

Rejection region:  2 >  2 ,k 1   .205,11  19.68

 2 = 347.33, p-value = 0. There is enough evidence to infer that the percentages have changed. 463


H 0 : p1  .23, p 2  .40, p 3  .15, p 4  .22

15.20

H1: At least one pi is not equal to its specified value. Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

63

0.23

73.600

-10.60

1.53

2

125

0.40

128.000

-3.00

0.07

3

45

0.15

48.000

-3.00

0.19

4

87

0.22

70.400

16.60

3.91

Total

320

1

320

0.00

5.70

p-value

0.1272

Rejection region: 

2

>  2 ,k 1  .205,3  7.81

 2 = 5.70, p-value = .1272. There is not enough evidence to infer that there has been a change in proportions.

15.21

H0: p1 = .773, p2= .132, p3= .095 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

1890

0.773

1961.874

-71.874

2.63

2

386

0.132

335.016

50.984

7.76

3

262

0.095

241.11

20.89

1.81

Total

2538

1

2538

0.00

12.20

p-value

0.0022

 2 = 12.20, p-value = .0022. There is sufficient evidence to conclude that the survey overrepresented at least one race.

15.22

H0: p1+4 = .525, p2 = .057, p3 = .100, p5 = .317 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1+4

1239

0.525

1330.350

-91.35

6.27

2

209

0.057

144.438

64.56

28.86

3

411

0.100

253.400

157.60

98.02

5

675

0.317

803.278

-128.28

20.49

Total

2534

0.999

2531.466

2.53

153.63

p-value

0.0000

 2 = 153.63, p-value = 0. There is sufficient evidence to conclude that the survey overrepresented at least one category.

464


15.23

H0: p0 = .123, p1 = .296, p2 = .194, p3+4 = .386 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

0

330

0.123

312.174

17.83

1.02

1

1269

0.296

751.248

517.75

356.83

2

186

0.194

492.372

-306.37

190.64

3+4

753

0.386

979.668

-226.67

52.44

Total

2538

0.999

2535.462

2.54

600.93

p-value

0.0000

 2 = 600.93, p-value = 0. There is sufficient evidence to conclude that the survey overrepresented at least one category.

15.24

H0: p1 = .616, p2 = .132, p3 = .176, p5 = .075 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

4425

0.616

3705.240

719.76

139.82

2

747

0.132

793.980

-46.98

2.78

3

556

0.176

1058.640

-502.64

238.65

5

287

0.075

451.125

-164.13

59.71

Total

6015

0.999

6008.985

6.02

440.96

p-value

0.0000

 2 = 440.96, p-value = 0. There is sufficient evidence to conclude that the survey overrepresented at least one race.

15.25

H0: p1 = .126, p2 = .296, p3 = .196, p5 = .383 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

546

0.126

757.890

-211.89

59.24

2

1599

0.296

1780.440

-181.44

18.49

3

1030

0.196

1178.940

-148.94

18.82

4

2840

0.383

2303.745

536.26

124.83

Total

6015

1.001

6021.015

-6.01

221.37

p-value

0.0000

 2 = 221.37, p-value = 0. There is sufficient evidence to conclude that the survey overrepresented at least one category.

465


15.26

H0: The two variables are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205,1  3.84

 2 = 19.10, p-value = 0. There is enough evidence to infer that the two variables are dependent. 15.27

H0: The two variables are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)(c 1   .205,1  3.84

 2 = 9.56, p-value = .0020. There is enough evidence to infer that the two classifications L and M are dependent.

15.28

H0: The two variables are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)(c 1   .205,1  3.84

 2 = 4.78, p-value = .0289. There is enough evidence to infer that the two classifications L and M are dependent.

15.29 The  2 statistic decreases.

466


15.30

H0: The two variables are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .210, 2  4.61

 2 = 4.40, p-value = .1110. There is not enough evidence to infer that the two classifications R and C are dependent.

15.31

H0: The two variables (responses and employee group) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205, 2  5.99

 2 = 2.27, p-value = .3221. There is not enough evidence to infer that responses differ among the three groups of employees.

15.32

H0: The two variables (shirt condition and shift) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205, 2  5.99

 2 = 2.35, p-value = .3087. There is not enough evidence to infer that there are differences in quality among the three shifts.

467


15.33

H0: The two variables (economic option and political affiliation) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)(c 1   .201,6  16.8

 2 = 70.67, p-value = 0. There is sufficient evidence to infer that political affiliation affects support for economic options.

15.34

H0: The two variables (inducement and return) are independent H1: The two variables are dependent

Rejection region:  >  2 ,( r 1)( c1   .205, 2  5.99 2

 2 = 19.68, p-value = 0. There is sufficient evidence to infer that the return rates differ among the different inducements.

15.35

H0: The two variables (newspaper and occupation) are independent H1: The two variables are dependent

468


Rejection region:  >  2 ,( r 1)( c 1   .205,6  12.59 2

 2 = 32.57, p-value = 0. There is sufficient evidence to infer that occupation and newspaper are related.

15.36

H0: The two variables (remedy/placebo and side effects) are independent H1: The two variables are dependent

Rejection region:  >  2 ,( r 1)(c 1   .205,3  7.81 2

 2 = .918, p-value = .8211. There is not enough evidence to infer that the reported side effects differ.

15.37

H0: The two variables (last purchase and second-last purchase) are independent H1: The two variables are dependent

Rejection region:  >  2 ,( r 1)( c 1   .205,9  16.9 2

469


 2 = .67, p-value = .9999. There is no evidence of a relationship.

15.38

H0: The two variables (education and smoker) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205,3  7.81

 2 = 41.76, p-value = 0. There is sufficient evidence to infer that the amount of education is a factor in determining whether a smoker will quit.

15.39

H0: The two variables (education and smoker) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205,15  25.0

 2 = 22.36, p-value = .0988. There is not enough evidence to infer that there is a relationship between an adult’s source of news and his or her heartburn condition.

15.40

H0: The two variables (university and degree) are independent

470


H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205,9  16.92

 2 = 43.36, p-value = 0. There is enough evidence to infer that undergraduate degree and the university applied to are related.

15.41

H0: The two variables (results and financial ties) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205, 2  5.99

 2 = 20.99, p-value = 0. There is sufficient evidence to infer that the research findings are related to whether drug companies fund the research.

15.42

H0: The two variables (degree and approach) are independent H1: The two variables are dependent

471


Rejection region:  >  2 ,( r 1)( c 1   .205,6  12.59 2

2 = 20.89, p-value = .0019. There is sufficient evidence to infer that there are differences in teaching approach among the four types of degree. The editor can design books and sales campaigns based on the distribution of degrees.

15.43

H0: The two variables (year and weapon) are independent H1: The two variables are dependent

2 = 4.76, p-value = .5751. There is not sufficient evidence to infer that the frequency of the use of weapons in robberies differed over the three years.

15.44

H0: The two variables are independent H1: The two variables are dependent

472


2 = 1.22, p-value = .9761. There is not enough evidence to conclude that the deteriorating condition of American bridges has changed over the years.

15.45

H0: The two variables are independent H1: The two variables are dependent

2 = 26.69, p-value = .0008. There is sufficient evidence to infer that there are differences in household types between the three countries.

473


15.46

H0: The two variables are independent H1: The two variables are dependent

2 = 311.4, p-value = 0. There is sufficient evidence to infer that there are differences in household types between the four countries.

15.47

H0: The two variables are independent H1: The two variables are dependent

2 = .395, p-value = .9413. There is not enough evidence to conclude that there are differences in obesity rates between the four Commonwealth nations

474


15.48

H0: The two variables are independent H1: The two variables are dependent

2 = 11.82, p-value = .0080. There is enough evidence to conclude that there are differences in smoking between the four Scandinavian countries.

15.49

H0: The two variables are independent H1: The two variables are dependent

2 = 2.32, p-value = .5084.

15.50

H0: The two variables are independent H1: The two variables are dependent

475


2 = 22.49, p-value = .0002. There is enough evidence to infer that there are differences in adult cigarette smoking between the four Commonwealth countries.

15.51

H0: The two variables are independent H1: The two variables are dependent

2 = 4.53, p-value = .1038. There is not enough evidence to infer that there are differences.

15.52

476


2 = 8.52, p-value = .0365. There is enough evidence to conclude that differences exist.

15.53

H0: The two variables are independent H1: The two variables are dependent

2 = 12.57, p-value = .0019.

15.54

H0: The two variables are independent H1: The two variables are dependent

477


2 = 29.70, p-value = 0.

15.55

H0: The two variables are independent H1: The two variables are dependent

2 = 7.16, p-value = .0279.

15.56

H0: The two variables are independent H1: The two variables are dependent

2 = 15.80, p-value = .0004.

15.57

H0: The two variables are independent H1: The two variables are dependent

478


2 = 2.02, p-value = .3646

15.58

H0: The two variables are independent H1: The two variables are dependent

2 = 40.19, p-value = 0.

15.59

H0: The two variables are independent H1: The two variables are dependent

2 = 159.79, p-value = 0. 479


15.60

H0: The two variables are independent H1: The two variables are dependent

2 = 146.54, p-value = 0.

15.61

H0: The two variables are independent H1: The two variables are dependent

2 = 80.36, p-value = 0.

15.62

H0: The two variables are independent H1: The two variables are dependent

480


2 = 29.42, p-value = 0.

15.63

H0: The two variables are independent H1: The two variables are dependent

2 = 2.95, p-value = .4000.

15.64

H0: The two variables are independent H1: The two variables are dependent

481


2 = .271, p-value = .9653.

15.65

H0: The two variables are independent H1: The two variables are dependent

2 = 5.13, p-value = .1625.

15.66

H0: The two variables are independent H1: The two variables are dependent

482


2 = 50.56, p-value = 0.

15.67

H0: The two variables are independent H1: The two variables are dependent

2 = 9.04, p-value = .0287.

15.68

H0: The two variables are independent H1: The two variables are dependent

483


2 = 5.99, p-value = .1121.

15.69

H0: The two variables are independent H1: The two variables are dependent

2 = 2.90, p-value = .4077.

15.70

H0: The two variables are independent H1: The two variables are dependent

484


2 = 4.41, p-value = .2204.

15.71

H0: The two variables are independent H1: The two variables are dependent

2 = 10.13, p-value = .0175.

15.72

H0: The two variables are independent H1: The two variables are dependent

485


2 = 1.97, p-value = .5796.

15.73

H0: The two variables are independent H1: The two variables are dependent

2 = 46.71, p-value = 0.

15.74

H0: The two variables are independent H1: The two variables are dependent

486


2 = .684, p-value = .7102.

15.75

H0: The two variables are independent H1: The two variables are dependent

2 = 5.00, p-value = .0254.

15.76

H0: The two variables are independent H1: The two variables are dependent

487


2 = 1.79, p-value = .1810.

15.77

H0: The two variables are independent H1: The two variables are dependent

2 = 20.03, p-value = 0.

15.78

H0: The two variables are independent H1: The two variables are dependent

488


2 = 4.51, p-value = .0337.

15.79

H 0 : The data are normally distributed H1 : The data are not normally distributed Expected

Observed

Value e i 6.68

Value f i 10

f i  ei 3.32

(f i  ei ) 2 / ei 1.65

-1.5 < Z  -0.5 .2417

24.17

18

-6.17

1.58

-0.5 < Z  0.5

.3829

38.29

48

9.71

2.46

0.5 < Z  1.5

.2417

24.17

16

-8.17

Z > 1.5

.0668

6.68

8

1.32

Total

1

100

100

Interval Z  -1.5

Probability .0668

2.76 0.26

 2 = 8.71

Rejection region:  2 >  2 ,k 3   .205,2  5.99

 2 = 8.71, p-value = .0128. There is enough evidence to infer that the data are not normally distributed.

15.80

H 0 : The data are normally distributed H1 : The data are not normally distributed Expected

Observed

Value e i 7.94

Value f i 6

f i  ei -1.94

(f i  ei ) 2 / ei 0.47

Interval Z  -1

Probability .1587

-1 < Z  0

.3413

17.07

27

9.93

5.78

0<Z  1

.3413

17.07

14

-3.07

0.55

Z>1

.1587

7.94

3

-4.94

3.07

Total

1

50

50

489

 2 = 9.87


Rejection region:  2 >  2 ,k 3  .210,1  2.71

 2 = 9.87, p-value = .0017. There is sufficient evidence to infer that the data are not normally distributed.

15.81

H 0 : Times are normally distributed H1 : Times are not normally distributed.

A B C D 1 Chi-Squared Test of Normality 2 Hours 3 7.15 4 Mean 1.65 5 Standard deviation 200 6 Observations 7 Probability Expected Observed Intervals 8 (z <= -1.5) 0.0668 13.36 11 9 0.2417 48.35 55 10 (-1.5 < z <= -0.5) (-0.5 < z <= 0.5) 0.3829 76.59 52 11 (0.5 < z <= 1.5) 0.2417 48.35 67 12 (z > 1.5) 0.0668 13.36 15 13 14 15 16.62 16 chi-squared Stat 2 17 df 0.0002 18 p-value 5.9915 19 chi-squared Critical

 2 = 16.62, p-value = .0002. There is sufficient evidence to infer that the amount of time at parttime jobs is not normally distributed.

15.82

H 0 : Costs are normally distributed H1 : Costs are not normally distributed

490


A B 1 Chi-Squared Test of Normality 2 Drug Cost 3 29.69 4 Mean 27.53 5 Standard deviation 900 6 Observations 7 Intervals 8 Probability 9 (z <= -2) 0.0228 (-2 < z <= -1) 10 0.1359 (-1 < z <= 0) 11 0.3413 (0 < z <= 1) 12 0.3413 (1 < z <= 2) 13 0.1359 14 (z > 2) 0.0228 15 16 chi-squared Stat 506.76 17 df 3 18 p-value 0 19 chi-squared Critical 7.8147

C

D

Expected Observed 20.48 0 122.31 9 307.21 599 307.21 248 122.31 17 20.48 27

 2 = 506.76, p-value = 0. There is sufficient evidence to infer that drug costs are not normally distributed. 15.83

Successful firms:

H 0 : Productivity in successful firms is normally distributed H1 : Productivity in successful firms is not normally distributed

A B C D 1 Chi-Squared Test of Normality 2 3 Successful 4 Mean 5.02 5 Standard deviation 1.39 6 Observations 200 7 8 Probability Expected Observed Intervals 9 (z <= -1.5) 0.0668 13.36 12 10 (-1.5 < z <= -0.5) 0.2417 48.35 52 11 (-0.5 < z <= 0.5) 0.3829 76.59 72 12 (0.5 < z <= 1.5) 0.2417 48.35 55 13 (z > 1.5) 0.0668 13.36 9 14 15 3.0288 16 chi-squared Stat 2 17 df 18 p-value 0.2199 19 chi-squared Critical 5.9915

 2 =3.03, p-value = .2199. There is not enough evidence to infer that productivity in successful firms is not normally distributed. 491


Unsuccessful firms:

H 0 : Productivity in unsuccessful firms is normally distributed H1 : Productivity in unsuccessful firms is not normally distributed

A B 1 Chi-Squared Test of Normality 2 Unsuccessful 3 7.80 4 Mean 3.09 5 Standard deviation 200 6 Observations 7 Probability Intervals 8 (z <= -1.5) 0.0668 9 0.2417 10 (-1.5 < z <= -0.5) (-0.5 < z <= 0.5) 0.3829 11 (0.5 < z <= 1.5) 0.2417 12 (z > 1.5) 0.0668 13 14 15 1.1347 16 chi-squared Stat 2 17 df 0.567 18 p-value 5.9915 19 chi-squared Critical

C

D

Expected Observed 13.36 12 48.35 47 76.59 83 48.35 44 13.36 14

 2 = 1.13, p-value = .5670. There is not enough evidence to infer that productivity in unsuccessful firms is not normally distributed.

15.84

H 0 : Reaction times are normally distributed H1 : Reaction times re not normally distributed

Phone

492


A B C D 1 Chi-Squared Test of Normality 2 3 Phone 4 Mean 0.646 5 Standard deviation 0.045 6 Observations 125 7 8 Intervals Probability Expected Observed 9 (z <= -1.5) 0.0668 8.35 8 10 (-1.5 < z <= -0.5) 0.2417 30.22 32 11 (-0.5 < z <= 0.5) 0.3829 47.87 47 12 (0.5 < z <= 1.5) 0.2417 30.22 29 13 (z > 1.5) 0.0668 8.35 9 14 15 16 chi-squared Stat 0.2351 17 df 2 18 p-value 0.8891 19 chi-squared Critical 5.9915

 2 = .235, p-value = .8891. There is not enough evidence to infer that reaction times of those using the cell phone are not normally distributed.

Not on phone

A B 1 Chi-Squared Test of Normality 2 Not 3 0.601 4 Mean 0.053 5 Standard deviation 145 6 Observations 7 Intervals Probability 8 9 (z <= -1.5) 0.0668 10 (-1.5 < z <= -0.5) 0.2417 11 (-0.5 < z <= 0.5) 0.3829 12 (0.5 < z <= 1.5) 0.2417 13 (z > 1.5) 0.0668 14 15 16 chi-squared Stat 3.1752 17 df 2 18 p-value 0.2044 19 chi-squared Critical 5.9915

C

D

Expected Observed 9.69 8 35.05 40 55.52 55 35.05 29 9.69 13

 2 = 3.18, p-value = .2044. There is not enough evidence to infer that reaction times of those not using the cell phone are not normally distributed.

493


15.85

H 0 : Matched pairs differences of sales are normally distributed H1 : Matched pairs differences of sales are not normally distributed

A B C D 1 Chi-Squared Test of Normality 2 Difference 3 19.75 4 Mean 30.63 5 Standard deviation 40 6 Observations 7 Probability Expected Observed Intervals 8 (z <= -1) 0.1587 6.35 6 9 (-1 < z <= 0) 0.3413 13.65 14 10 (0 < z <= 1) 0.3413 13.65 14 11 (z > 1) 0.1587 6.35 6 12 13 14 15 0.0553 16 chi-squared Stat 1 17 df 0.8140 18 p-value 2.7055 19 chi-squared Critical

 2 = .055, p-value = .8140. There is not enough evidence to infer that matched pairs difference of sales is not normally distributed.

15.86

H0: p1= 1/3. P2 = 1/3, p3 = 1/3 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

14

0.333

10

4

1.60

2

10

0.333

10

0

0.00

3

6

0.333

10

-4

1.60

Total

30

1

30

0.00

3.20

p-value

0.2019

Rejection region:  2 >  2 ,k 1   .210,2  4.61

 2 = 3.20, p-value = .2019. There is not enough evidence to infer that the game is unfair.

15.87

H0: p1= .2. P2 = .2, p3 = .2, p4 = .2, p5 = .2 H1: At least one pi is not equal to its specified value.

Cell

Frequency F

Hypothesized P

Expected value E

F-E

((F-E)^2)/E

1

87

0.2

72.4

14.6

2.94

2

62

0.2

72.4

-10.4

1.49

494


3

71

0.2

72.4

-1.4

0.03

4

68

0.2

72.4

-4.4

0.27

5

74

0.2

72.4

1.6

0.04

Total

362

1

362

0.00

4.77

p-value

0.3119

Rejection region:  2 >  2 ,k 1   .205, 4  9.49

 2 = 4.77, p-value = .3119. There is not enough evidence to infer that absenteeism is higher on some days of the week.

15.88

H0: The two variables (shift and day) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .210, 4  7.78

 2 = 5.42, p-value = .2465. There is not enough evidence to infer that there is a relationship between the days an employee is absent and the shift on which the employee works.

15.89

H0: The two variables (satisfaction and relationship) are independent H1: The two variables are dependent

Rejection region:  2 >  2 ,( r 1)( c 1   .205,6  12.59

 2 = 74.47, p-value = 0. There is sufficient evidence to infer that the level of job satisfaction depends on boss/employee gender relationship.

15.90

H0: The two variables (Country and stress) are independent H1: The two variables are dependent

495


A B C 1 Contingency Table 2 3 Stress 4 Country 1 5 1 266 6 2 347 7 3 153 8 4 164 9 5 92 10 TOTAL 1022 11 12 13 chi-squared Stat 14 df 15 p-value 16 chi-squared Critical

D

E

2 315 276 187 128 79 985

TOTAL 581 623 340 292 171 2007

20.3755 4 0.0004 9.4877

 2 = 20.38, p-value = .0004. There is enough evidence to infer that Americans and Canadians differ in their sources of stress.

15.91

H0: The two variables (method and quit) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 Quit 3 1 4 Method 1 104 5 2 14 6 TOTAL 118 7 8 9 chi-squared Stat 10 df 11 p-value 12 chi-squared Critical 13

D

E

F

G

2 125 17 142

3 32 5 37

4 49 9 58

TOTAL 310 45 355

0.5803 3 0.9009 7.8147

 2 = .580, p-value = .9009. There is not enough evidence to infer that the four methods differ in their success rates.

15.92

H0: The two variables (education and section) are independent H1: The two variables are dependent

496


A B C 1 Contingency Table 2 3 Section 4 Education 1 5 1 4 6 2 27 7 3 1 8 4 10 9 TOTAL 42 10 11 12 chi-squared Stat 13 df 14 p-value 15 chi-squared Critical

D

E

F

G

2 21 32 20 44 117

3 31 18 42 22 113

4 14 2 22 3 41

TOTAL 70 79 85 79 313

86.6154 9 0 16.919

 2 = 86.62, p-value = 0. There is sufficient evidence to infer that educational level affects the way adults read the newspaper.

15.93a The expected frequency is 1/49.

H 0 : p1 = 1/49, p2 = 1/49, . . . , p49 = 1/49

b

H1:At least one pi is not equal to its specified value. Number i f i

ei

(f i  e i )

(f i  ei ) 2 / ei

1

5

312(1/49) = 6.37

-1.38

0.29

2

6

312(1/49) = 6.37

-0.38

0.02

3

7

312(1/49) = 6.37

0.63

0.06

.

.

.

.

.

.

47

6

312(1/49) = 6.37

-0.37

0.02

48

10

312(1/49) = 6.37

3.63

2.07

49

6

312(1/49) = 6.37

-0.37

0.02

Total

312

 2 = 38.22

312

 2 = 38.22, p-value = .8427. There is not enough evidence to infer that the numbers were not generated randomly.

15.94

H0: The two variables are independent H1: The two variables are dependent

497


2 = 15.39, p-value = .0005.

15.95

H0: The two variables are independent H1: The two variables are dependent

2 = 11.36, p-value = .0034.

15.96

H0: The two variables are independent H1: The two variables are dependent

498


2 = 166.8, p-value = 0.

15.97

H0: The two variables are independent H1: The two variables are dependent

2 = 51.16, p-value = 0.

15.98 Binomial probabilities with n = 5 and p = .5: P(X = 0) = .0313, P(X = 1) = .1563, P(X = 2) = .3125, P(X = 3) = .3125, P(X = 4) = .1563, P(X = 5) = .0313

H0: p0 = .0313, p1 = .1563, p2 = .3125, p3 = .3125, p4 = .1563, p5 = .0313 H1: At least one pi is not equal to its specified value. Cell i

fi

ei

(f i  e i )

(f i  ei ) 2 / ei

0

8

200(.0313) = 6.26

1.74

0.48

1

35

200(.1563) = 31.26

3.74

0.45

2

57

200(.3125) = 62.50

-5.50

0.48

3

69

200(.3125) = 62.50

6.50

0.68

499


4

28

200(.1563) = 31.26

-3.26

0.34

5

3

200(.0313) = 6.26

-3.26

1.70

Total

200

 2 = 4.13

200

Rejection region:  2 >  2 ,61   .205,5  11.1

 2 = 4.13, p-value = .5310. There is not enough evidence to infer that at the number of boys in families with 5 children is not a binomial random variable with p =.5.

15.99

H0: The two variables (cold and group) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 3 Group 4 Cold 1 5 1 17 6 2 12 7 3 9 8 4 16 9 TOTAL 54 10 11 12 chi-squared Stat 13 df 14 p-value 15 chi-squared Critical

D

E

2 11 13 18 18 60

TOTAL 28 25 27 34 114

4.139 3 0.2468 7.8147

 2 = 4.14, p-value = .2468. There is not enough evidence to infer there are differences between the four groups.

15.100 H0: The two variables (faculty and retire) are independent H1: The two variables are dependent

500


A B C 1 Contingency Table 2 Retire 3 1 4 Faculty 1 174 5 2 13 6 TOTAL 187 7 8 9 chi-squared Stat 10 df 11 p-value 12 chi-squared Critical 13

D

E

F

G

H

2 51 7 58

3 113 22 135

4 42 7 49

5 86 6 92

TOTAL 466 55 521

9.732 4 0.0452 9.4877

 2 = 9.73, p-value = .0452. There is enough evidence to infer that whether a professor wishes to retire is related to the faculty.

15.101 H0: The two variables (tree choice and age category) are independent H1: The two variables are dependent A B C 1 Contingency Table 2 Tree choice 3 4 Age category 1 5 1 136 6 2 196 TOTAL 332 7 8 9 10 chi-squared Stat df 11 p-value 12 13 chi-squared Critical

D

E

F

2 309 339 648

3 158 370 528

TOTAL 603 905 1508

38.41 2 0 5.9915

 2 = 38.41, p-value = 0. There is enough evidence to infer that there are differences in the choice of Christmas tree between the three age categories.

15.102 H0: The two variables (network and ask) are independent H1: The two variables are dependent

501


A B C 1 Contingency Table 2 Ask 3 1 4 Network 1 19 5 2 104 6 TOTAL 123 7 8 9 chi-squared Stat 10 df 11 p-value 12 chi-squared Critical 13

D

E

F

2 30 107 137

3 43 123 166

TOTAL 92 334 426

4.573 2 0.1016 5.9915

 2 = 4.57, p-value = .1016. There is not enough evidence to conclude that there are differences in responses between the three network news shows.

15.103 a H0: The two variables (education and group) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 3 Group 4 Education 1 5 1 5 6 2 70 7 TOTAL 75 8 9 10 chi-squared Stat 11 df 12 p-value 13 chi-squared Critical

D

E

F

G

2 113 305 418

3 73 189 262

4 40 55 95

TOTAL 231 619 850

26.7059 3 0 7.8147

 2 = 26.71, p-value = 0. There is enough evidence to infer that there are differences in educational attainment between those who belong and those who do not belong to the health conscious group. b

H0: The two variables (education and buy Special X) are independent H1: The two variables are dependent

502


A B C 1 Contingency Table 2 Buy Sp X 3 1 4 Education 1 70 5 2 5 6 TOTAL 75 7 8 9 chi-squared Stat 10 df 11 p-value 12 chi-squared Critical 13

D

E

F

G

2 376 42 418

3 229 33 262

4 87 8 95

TOTAL 762 88 850

2.9416 3 0.4007 7.8147

 2 = 2.94, p-value = .4007. There is not enough evidence to infer that there is a relationship between the four education groups and whether a person buys Special X.

15.104 a H0: The two variables (gender and vote) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 3 Votes 4 Gender 1 5 1 189 6 2 203 7 TOTAL 392 8 9 10 chi-squared Stat 11 df 12 p-value 13 chi-squared Critical

D

E

2 169 204 373

TOTAL 358 407 765

0.6483 1 0.4207 3.8415

 2 = .648, p-value = .4207. There is not enough evidence to infer that voting and gender are related. b

H0: The two variables (education and vote) are independent H1: The two variables are dependent

503


A B C 1 Contingency Table 2 Votes 3 1 4 Educ 1 48 5 2 34 6 TOTAL 82 7 8 9 chi-squared Stat 10 df 11 p-value 12 chi-squared Critical 13

D

E

F

G

2 164 178 342

3 107 134 241

4 39 61 100

TOTAL 358 407 765

7.7214 3 0.0521 7.8147

 2 = 7.72, p-value = .0521. There is not enough evidence to infer that voting and educational level are related. c

H0: The two variables (income category and vote) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 3 Votes 4 Income 1 5 1 38 6 2 21 7 TOTAL 59 8 9 10 chi-squared Stat 11 df 12 p-value 13 chi-squared Critical

D

E

F

G

2 186 185 371

3 105 128 233

4 29 73 102

TOTAL 358 407 765

23.108 3 0 7.8147

 2 = 23.11, p-value = 0. There is enough evidence to infer that voting and income are related.

15.105 H0: The two variables are independent H1: The two variables are dependent

504


2 = 5.49, p-value = .2409.

15.106 H0: The two variables are independent H1: The two variables are dependent

2 = 57.45, p-value = 0.

15.107 H0: The two variables are independent H1: The two variables are dependent

2 = 12.93, p-value = .0441. 505


15.108 H0: The two variables are independent H1: The two variables are dependent

2 = 28.97, p-value = 0.

15.109 H0: The two variables are (type of work and segment) independent H1: The two variables are dependent

A B C 1 Contingency Table 2 Segment 3 1 4 Work 1 157 5 2 219 6 3 256 7 TOTAL 632 8 9 10 11 chi-squared Stat 12 df 13 p-value 14 chi-squared Critical

D

E

F

2 44 53 102 199

3 217 264 524 1005

TOTAL 418 536 882 1836

23.0946 4 0.0001 9.4877

 2 = 23.0946, p-value = .0001. There is enough evidence to infer that there are differences in employment between the three market segments.

15.110 H0: The two variables (value and segment) are independent H1: The two variables are dependent

506


A B C 1 Contingency Table 2 Segment 3 1 4 Value 1 147 5 2 221 6 3 339 7 TOTAL 707 8 9 10 chi-squared Stat 11 df 12 p-value 13 chi-squared Critical 14

D

E

F

2 135 155 254 544

3 136 160 289 585

TOTAL 418 536 882 1836

4.5122 4 0.3411 9.4877

 2 = 4.51, p-value = .3411. There is not enough evidence to infer that there are differences in the definition of value between the three market segments.

15.111 H0: The two variables (breakfast and group) are independent H1: The two variables are dependent

A B C 1 Contingency Table 2 3 Group 4 Breakfast 1 5 1 3 6 2 44 7 3 17 8 4 25 9 TOTAL 89 10 11 12 chi-squared Stat 13 df 14 p-value 15 chi-squared Critical

D

E

F

G

2 15 53 10 77 155

3 29 95 79 77 280

4 222 292 135 77 726

TOTAL 269 484 241 256 1250

206.4984 9 0 16.919

 2 = 206.5, p-value = 0. There is enough evidence to infer that there are differences in frequency of healthy breakfasts between the three four segments.

507


Chapter 16 16.1 a The slope coefficient tells us that for additional inch of father’s height the son’s height increases on average by .516. The y-intercept is meaningless. b On average the son will be shorter than his father. c On average the son will be taller than his father.

16.2 a Scatter Diagram 20

Sales

15 10 5 0 0

20

40

60

80

100

120

Advertising

b

Total

xi 23 46 60 54 28 33 25 31 36 88 90 99 613 n

xi2 529 2,116 3,600 2,916 784 1,089 625 961 1,296 7,744 8,100 9,801 39,561

yi 9.6 11.3 12.8 9.8 8.9 12.5 12.0 11.4 12.6 13.7 14.4 15.9 144.9 n

x i = 613

i 1

 i 1

x i yi 220.8 519.8 768.0 529.2 249.2 412.5 300.0 353.4 453.6 1205.6 1296.0 1,574.1 7,882.2

n

n

y i = 144.9

yi2 92.16 127.69 163.84 96.04 79.21 156.25 144.00 129.96 158.76 187.69 207.36 252.81 1,795.77

x i2 = 39,561

i 1

 x y = 7,882.2 i

i

i 1

n n   xi yi   n 1  (613 )(144 .9  1   i 1 s xy  x i y i  i 1  = 12  1 7,882 .2    43.66 n  1  i 1 n 12       

 

509


2     n    xi      n 2  1  x i2   i 1   = 1 39,561  (613 )   749 .7 s 2x  n  1  i 1 n  12  1  12       

b1 

s xy

=

s 2x

43 .66  .0582 749 .7

x

 x  613  51.08

y

 y  144 .9  12.08

i

n

12

i

n

12

b 0  y  b1x = 12.08 – (.0582)(51.08) = 9.107 The sample regression line is

ŷ = 9.107 + .0582x The slope tells us that for each additional thousand dollars of advertising sales increase on average by .0582 million. The y-intercept has no practical meaning.

16.3 a

Total

xi 8.5 7.8 7.6 7.5 8.0 8.4 8.8 8.9 8.5 8.0 82.0

x i2 72.25 60.84 57.76 56.25 64.00 70.56 77.44 79.21 72.25 64.00 674.56

yi 115 111 185 201 206 167 155 117 133 150 1,540

n

 x = 82.0 i

i 1

n

n

yi2 13,225 12,321 34,225 40,401 42,436 27,889 24,025 13,689 17,689 22,500 248,400

x i yi 977.5 865.8 1,406.0 1,507.5 1,648.0 1,402.8 1,364.0 1,041.3 1,130.5 1,200.0 12,543.4

n

 y = 1,540  x = 674.56  x y = 12,543.4 2 i

i

i 1

i 1

i

i

i 1

n n   xi yi   n (82 .0)(1,540  1  1   i 1 12 ,543 .4  = s xy  x i y i  i 1   9.40    10  1  10 n  1  i 1 n      

 

2     n   xi   n    2  1   x i2   i 1   = 1 674 .56  (82 .0)   .24 s 2x  n  1  i 1 n  10  1  10       

510


b1 

s xy

=

s 2x

9.40  39 .17 .24

x

 x  82.0  8.20

y

 y  1,540  154 .0

i

n

10

i

n

10

b 0  y  b1x = 154.0 – (–39.17)(8.20) = 475.2 The sample regression line is

ŷ = 475.2 – 39.17x b. The slope coefficient tells us that for each additional 1 percentage point increase in mortgage rates, the number of housing starts decreases on average by 39.17. The y-intercept has no meaning.

16.4a

Scatter Diagram 20 15

Overweight

10 5 0 -5

0

10

20

30

40

50

-10 -15 Television

b

xi 42 34 25 35 37 38 31 33 19 29 38 28

yi 18 6 0 –1 13 14 7 7 –9 8 8 5

x i2 1,764 1,156 625 1,225 1,369 1,444 961 1,089 361 841 1,444 784

511

yi2 324 36 0 1 169 196 49 49 81 64 64 25

x i yi 756 204 0 –35 481 532 217 231 –171 232 304 140


Total

29 36 18 472

3 14 –7 86

n

841 1,296 324 15,524

n

x i = 472

i 1

x i2 = 15,524

i 1

i 1

87 504 –126 3,356

n

n

y i = 86

9 196 49 1,312

 x y = 3,356 i

i

i 1

n n   xi yi   n (472 )(86  1  1   i 1 x i y i  i 1 s xy   = 15  1 3,356  15   46 .42 n  1  i 1 n       

 

2     n    xi  n     2  1  x i2   i 1   = 1 15,524  (472 )   47 .98 s 2x  n  1  i 1 n  15  1  15       

b1 

s xy

=

s 2x

46 .42  .9675 47 .98

x

 x  472  31.47

y

 y  86  5.73

i

n

15

i

n

15

b 0  y  b1x = 5.73 – (.9675)(31.47) = –24.72 The sample regression line is

ŷ = –24.72 + .9675x The slope coefficient indicates that for each additional hour of television weight increases on average by .9675 pounds. The y-intercept is the point at which the regression line hits the y–axis; it has no practical meaning.

16.5a

Total

xi 80 68 78 79 87 74 86 92 77 84 805

yi 20,533 1,439 13,829 21,286 30,985 17,187 30,240 37,596 9,610 28,742 211,447

x i2 6,400 4,624 6,084 6,241 7,569 5,476 7,396 8,464 5,929 7,056 65,239

512

yi2 421,604,089 2,070,721 191,241,241 453,093,796 960,070,225 295,392,969 914,457,600 413,459,100 92,352,100 826,102,564 5,569,844,521

x i yi 1,642,640 97,852 1,078,662 1,681,594 2,695,695 1,271,838 2,600,640 3,458,832 739,970 2,414,328 17,682,051


n

n

n

n

 y = 211,447  x = 65,239  x y = 17,682,051

 x = 805

2 i

i

i

i 1

i 1

i 1

i

i

i 1

n n   xi yi   n (805 )( 211,447  1  1   i 1 17 ,682 ,051  = s xy  x i y i  i 1    73,396   10 10  1  n n  1  i 1      

 

2   n     xi   n    2  1   x i2   i 1   = 1 65,239  (805 )   48 .50 s 2x  n  1  i 1 n  10  1  10       

b1 

s xy

=

s 2x

73,396  1,513 48 .50

x

 x  805  80.5

y

 y  211,447  21,145

i

n

10

i

n

10

b 0  y  b1x = 21,145 – (1,513)(80.5) = –100,652 The sample regression line is

ŷ = –100,652 + 1,513x b. For each additional one degree increase in temperature the number of beers sold increases on average by 1,513. The y-intercept is the point at which the regression line hits the y–axis; it has no practical meaning.

513


16.6 a Scatter Diagram 30

Test scores

25 20 15 10 5 0 0

20

40

60

80

Lengths

b b1 

s xy s 2x

51 .86 = .2675, b 0  y  b1x = 13.80 – .2675(38.00) = 3.635 193 .9

=

Regression line: ŷ = 3.635 + .2675x (Excel: ŷ = 3.636 + .2675x) c b1 = .2675; for each additional second of commercial, the memory test score increases on average by .2675. b0 = 3.64 is the y-intercept.

16.7a b1 

s xy

=

s 2x

86 .93  1.465 , b 0  y  b1x = 210.4 – 1.465(13.68) =190.4. 59 .32

Regression line: ŷ = 190.4 + 1.465x (Excel: ŷ = 190.4 + 1.465x)

b For each additional floor prices increase on average by $1.465 thousand ($1,465). The yintercept has no practical meaning.

16.8 b1 

s xy s 2x

=

9.67  .0899 , b 0  y  b1x = 11.55 – .0899(45.49) =7.460. 107 .51

Regression line: ŷ =7.460 + .0899x (Excel: ŷ = 7.462 + .0900x) The slope coefficient tells us that for each additional year of age time increases on average by .0899 minutes. The y-intercept has no meaning.

16.9 a b1 

s 2xy s 2x

=

6.44  .1169 , b 0  y  b1x = 26.28 – (–.11697)(37.29) = 30.64. 55 .11

514


Regression line: ŷ = 30.64 – .1169x (Excel: ŷ = 30.63 – .1169x) b The slope coefficient indicates that for each additional year of age, the employment period decreases on average by .1169. b0 = 30.63 is the y-intercept.

16.10a b1 

s xy s 2x

=

20 .55  .1898 , b 0  y  b1x = 14.43 – .1898(37.64) = 7.286. 108 .3

Regression line: ŷ = 7.286 +.1898x (Excel: ŷ = 7.287 +.1897x) b For each additional cigarette the number of days absent from work increases on average by .1898. The y-intercept has no meaning.

16.11 b1 

s xy s 2x

=

22 .83  5.347 , b 0  y  b1x = 49.22 – 5.347(4.885) = 23.10. 4.270

Regression line: ŷ = 23.10 + 5.347x (Excel: ŷ = 23.11 + 5.347x) For each addition kilometer a house is away from its nearest fire station the percentage damage increases on average by 5.347.

16.12a b1 

s xy s 2x

=

30 ,945  44 .97 , b 0  y  b1x = 6,465 – 44.97(53.93) = 4040. 688 .2

Regression line: ŷ = 4040 + 44.97x (Excel: ŷ = 4040 + 44.97x) b. For each additional thousand square feet the price increases on average by $44.97 thousand.

16.13 b1 

s xy s 2x

=

81.78  .00138 , b 0  y  b1 x = 27.73 – (–.00138)(1199) = 29.39. 59,153

Regression line: ŷ = 29.39–.00138x (Excel: 29.39–.00138x) For each additional hour the price decreases on average by .00138 thousand dollars or $1.38.

16.14 b1 

s xy s 2x

=

310 .0  64 .05, b 0  y  b1x = 762.6 –64.05(4.75) = 458.4. 4.84

Regression line: ŷ = 458.4 + 64.05x (Excel: ŷ = 458.9 + 64.00x) For each additional occupant the electrical use increases on average by 64.05.

16.15 b1 

s xy s 2x

=

225 .7  1.959 , b 0  y  b1x = 270.3 –1.959(59.42) = 153.9. 115 .2

Regression line: ŷ = 153.9 + 1.959x (Excel: : ŷ = 153.9 + 1.958x) For each additional $1,000 of income the weekly food budget increases on average by $1.96. 515


16.16 a b1 

s xy s 2x

=

10 .78  .3039 , b 0  y  b1x = 17.20 – (–.3039)(11.33) = 20.64. 35 .47

Regression line: ŷ = 20.64 – .3039x (Excel: ŷ = 20.64 – .3038x) b The slope indicates that for each additional one percentage point increase in the vacancy rate rents on average decrease by $.3039.

16.17 a b1 

s xy s 2x

=

6.020  .604 , b 0  y  b1x = 59.59 –.604(68.95) =17.94. 9.966

Regression line: ŷ = 17.94 + .604x (Excel: ŷ = 17.93 + .604x) b For each additional inch of height income increases on average by $.604 thousand or $604.

16.18 b1 

s xy s 2x

=

.8258  .0514 , b 0  y  b1x = 93.89 –.0514(79.47) = 89.81. 16 .07

Regression line: ŷ = 89.81 + .0514x (Excel: ŷ = 89.81 + .0514x) For each additional mark on the test the number of non-defective products increases on average by .0514.

16.19 For each commercial length, the memory test scores are normally distributed with constant variance and a mean that is a linear function of the commercial lengths.

16.20 For each number of years of education incomes are normally distributed with constant variance and a mean that is a linear function of the number of years of education.

16.21 For each number of hours prices are normally distributed with constant variance and a mean that is a linear function of the number of hour.

16.22 b x i

yi

x i2

yi2

x i yi

1 3 4 6 9 8 10 41

1 8 15 33 75 70 95 297

1 9 16 36 81 64 100 307

1 64 225 1089 5625 4900 9025 20,929

1 24 60 198 675 560 950 2,468

Total

n

 i 1

n

x i = 41

 i 1

n

y i = 297

n

n

x i2 = 307

i 1

516

 i 1

y i2 = 20,929

 x y = 2,468 i

i 1

i


n n   xi yi   n (41)( 297  1  1   i 1 2,468  = s xy  x i y i  i 1    121 .4   7 7 1  n  1  i 1 n      

 

2     n   xi       n 1 (41) 2  1  i 1      x i2  s 2x  = 307    11 .14 n  1  i 1 n  7  1  7       

2     n    yi      n 2  1  y i2   i 1   = 1 20,929  (297 )   1,388 .0 s 2y  n  1  i 1 n  7  1  7       

b1 

s xy s 2x

=

121 .4  10 .90 11 .14

 s 2xy   (121 .4) 2  SSE  (n  1) s 2y  2  = (7  1)1,388 .0   390 .1   11 .14  s x    s 

SSE = n2

390 .1  8.83 (Excel: s  = 8.85) 72

H 0 : 1  0 H 1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,5  2.571 or t   t  / 2, n  2   t .025,5  2.571 s b1 

t

s (n  1)s 2x

=

8.83 (7  1)(11 .14 )

 1.08

b1  1 10 .90  0 =  10 .09 (Excel: t = 10.07, p–value = 0. There is enough evidence to infer a 1.08 s b1

linear relationship.

517


Scatter Diagram 120

100 80 60 40 20

0 -20

0

2

4

6

8

10

12

There does appear to be a linear relationship.

16.23a

Scatter Diagram 300

250 200 150 100 50

0 -50

0

1

2

3

4

5

6

7

There does appear to be a linear relationship.

b

xi

yi

x i2

yi2

x i yi

Total

3 5 2 6 1 4 21

25 110 9 250 3 71 468

9 25 4 36 1 16 91

625 12100 81 62500 9 5041 80,356

75 550 18 1500 3 284 2,430

518


n

n

 x = 21

 y = 468 i

i

i 1

i 1

n

 x = 91 2 i

i 1

n

n

 y = 80,356  x y = 2,430 2 i

i 1

i

i

i 1

n n   xi yi   n (21)( 468 )  1  1   i 1 2,430  = s xy  x i y i  i 1    158 .4   6 6 1  n n  1  i 1      

 

2   n     xi   n    2  1   x i2   i 1   = 1 91  (21)   3.50 s 2x  n  1  i 1 n  6  1  6       

2   n      yi      n 2  1  y i2   i 1   = 1 80,356  (468 )   8,770 s 2y  n  1  i 1 n  6  1  6       

b1 

s xy s 2x

=

158 .4  45 .26 3.5

 s 2xy   (158 .4) 2  SSE  (n  1) s 2y  2  = (6  1) 8,770   8,006   3.50  s x    s 

8006  44 .74 (Excel: s  = 44.75) 62

SSE = n2

H 0 : 1  0 H 1 : 1  0 Rejection region: t  t  / 2,n  2  t .025, 4  2.776 or t   t  / 2,n  2   t .025, 4  2.776 s b1 

t

s (n  1)s 2x

=

44 .74 (6  1)(3.50 )

 10 .69

45 .26  0 b1  1  4.23 (Excel: t = 4.23, p–value = .0134. There is enough evidence to = 10 .69 s b1

infer a linear relationship.

519


2     n    yi      n 2  1  y i2   i 1   = 1 1,795 .77  (144 .9)   4.191 16.24 a s 2y  n  1  i 1 n  12  1  12       

 s 2xy   (43 .66 ) 2  SSE  (n  1) s 2y  2  = (12  1) 4.191   18 .13   749 .7  s x    SSE = n2

s 

18 .13  1.347 (Excel: s  = 1.347) 12  2

H 0 : 1  0

b

H 1 : 1  0 Rejection region: t  t  / 2, n  2  t.025,10  2.228 or t   t  / 2, n  2   t.025,10  2.228 s

s b1 

t

(n  1)s 2x

=

1.347 (12  1)(749 .7)

 .0148

b1  1 .0582  0  3.93 (Excel: t = 3.93, p–value = .0028. There is enough evidence to = .0148 s b1

infer a linear relationship between advertising and sales. c b1  t  / 2,n 2 s b1  .0582  2.228(.0148)  .0582  .0330 LCL = .0252, UCL = .0912 d R2 

s 2xy

(43 .66 ) 2 =  .6067 (Excel: R 2 = .6066). 60.67% of the variation in sales is s 2x s 2y (749 .7)( 4.191)

explained by the variation in advertising. e There is evidence of a linear relationship. For each additional dollar of advertising sales increase, on average by .0582.

2   n     yi   n    2  1   y i2   i 1   = 1 248 ,400  (1,540 )   1249 16.25 s 2y  n  1  i 1 n  10  1  10       

R  2

s 2xy

(9.40 ) 2 =  .2948 (Excel: R 2 = .2948). 2 2 (. 24 )( 1 , 249 ) s xs y

 s 2xy   (9.49 ) 2  SSE  (n  1) s 2y  2  = (10  1)1,249   7,864   .24  s x   

520


s 

SSE = n2

7,864  31 .35 (Excel: s  = 31.48) 10  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2,n  2  t .025,8  2.306 or t   t  / 2, n  2   t .025,8  2.306 s b1 

t

s (n  1)s 2x

=

31 .35 (10  1)(. 24 )

 21 .33

39 .17  0 b1  1  1.84 (Excel: t = 1.83, p–value = .1048. There is not enough evidence = .21 .33 s b1

to infer a linear relationship between interest rates and housing starts.

2   n      yi  n     2  1  y i2   i 1   = 1 1,312  (86 )   58 .50 16.26 s 2y  n  1  i 1 n  15  1  15       

 2 s 2xy   (46 .42 ) 2  SSE  (n  1) s y  2  = (15  1) 58 .50   190 .2   47 .98  s x    s 

SSE = n2

190 .2  3.825 15  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,13  2.160 or t   t  / 2, n  2   t .025,13  2.160 s b1 

t

s (n  1)s 2x

=

3.825 (15  1)( 47.98)

 .1476

b1  1 .9675  0  6.55 (Excel: t = 6.55, p–value = 0.) There is enough evidence to = .1476 s b1

conclude that there is a linear relationship between hours of television viewing and how overweight the child is.

521


2     n    yi      n 2  1  y i2   i 1   = 1 5,569 ,844 ,521  (211,447 )   122 ,095 ,682 16.27 s 2y  n  1  i 1 n  10  1  10       

 s 2xy   (73,396 ) 2  SSE  (n  1) s 2y  2  = (10  1)122 ,095 ,682   99,216 ,698   48 .50  s x    s 

99 ,216 ,698  3,522 10  2

SSE = n2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2,n  2  t .025,8  2.306 or t   t  / 2, n  2   t .025,8  2.306 s

s b1 

t

(n  1)s 2x

=

3522 (10  1)( 48 .50 )

 168.6

b1  1 1,513  0  8.97 (Excel: t = 8.98, p–value = 0.) There is evidence of a linear = 168 .6 s b1

relationship between temperature and the number of beers sold.

 s 2xy   (51 .86 ) 2  16.28 a SSE  (n  1) s 2y  2  = (60  1) 47 .96   2,011     193 . 9 s x     s 

2,011  5.888 (Excel: s  = 5.888). Relative to the values of the dependent 60  2

SSE = n2

variable the standard error of estimate appears to be large indicating a weak linear relationship. b R2 

s 2xy s 2x s 2y

=

(51 .86 ) 2  .2892 (Excel: R 2 = .2893). (193 .9)( 47 .96 )

H 0 : 1  0

c

H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,58  2.000 or t   t  / 2, n  2   t .025,58  2.000 s b1 

t

s (n  1)s 2x

=

5.888 (60  1)(193 .9)

 .0550

b1  1 .2675  0  4.86 (Excel: t = 4.86, p–value = 0). There is enough evidence to infer a = .0550 s b1

linear relationship between memory test scores and length of commercial.

522


d b1  t  / 2,n 2 s b1  .2675  1.671(.0550)  .2675  .0919 LCL = .1756, UCL = .3594

 s 2xy   (86 .93) 2  16.29 SSE  (n  1) s 2y  2  = (50  1) 496 .4   18,081   59 .32  s x    s 

SSE = n2

18,081  19 .41 (Excel: s  = 19.41). Relative to the values of the dependent 50  2

variable the standard error of estimate appears to be large indicating a weak linear relationship. R2 

s 2xy

(86.93) 2 =  .2566 (Excel: R 2 = .2566). 2 2 s x s y (59.32)( 496 .4)

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025, 48  2.009 or t   t  / 2, n  2   t .025, 48  2.009 s b1 

t

s (n  1)s 2x

=

19 .41 (50  1)(59 .32 )

 .3600

b1  1 1.465  0  4.07 (Excel: t = 4.07, p–value = .0002). There is evidence of a linear = .3600 s b1

relationship. The relationship however, is weak.

 s 2xy   (9.67 ) 2    9500 .8 16.30 SSE  (n  1) s 2y  2  = (229  1) 42 .54    107 .51  s x    s 

SSE = n2

9500 .8  6.47 229  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2,n 2  t .025,227  1.96 or t   t  / 2,n 2   t .025,227  1.96 s b1 

t

s (n  1)s 2x

=

6.47 ( 229  1)(107 .51)

 .0413

b1  1 .0899  0 =  2.17 (Excel: t =2.18, p–value = .0305.) There is evidence of a linear .0413 s b1

relationship between age and time to complete census.

 s 2xy   (6.44 ) 2  16.31 SSE  (n  1) s 2y  2  = (80  1) 4.00   256 .5   55 .11  s x    523


s 

SSE = n2

256 .5  1.813 (Excel: s  = 1.813). Relative to the values of the dependent 80  2

variable the standard error of estimate appears to be large indicating a weak linear relationship. R2 

s 2xy

(6.44 ) 2 =  .1881 (Excel: R 2 = .1884). There is a weak linear relationship s 2x s 2y (55 .11)( 4.00 )

between age and number of weeks of employment.

 s 2xy   (20 .55) 2  16.32 SSE  (n  1) s 2y  2  = (231  1)19 .80   3657   108 .3  s x    s 

SSE = n2

3657  3.996 231  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025, 229  1.960 or t   t  / 2,n  2   t .025, 229  1.960 s b1 

t

s (n  1)s 2x

=

3.996 (231  1)(108 .3)

 .02532

b1  1 .1898  0  7.50 (Excel: t =7.49, p–value = 0.) There is evidence of a positive linear = .02532 s b1

relationship between cigarettes smoked and the number of sick days.

 s 2xy   (22 .83) 2  16.33 SSE  (n  1) s 2y  2  = (85  1) 243 .9   10,234   4.270  s x    s 

SSE = n2

10 ,234  11 .10 (Excel: s  = 11.11). 85  2

H 0 : 1  0

a

H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,83  1.990 or t   t  / 2, n  2   t .025,83  1.990 s b1 

t

s (n  1)s 2x

=

11.10 (85  1)( 4.270 )

 .5861

b1  1 5.347  0  9.12 (Excel: t =9.12, p–value = 0.) There is evidence of a linear = .5861 s b1

relationship between distance and fire damage. b b1  t  / 2,n 2 s b1  5.347  1.988(.5861)  5.347  1.166 LCL = 4.18, UCL = 6.51 524


c R2 

s 2xy

(22 .83) 2 =  .5005 (Excel: R 2 = .5004). There is a moderately strong linear 2 2 ( 4 . 270 )( 243 . 9 ) s xs y

relationship between distance and fire damage. 2  2 s 2xy  s   = (40  1)11,918 ,489  (30,945 )   410 ,554 ,683   SSE ( n 1 ) 16.34 y   688 .2  s 2x   

410 ,554 ,683  3,287 (Excel: s  = 3,287). There is a weak linear relationship. 40  2

a s 

SSE = n2

b

H 0 : 1  0 H1 : 1  0

Rejection region: t  t  / 2, n  2  t .025,38  2.021 or t   t  / 2, n  2   t .025,38  2.021 s b1 

t

s (n  1)s 2x

=

3,287 (40  1)(688 .2)

 20 .06

44 .97  0 b1  1  2.24 (Excel: t = 2.24, p–value = .0309.) There is enough evidence of a = 20 .06 s b1

linear relationship. c R2 

s 2xy

(30,945 ) 2 =  .1167 (Excel: R 2 = .1168) 11.67% of the variation in s 2x s 2y (688 .2)(11,918 ,489 )

percent damage is explained by the variation in distance to the fire station.

 s 2xy   (81 .78) 2  16.35 SSE  (n  1) s 2y  2  = (60  1) 3.623   207 .1   59,153  s x    s 

SSE = n2

207 .1  1.890 (Excel: s  =1.889 ). 60  2

H 0 : 1  0 H1 : 1  0 Rejection region: t   t ,n  2   t .05,58  1.671 s b1 

t

s (n  1)s 2x

=

1.890 (60  1)(59,153 )

 .001012

.00138  0 b1  1  1.364 (Excel: t = –1.367, p–value = .1769/2 = .0885.) There is not = .001012 s b1

enough evidence to infer that as hours of engine use increase the price decreases.

525


 s 2xy   (310 .0) 2  16.36 SSE  (n  1) s 2y  2  = (200  1) 56,725   7,337 ,056   4.84  s x   

s  R2 

7,337 ,056  191 .1 (Excel: s  = 192.5). 200  2

SSE = n2 s 2xy

(310 .0) 2 =  .3500 (Excel: R 2 = .3496) 35.00% of the variation in the s 2x s 2y (4.84 )(56,725 )

electricity use is explained by the variation in the number of occupants.

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,198  1.972 or t   t  / 2, n  2   t .025,198  1.972 s b1 

t

s (n  1)s 2x

191 .1

=

(200  1)( 4.84)

 6.16

b1  1 64 .05  0  10 .39 (Excel: t =10.32, p–value = 0.) There is enough evidence of a = 6.16 s b1

linear relationship.

16.37 a R 2 

s 2xy s 2x s 2y

=

(225 .7) 2  .2461 (Excel: R 2 = .2459) 24.61% of the variation in food (115 .2)(1,797 )

budgets is explained by the variation in household income.

 s 2xy   (225 .7) 2  b SSE  (n  1) s 2y  2  = (150  1)1,797   201,866   115 .2  s x    s 

SSE = n2

201,866  36 .93 (Excel: s  = 36.94 ). 150  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,148  1.977 or t   t  / 2, n  2   t .025,148  1.977 s b1 

t

s (n  1)s 2x

=

36 .93 (150  1)(115 .2)

 .2819

b1  1 1.959  0  6.949 (Excel: t = 6.95, p–value = 0.) There is evidence of a linear = .2819 s b1

relationship between food budget and household income.

526


 s 2xy   (10 .78) 2  16.38 SSE  (n  1) s 2y  2  = (30  1)11 .24   230 .9   35 .47  s x    s 

SSE = n2

230 .9  2.872 (Excel: s  = 2.873 ). 30  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025, 28  2.048 or t   t  / 2, n  2   t .025, 28  2.048 s b1 

t

s (n  1)s 2x

=

2.872 (30  1)(35.47 )

 .08955

.3039  0 b1  1  3.39 (Excel: t = –3.39, p–value = .0021.) There is sufficient evidence = .08955 s b1

to conclude that office rents and vacancy rates are linearly related.

 s 2xy   (6.020 ) 2  16.39 SSE  (n  1) s 2y  2  = (250  1) 71 .95   17 ,010   9.966  s x    s 

SSE = n2

17 ,010  8.28 (Excel: s  = 8.28). 250  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t , n  2  t .05, 248  1.645 s b1 

t

s (n  1)s 2x

=

8.28 (250  1)(9.966

 .166

b1  1 .604  0  3.64 (Excel: t = 3.63, p–value = .00034/2 = .00017) There is enough = .166 s b1

evidence to conclude that height and income are positively linearly related.

16.40 a R 2 

s 2xy

(.8258 ) 2 =  .0331 (Excel: R 2 = .0331) 3.31% of the variation in s 2x s 2y (16 .07 )(1.283 )

percentage of defectives is explained by the variation in aptitude test scores.

 2 s 2xy   (.8258 ) 2  b. SSE  (n  1) s y  2  = (45  1)1.283   54 .58   16 .07  s x    s 

SSE = n2

54 .58  1.127 (Excel: s  = 1.127). 45  2 527


H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025, 43  2.014 or t   t  / 2, n  2   t .025, 43  2.014 s

s b1 

t

(n  1)s 2x

=

1.127 (45  1)(16 .07 )

 .04238

b1  1 .0516  0  1.22 (Excel: t = 1.21, p–value = .2319) There is not enough evidence to = .04238 s b1

conclude that aptitude test scores and percentage of defectives are linearly related.

H0 :   0

16.41

H1 :   0 Rejection region: t   t  2, n  2   t .05,58  1.671

r

s xy sxsy

 .1767

(59,153 )(3.623 )

n2

tr

 81 .78

1 r 2

 (.1767 )

60  2 1  (.1767 ) 2

 1.367 (Excel: t = –1.367, p–value = .0885) There is not

enough evidence to infer a negative linear relationship.

H0 :   0

16.42

H1 :   0 Rejection region: t  t  / 2, n  2  t .025,58  2.000 or t   t  / 2, n  2   t .025,58  2.000

r

s xy sxsy

tr

51 .86

 .5378

(193 .9)( 47 .96 )

n2 1 r

2

 (.5378 )

60  2 1  (.5378 ) 2

 4.86 (Excel: t = 4.86, p–value = 0) This result is identical

to the one produced in Exercise 16.6.

16.43

H0 :   0 H1 :   0

Rejection region: t  t  / 2, n  2  t .025,148  1.977 or t   t  / 2, n  2   t .025,148  1.977

528


r

s xy sxsy

 .4961

(115 .2)(1,797 )

n2

tr

225 .7

1 r

2

 (.4961 )

150  2 1  (.4961 ) 2

 6.95 (Excel: t = 6.95, p–value = 0.) There is evidence of a

linear relationship between food budget and household income.

H0 :   0

16.44

H1 :   0 Rejection region: t  t , n  2  t .05, 229  1.645

r

s xy sxsy

tr

20 .55

 .4438

(108 .3)(19 .80 )

n2 1 r

2

 (.4438 )

231  2 1  (.4438 ) 2

 7.49 (Excel: t = 7.49, p–value = 0.) There is evidence of a

positive linear relationship between cigarettes smoked and the number of sick days.

16.45

a.

H0:β1 = 0 H1:β1 ≠ 0

529


t = 11.30, p-value = 0. There is enough evidence to conclude that there is a linear relationship between the two variables. b. LCL = 4.59, UCL = 6.53

16.46

a.

H0:β1 = 0 H1:β1 ≠ 0

t = 7.64, p-value = 0. There is enough evidence to infer that there is a linear relationship between grade and price. b. R2 = .4932; 49.32% of the variation in price is explained by the variation in grade.

16.47

530


a.

H0:β1 = 0 H1:β1 ≠ 0

t = 21.19, p-value = 0. There is enough evidence to infer that there is a linear relationship between age and number of days watching national news on television. b. R2 = .5854; 58.54% of the variation in number of days is explained by the variation in age.

16.48

531


H0:β1 = 0 H1:β1 ≠ 0 t = 10.81, p-value = 0. There is enough evidence to infer that age and intention to vote are linearly related.

16.49

a.

H0:β1 = 0 H1:β1 ≠ 0

t = 5.89, p-value = 0. There is not enough evidence to infer that there is a positive linear relationship between temperature and distance. b. R2 = .3166; 31.66% of the variation in distance is explained by the variation in temperature.

16.50

H0:β1 = 0 H1:β1 ≠ 0

t = 5.04, p-value = 0. There is enough evidence to infer that there is a linear relationship between income and position on the question should the government reduce income differences between rich and poor.

532


16.51

LCL = 554.3, UCL = 837.1

16.52

H0:β1 = 0 H1:β1 < 0

t = -8.58, p-value = 0. There is enough evidence to conclude that more educated people watch less television.

16.53

H0:β1 = 0 H1:β1 > 0

a. t = 13.81, p-value = 0. There is enough evidence to infer that more hours of work leads to higher income. b. LCL = 870.4, UCL = 1159.

16.54

H0:β1 = 0 H1:β1 ≠ 0

t = -.239, p-value = .8115. There is not enough evidence to infer a linear relationship between age and hours of work per week.

16.55

H0:β1 = 0 H1:β1 > 0

533


t = 5.67, p-value = 0. There is enough evidence to infer a positive linear relationship between age and hours of watching television per day.

16.56

H0:β1 = 0 H1:β1 > 0

t = 20.26, p-value = 0. There is enough evidence of a positive linear relationship between total family income and the number of earners is the family. LCL = 18,260 UCL = 22,173

16.57

H0:β1 = 0 H1:β1 > 0

t = 3.45, p-value = .00058/2 = .00029. There is enough evidence to conclude that more educated people are more likely to support government action to reduce income differences across the country differences.

16.58

H0:β1 = 0 H1:β1 > 0

t = 27.08, p-value = 0. There is sufficient evidence to conclude that a married person’s years of education are positively linearly related to his or her spouse’s level of education.

534


16.59

H0:β1 = 0 H1:β1 > 0

t = 1.17, p-value = .2439. There is not enough evidence to infer that as people become richer they tend to have more children.

16.60

H0:β1 = 0 H1:β1 > 0

t = 2.60, p-value = .0096. There is enough evidence to conclude that if one spouse works longer hours so does the spouse.

16.61

H0:β1 = 0 H1:β1 > 0

t = 17.99, p-value = 0. There is sufficient evidence to infer a positive linear relationship between years of education and the age when one has his or her first child.

16.62

H0:β1 = 0 H1:β1 < 0

t = -10.04, p-value = 0. There is enough evidence to conclude that more educated people have fewer children.

16.63

H0:β1 = 0 H1:β1 > 0

535


t = 21.94, p-value = 0. There is enough evidence to infer that the amount of education one completes is positively linearly related to his or her father. 16.64

H0:β1 = 0 H1:β1 > 0

t = 23.41, p-value = 0. There is enough evidence to conclude that there is a positive linear relationship between the years of education and the years of education of one’s mother.

16.65

H0:β1 = 0 H1:β1 > 0

t = 2.87, p-value = .0041/2 = .0021. There is sufficient evidence to conclude that harder working Americans are more likely to urge to want government not to reduce income differences.

16.66

H0:β1 = 0 H1:β1 > 0

t = 12.40, p-value = 0. There is enough evidence to conclude that more education leads to higher incomes. b. LCL = 5330, UCL = 7334

16.67

H0:β1 = 0 H1:β1 > 0

536


t= 10.83, p-value = 0. There is enough evidence to conclude that more education increases financial assets. b. LCL = 6181, UCL = 8917 c.

The coefficient of determination is .0889, which means that 8.89% of the variation in financial assets is explained by the variation in education.

16.68

H0:β1 = 0 H1:β1 > 0

t = 11.10, p-value = 0. There is enough evidence to conclude that education and debt are positively linearly related. b. LCL = 11,833, UCL = 16,913 c,

The coefficient of determination is .0931, which means that 9.31% of the variation in debt is explained by the variation in education.

16.69

H0:β1 = 0 H1:β1 < 0

t = -12.22, p-value = 0. There is evidence of a negative linear relationship.

16.70

H0:β1 = 0 H1:β1 > 0

a.t = 6.67, p-value = 0. There is enough evidence to infer a positive linear relationship. b. LCL = 815.6, UCL = 1495.

537


16.71

H0:β1 = 0 H1:β1 < 0

t = -9.12, p-value = 0. There is enough evidence to infer that age and amount spent on food away from home are negatively related.

16.72

H0:β1 = 0 H1:β1 < 0

t = -9.08, p-value = 0. There is enough evidence of a negative linear relationship.

16.73

H0:β1 = 0 H1:β1 > 0

t = 8.90, p-value = 0. There is enough evidence to conclude that as one grows older one increases unrealized capital gains.

16.74

a.

H0:β1 = 0 H1:β1 > 0

t = 18.01, p-value = 0. There is enough evidence of a positive linear relationship. b. LCL = 1324, UCL = 1648

16.75

538


a.

H0:β1 = 0 H1:β1 > 0

t = 2.89, p-value = .0039/2 = .0040. There is enough evidence of a positive linear relationship. b.LCL = 58.52, UCL = 306.6.

16.76 The prediction interval provides a prediction for a value of y. The confidence interval estimator of the expected value of y is an estimator of the population mean for a given x.

16.77 Yes, because

1

2 1 (x g  x)   n ( n  1)s 2x

2 1 (x g  x)  n ( n  1)s 2x

16.78 ŷ  b 0  b1x g  9.107  .0582 (80 )  13 .76 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 13 .76  1.812 (1.347 ) 1 

2 1 (x g  x) (where t  / 2,n  2  t .05,10  1.812 )  n (n  1)s 2x

1 (80  51 .08 ) 2   13 .76  2.657 12 (12  1)( 749 .7)

Lower prediction limit = 11.10, Upper prediction limit = 16.42 (Excel: 11.10, 16.41) 16.79 ŷ  b 0  b1x g  475 .2  39 .17 (7)  201 .0 Confidence interval estimate: ŷ  t  / 2,n 2 s 

 201 .0  1.860 (31 .35)

2 1 (x g  x) (where t  / 2, n  2  t .05,8  1.860 )  n (n  1)s 2x

1 (7  8.20 ) 2   201 .0  51 .06 10 (10  1)(. 24 )

LCL = 149.94, UCL = 252.06 (Excel: 149.75, 252.25) 16.80 ŷ  b 0  b1x g  24 .72  .9675 (35)  9.143 a Prediction interval: ŷ  t  / 2,n  2 s  1 

= 9.143  1.771(3.825 ) 1 

2 1 (x g  x) (where t  / 2,n  2  t .05,13  1.771)  n (n  1)s 2x

1 (35  31 .47 ) 2   9.143  7.057 15 (15  1)( 47 .98 )

Lower prediction limit =2.086, Upper prediction limit = 16.200 (Excel: 2.095, 16.209) b Confidence interval estimate: ŷ  t  / 2,n 2 s 

2 1 (x g  x)  n (n  1)s 2x

539


 9.143  1.771(3.825 )

1 (35  31 .47 ) 2   9.143  1.977 15 (15  1)( 47 .98 )

LCL = 7.166, UCL = 11.120 (Excel: 7.174, 11.130) 16.81 ŷ  b 0  b1x g  100 ,652  1,513 (75)  12,823 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 12 ,823  1.860 (3,522 ) 1 

2 1 (x g  x) (where t  / 2, n  2  t .05,8  1.860 )  n (n  1)s 2x

1 (75  80 .5) 2  12 ,823  7,084  10 (10  1)( 48 .50 )

Lower prediction limit = 5,739, Upper prediction limit = 19,907 (Excel: 5,740, 19,902) 16.82 ŷ  b 0  b1 x g  3.636 + .2675(30) = 11.61 a Prediction interval: ŷ  t  / 2,n  2 s  1 

= 11 .61  2.000 (5.888 ) 1 

2 1 (x g  x) (where t  / 2, n  2  t .025,58  2.000 )  n (n  1)s 2x

1 (30  38 ) 2   11 .61  11 .91 60 (60  1)(193 .9)

Lower prediction limit = -.3 (changed to 0), Upper prediction limit = 23.52 (Excel: -.26, 23.58) b Confidence interval estimate: ŷ  t  / 2,n 2 s 

 11 .61  2.000 (5.888 )

2 1 (x g  x)  n (n  1)s 2x

1 (30  38 ) 2   11 .61  1.757 60 (60  1)(193 .9)

LCL = 9.85, UCL = 13.37 (Excel: 9.90, 13.42) 16.83 ŷ  b 0  b1 x g  190.4 + 1.465(20) = 219.7 a Prediction interval: ŷ  t  / 2,n  2 s 

= 219 .7  2.009 (19 .41) 1 

2 1 (x g  x) (where t  / 2, n  2  t .025, 48  2.009 ) 1  n (n  1)s 2x

1 ( 20  13 .68 ) 2   219 .7  39 .65 50 (50  1)( 59 .32 )

Lower prediction limit = 180.1, Upper prediction limit = 259.4 (Excel: 180.0, 259.4) b ŷ  b 0  b1 x g  190.4 + 1.465(15) = 212.4

540


Confidence interval estimate: ŷ  t  / 2,n 2 s 

 212 .4  2.678 (19 .41)

2 1 (x g  x) (where t  / 2, n  2  t .005, 48  2.678 )  n (n  1)s 2x

1 (15  13 .68 ) 2   212 .4  7.460 50 (50  1)( 59 .32 )

LCL = 204.9, UCL = 219.9 (Excel: 204.9, 219.8) 16.84 ŷ  b 0  b1 x g  7.46 + .0899(40) = 11.06 Confidence interval estimate: ŷ  t  / 2,n 2 s 

 11 .06  1.645 (6.47 )

2 1 (x g  x) (where t  / 2,n 2  t .05,227  1.645 )  n (n  1)s 2x

1 ( 40  45 .49 ) 2  11 .06  .80  229 ( 229  1)(107 .51)

LCL = 10.26, UCL = 11.86 (Excel: 10.26, 11.86) 16.85 ŷ  b 0  b1 x g  30.64 – .1169(22) = 28.07 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 28 .07  1.990 (1.813 ) 1 

2 1 (x g  x) (where t  / 2,n  2  t .025,78  1.990 )  n (n  1)s 2x

1 ( 22  37 .29 ) 2   28 .07  3.73 80 (80  1)( 55 .11)

Lower prediction limit = 24.34, Upper prediction limit = 31.80 (Excel: 24.33, 31.79) 16.86 ŷ  b 0  b1 x g  7.286 + .1898(40) = 14.88 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 14 .88  1.960 (3.996 ) 1 

2 1 (x g  x) (where t  / 2,n  2  t .025, 229  1.960 )  n (n  1)s 2x

1 ( 40  37 .64 ) 2  14 .88  7.85  231 ( 231  1)(108 .3)

Lower prediction limit = 7.03, Upper prediction limit = 22.73 (Excel: 6.98, 22.77) 16.87 ŷ  b 0  b1 x g  23.10 + 5.347(8) = 65.88 a Prediction interval: ŷ  t  / 2,n  2 s 

= 65 .88  1.990 (11 .10 ) 1 

2 1 (x g  x) (where t  / 2, n  2  t .025,83  1.990 ) 1  n (n  1)s 2x

1 (8  4.885 ) 2   65 .88  22 .51 85 (85  1)( 4.270 )

541


Lower prediction limit = 43.37, Upper prediction limit = 88.39 (Excel:43.37, 88.39) b ŷ  b 0  b1 x g  23.10 + 5.347(5) = 49.84 Confidence interval estimate: ŷ  t  / 2,n 2 s 

2 1 (x g  x)  n (n  1)s 2x

1 (5  4.885 ) 2   49 .84  2.40 85 (85  1)( 4.270 )

 49 .84  1.990 (11 .10 )

LCL = 47.44, UCL = 52.24 (Excel: 47.44, 52.24) 16.88 ŷ  b 0  b1 x g  4,040 + 44.97(60) = 6,738 Confidence interval estimate: ŷ  t  / 2,n 2 s 

6,738  2.021(3,287 )

2 1 (x g  x) (where t  / 2,n  2  t .025,38  2.021)  n (n  1)s 2x

1 (60  53 .93) 2  6,738  1,079  40 ( 40  1)( 688 .2)

LCL = 5,659, UCL = 7,817 (Excel: LCL = 5,657, UCL = 7,818) 16.89 ŷ  b 0  b1 x g  29.39 – .00138(400) = 28.84 Prediction interval: ŷ  t  / 2,n  2 s 

2 1 (x g  x) (where t  / 2, n  2  t .005,58  2.660 ) 1  n (n  1)s 2x

= 28 .84  2.660 (1.889 ) 1 

1 ( 400  1,199 ) 2   28 .84  5.50 60 (60  1)( 59 ,153 )

Lower prediction limit = 23.34, Upper prediction limit = 34.34 (Excel: 23.33, 34.35) 16.90 ŷ  b 0  b1 x g  458.4 + 64.05(4) = 714.6 Confidence interval estimate: ŷ  t  / 2,n 2 s 

 714 .6  1.653 (191 .1)

2 1 (x g  x) (where t  / 2, n  2  t .05,198  1.653 )  n (n  1)s 2x

1 ( 4  4.75) 2   714 .6  23 .6 200 (200  1)( 4.84 )

LCL = 691.0, UCL = 738.2 (Excel: 691.2, 738.7) 16.91 ŷ  b 0  b1 x g  153.9 + 1.959(60) = 271.4 Prediction interval: ŷ  t  / 2,n  2 s  1 

2 1 (x g  x) (where t  / 2, n  2  t .05,148  1.656 )  n (n  1)s 2x

542


= 271 .4  1.656 (36 .93) 1 

1 (60  59 .42 ) 2   271 .4  61 .4 150 (150  1)(115 .2)

Lower prediction limit = 210.0, Upper prediction limit = 332.8 (Excel: 210.0, 332.7) 16.92 ŷ  b 0  b1 x g  20.64 – .3039(8) = 18.21 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 18 .21  2.048 (2.872 ) 1 

2 1 (x g  x) (where t  / 2, n  2  t .025, 28  2.048 )  n (n  1)s 2x

1 (8  11 .33) 2   18 .21  6.01 30 (30  1)( 35 .47 )

Lower prediction limit = 12.20, Upper prediction limit = 24.22 (Excel: 12.19, 24.22) 16.93 a ŷ  b 0  b1 x g  17.94 + .604(74) = 62.64 Confidence interval: ŷ  t  / 2,n 2 s 

= 62 .64  1.96 (8.28

2 1 (x g  x) (where t  / 2, n  2  t .025, 248  1.96 )  n (n  1)s 2x

1 (74  68 .95) 2   62 .64  1.94 250 ( 250  1)( 9.966 )

Lower confidence limit = 60.70, Upper confidence limit = 64.58 (Excel: 60.70, 64.59) b ŷ  b 0  b1 x g  17.94 + .604(68) = 59.01 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 59 .01  1.96 (8.28 1 

2 1 (x g  x) (where t  / 2, n  2  t .025, 248  1.96 )  n (n  1)s 2x

1 (68  68 .95) 2  59 .01  16 .26  250 ( 250  1)( 9.966 )

Lower prediction limit = 42.75, Upper prediction limit = 75.27 (Excel: 42.67, 75.36) 16.94 ŷ  b 0  b1 x g  89.81 + .0514(80) = 93.92 Confidence interval estimate: ŷ  t  / 2,n 2 s 

 93 .92  2.014 (1.127 )

2 1 (x g  x) where t  / 2,n 2  t .025,43  2.014  n (n  1)s 2x

1 (80  79 .47 ) 2   93 .92  .34 45 (45  1)(16 .07 )

LCL = 93.58, UCL = 94.26 (Excel: 93.57, 94.26)

16.95 543


LCL = 74.163, UCL = 80.057.

16.96

Lower prediction interval = -235.426, upper predication interval = 2070.219

16.97

LCL = 1.892, UCL = 2.103

16.98

544


Lower prediction interval = 2.468, upper predication interval = 8.604

16.99

Lower prediction interval = 211.350, upper predication interval = 219.702.

16.100 In all cases the linear relationship is far too weak to produce accurate predictions.

16.101

Lower prediction interval = 7.679, upper predication interval = 19.723. LCL = 13.578, UCL = 13.823

16.102

545


Lower prediction interval = -1.680, upper predication interval = 8.262. LCL = 3.150, UCL = 3.432.

16.103

Lower prediction interval = -28,401, upper predication interval = 119,175. LCL = 43,322, UCL = 47,387.

16.104

Lower prediction interval = 13.64, upper predication interval = 69.99 LCL = 40.69, UCL = 42.93

16.105

546


Lower prediction interval = -1.722, upper predication interval = 8.346. LCL = 3.144, UCL = 3.480.

16.106

Lower prediction interval = 26,400, upper predication interval = 205,033. LCL = 110,196, UCL = 121,237.

16.107

Lower prediction interval = -33,469, upper predication interval = 123,084. LCL = 42,645, UCL = 46,970.

16.108

Lower prediction interval = -1.021, upper predication interval = 5.224. LCL = 2.019, UCL = 2.184.

547


16.109

Lower prediction interval = 9.72, upper predication interval = 20.47. LCL = 14.94, UCL = 15.25.

16.110

Lower prediction interval = 11.39, upper predication interval = 22.13. LCL = 16.50, UCL = 17.03.

16.111

Lower prediction interval = -.06, upper predication interval = 7.84. LCL = 3.75, UCL = 4.03.

16.112 The relationship between the dependent and independent variables is too weak to provide accurate predictions.

16.113 a x i –5 –2 0 3

yi 15 9 7 6

x i2 25 4 0 9

548

yi2 225 81 49 36

x i yi –75 –18 0 18


Total

4 7 7

4 1 42 n

16 49 103 n

n

xi = 7

i 1

16 1 408

y i = 42

n

 x y = –52

x i2 = 103

i

i 1

i 1

16 7 –52 i

i 1

n n   x yi   n i (7)( 42  1  1   i 1 x i y i  i 1 s xy   = 6  1  52  6   20 .20 n n  1  i 1       

 

2     n    xi  n     2  1  x i2   i 1   = 1 103  (7)   18 .97 s 2x  n  1  i 1 n  6  1  6       

b1 

s xy s 2x

20 .2  1.065 18 .97

=

x

 x  7  1.167

y

 y  42  7.000

i

n

6

i

n

6

b 0  y  b1x = 7.000 – (–1.065)(1.167) = 8.253 The sample regression line is

ŷ = 8.253 – 1.065x 2     n   yi   n    1  (42 ) 2  1  i 1    2 2  yi  = b, c, &d s y    22 .80 408  n  1  i 1 n  6  1  6       

 s 2xy   (20 .20 ) 2  SSE  (n  1) s 2y  2  = (6  1) 22 .80   6.451   18 .97  s x    s 

xi –5 –2 0 3

SSE = n2

6.451  1.270 (Excel: s  = 1.268) 62

yi 15 9 7 6

e i  y i  ŷ i 1.42 –1.38 –1.253 .942

ŷ i  8.253  1.065 x 13.58 10.38 8.253 5.058

549

ei / s 1.118 –1.087 –.987 .742


4

4

7 1 There are no outliers.

3.993

.007

.0055

.798

.202

.159

16.114

xi 23 46 60 54 28 33 25 31 36 88 90 99

yi 9.6 11.3 12.8 9.8 8.9 12.5 12.0 11.4 12.6 13.7 14.4 15.9

ŷ i  9.107  .0582 x 10.45 11.78 12.60 12.25 10.74 11.03 10.56 10.91 11.20 14.23 14.35 14.87

16.115

xi 8.5 7.8 7.6 7.5 8.0 8.4 8.8 8.9 8.5 8.0

yi 115 111 185 201 206 167 155 117 133 150

ŷ = 475.2 – 39.17x 142.3 169.7 177.5 181.4 161.8 146.2 130.5 126.6 142.3 161.8

16.116 a & b

xi 42 34 25 35 37 38 31 33 19 29 38 28 29 36 18

e i  y i  ŷ i –.85 –.48 .20 –2.45 –1.84 1.47 1.44 .49 1.40 –.53 .06 1.03 e i  y i  ŷ i –27.3 –58.7 7.5 19.6 44.2 20.8 24.5 –9.6 –9.3 –11.8

ŷ = – 24.72 + .9675x 15.92 8.18 –.53 9.14 11.08 12.05 5.27 7.21 –6.34 3.34 12.05 2.37 3.34 10.11 –7.31

yi 18 6 0 –1 13 14 7 7 –9 8 8 5 3 14 –7

c

550

e i  y i  ŷ i 2.09 –2.18 .53 –10.14 1.92 1.96 1.73 –.21 –2.66 4.66 –4.05 2.63 –.34 3.89 .31


Plot of Residuals vs Predicted 6 4

Residuals

2 0 -10

-5

-2 0

5

10

15

20

-4 -6 -8 -10 -12 Predicted

16.117

xi 80 68 78 79 87 74 86 92 77 84

ŷ = –100,652 + 1,513x 20,388 2,232 17,362 18,875 30,979 11,310 29,466 38,544 15,849 26,440

yi 20,533 1,439 13,829 21,286 30,985 17,187 30,240 37,596 9,610 28,742

e i  y i  ŷ i 145 –793 –3,533 2,411 6 5,877 774 –948 –6,239 2,302

The histograms drawn below are of the standardized residuals, which make it easier to see whether the shape is extremely nonnormal. It also makes it easier to identify outliers. The shape of the resulting histogram is identical to the histogram of the residuals using the equivalent class limits.

16.118 b & c

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standardized Residuals

Because the histogram is approximately bell shaped the errors appear to be normally distributed. There are two residuals whose absolute value exceeds 2.0.

551


d Plot of Residuals vs Predicted 20 15

Residuals

10 5 0 -5 0

5

10

15

20

25

-10 -15 -20 Predicted

There is no indication of heteroscedasticity.

16.119 a

Frequency

Histogram 40 20 0 -3

-2

0

1

1

2

3

Standardized Residuals

The histogram is not bell shaped. However, residuals do not appear to be extremely nonnormal. c

552


Residuals

Plot of Residuals vs Predicted 60 50 40 30 20 10 0 -10180 -20 -30 -40 -50

190

200

210

220

230

240

3

4

Predicted

There is no clear indication of heteroscedasticity.

16.120

Frequency

Histogram 100 50 0 -3

-2

-1

0

1

2

Standard Residuals The error variable appears to be normally distributed.

Plot of Residuals versus Predicted 25 20 15

Residuals

10

5 0 -5 8

9

10

11

12

-10 -15 -20

Predicted

553

13

14

15


The variance of the error variable is constant.

16.121 b & c

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable appears to be normally distributed. There are three observations whose standardized residuals are greater than 2.0.

d Plot of Residuals vs Predicted 6

Residuals

4 2 0 -2

24

25

26

27

28

-4 -6 Predicted

554

29


The variance of the error variable is constant.

16.122

Frequency

Histogram 100 50 0 -3

-2

-1

0

1

2

More

Standard Residuals

The error variable appears to be normally distributed. Plot of Residuals vs Predicted 15 10

Residuals

5 0 -5

8

12

16

20

-10 -15 Predicted

The variance of the error variable is constant.

16.123b

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

Standard Residuals

The errors appear to be normally distributed. c There are no outliers. d

555

2

3


Plot of Residuals vs Predicted 40 30 Residuals

20 10 0 -10

20

30

40

50

60

70

80

-20 -30 Predicted

The variance of the error variable appears to decrease somewhat as the predicted values increase. However, the effect is not large enough to be a problem.

16.124

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

2

3

8000

9000

Standard Residuals

The error variable appears to be normally distributed. Plot of Residuals vs Predicted 8000 6000

Residuals

4000 2000 0 -20004000

5000

6000

7000

-4000 -6000 -8000 Predicted

There is no clear sign of heteroscedasticity. 556


16.125

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable appears to be normally distributed. Plot of Residuals vs Predicted 8 6

Residuals

4 2 0 -2

27

27.5

28

28.5

-4 -6 Predicted

The variance of the error variable is constant.

16.126

Frequency

Histogram 100 0 -3

-2

-1

0

1

Standard Residuals

The error variable appears to be normal.

557

2

3


Plot of Residuals vs Predicted 600

400 200 0 -200

400

500

600

700

800

900

-400

-600 -800

The variance of the error variable is constant.

16.127

Frequency

Histogram 100 50 0 -3

-2

-1

0

1

Standard Residuals

The error variable appears to be normal.

558

2

3

1000


Plot of Residuals vs Predicted 150

Residuals

100 50 0 200 -50

250

300

350

-100 -150 Predicted

The variance of the error variable is constant.

16.128

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable appears to be normal.

Plot of Residuals vs Predicted 6

Residuals

4 2 0 -2

14

18

22

-4 -6 Predicted

The variance of the error variable is constant.

559


16.129

Frequency

Histogram 100 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable appears to be normal. 25

Residuals vs Predicted

20 15 10 5 0 -5 54

56

58

60

62

64

-10 -15 -20 -25

The variance of the error variable is constant.

16.130

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

Standard Residuals

The error variable appears to be normal.

560

2

3

66


Plot of Residuals vs Predicted 3 2

Residuals

1 0 93.2 -1

93.4

93.6

93.8

94

94.2

94.4

-2 -3 -4 Predicted

The variance of the error variable is constant.

16.131 a b1 

s xy s 2x

=

74 .02  21 .33, b 0  y  b1x = 384.81 – 21.33(4.12) = 296.93 3.47

Regression line: ŷ = 296.93 + 21.33x (Excel: ŷ = 296.92 + 21.36x) b On average each additional ad generates 21.33 customers.

 2 s 2xy   (74 .02 ) 2  c SSE  (n  1) s y  2  = (26  1)18,552   424 ,326   3.47  s x    s 

SSE = n2

424 ,326  132 .97 (Excel: s  = 132.96). 26  2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t , n  2  t .05, 24  1.711 s b1 

t

s (n  1)s 2x

=

132 .97 (26  1)(3.47 )

 14 .28

21 .33  0 b1  1  1.49 (Excel: t = 1.50, p–value = .1479/2 = .0740.) There is not enough = 14 .28 s b1

evidence to conclude that the larger the number of ads the larger the number of customers. d R2 

s 2xy

(74 .02 ) 2 =  .0851 (Excel: R 2 = .0852). There is a weak linear relationship s 2x s 2y (3.47 )(18,552 )

between the number of ads and the number of customers. e The linear relationship is too weak for the model to produce predictions. 561


16.132 a b1 

s xy s 2x

=

936 .82  2.47 b 0  y  b1x = 395.21 – 2.47(113.35) = 115.24. 378 .77

Regression line: ŷ = 115.24 + 2.47x (Excel: ŷ = 114.85 + 2.47x) b b1 = 2.47; for each additional month of age, repair costs increase on average by $2.47.

b0 = 114.85 is the y-intercept. c R2 

s 2xy

(936 .82)2 =  .5659 (Excel: R 2 = .5659) 56.59% of the variation in repair s 2x s 2y (378 .77 )( 4,094 .79)

costs s explained by the variation in ages.

 s 2xy   (936 .82) 2  d SSE  (n  1) s 2y  2  = (20  1) 4,094 .79   33,777   378 .77  s x    s 

33,777  43 .32 (Excel: s  = 43.32). 20  2

SSE = n2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t .025,18  2.101 or t   t  / 2, n  2   t .025,18  2.101 s b1 

t

s (n  1)s 2x

=

43 .32

 .511

(20  1)(378 .77 )

2.47  0 b1  1  4.84 (Excel: t = 4.84, p–value = .0001. There is enough evidence to infer = .511 s b1

that repair costs and age are linearly related. e ŷ  b0  b1xg  115.24  2.47(120)  411.64 Prediction interval: ŷ  t  / 2,n  2 s  1 

= 411 .64  2.101(43 .32) 1 

2 1 (x g  x) (where t  / 2, n  2  t.025,18  2.101)  n (n  1)s 2x

1 (120  113 .35)2   411 .64  93 .54 20 (20  1)(378 .77 )

Lower prediction limit = 318.1, upper prediction limit = 505.2 (Excel: 318.1, 505.2)

16.133 a b1 

s xy s 2x

=

2538  .123 b 0  y  b1x = 318.60 –.123(300) = 281.7. 20,690

Regression line: ŷ = 281.7 + .123x (Excel: ŷ = 281.8 + .123x)

562


The slope is .123, which tells us that for each additional unit of fertilizer, corn yield increases on average by .123. The y-intercept is 281.7, which has no real meaning.

 s 2xy   (2,538 ) 2  b SSE  (n  1) s 2y  2  = (30  1) 5,230   142 ,641   20,690  s x    s 

142 ,641  71 .37 (Excel: s  = 71.38). 30  2

SSE = n2

H 0 : 1  0 H1 : 1  0 Rejection region: t  t  / 2, n  2  t.025, 28  2.048 or t   t  / 2, n  2   t.025, 28  2.048 s b1 

t

s (n  1)s 2x

=

71 .37 (30  1)( 20,690 )

 .0921

b1  1 .123  0  1.34 (Excel: t = 1.33, p–value = .1938. There is not enough evidence to = .0921 s b1

infer a linear relationship between amount of fertilizer and corn yield. c R2 

s 2xy s 2x s 2y

=

(2,538 ) 2  .0595 ( Excel: R 2 = .0595) 5.95% of the variation in corn yield (20,690 )(5,230 )

is explained by the variation in amount of fertilizer. d The model is too poor to be used to predict. 16.134a H 0 :   0

H1 :   0

A 1 Correlation 2 3 Tar and Nicotine 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

B

0.9766 21.78 23 0 1.7139 0 2.0687

r = .9766, t = 21.78, p–value = 0. There is sufficient evidence to infer that levels of tar and nicotine are linearly related.

b

H0 :   0

563


H1 :   0

A 1 Correlation 2 3 Nicotine and CO 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

B

0.9259 11.76 23 0 1.7139 0 2.0687

r = .9259, t = 11.76, p–value = 0. There is sufficient evidence to infer that levels of nicotine and carbon monoxide are linearly related.

H0 :   0

16.135

H1 :   0 Rejection region: t  t , n  2  t .05, 428  1.645 or

r

s xy s xs y

tr

255 ,877

(99 .11)( 2,152 ,602 ,614 )

n2 1 r

2

 (.5540 )

430  2 1  (.5540 ) 2

 .5540 (Excel: .5540)

 13 .77 (Excel: t = 13.77, p–value = 0). There is enough

evidence of a positive linear relationship. The theory appears to be valid.

H0 :   0

16.136

H1 :   0 Rejection region: t  t  / 2, n  2  t .025, 48  2.009 or t   t  / 2, n  2   t .025, 48  2.009

r

s xy sxsy

tr

13 .08 (90 .97 )(11 .84 )

n2 1 r

2

 (.3985 )

 .3985 (Excel: .3984)

50  2 1  (.3985 ) 2

 3.01 (Excel: t = 3.01, p–value = .0042). There is enough

evidence of a linear relationship. The theory appears to be valid.

16.137

H0 :   0 H1 :   0

564


A B C 1 Correlation 2 3 Fund and Gold 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

D

0.7929 6.63 26 0 1.7056 0 2.0555

t = 6.63, p-value = 0. There is enough evidence to conclude that there is a positive linear relationship between the value of the fund and the price of gold.

16.138

H0 :   0 H1 :   0

A B C 1 Correlation 2 3 Time and Sales 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

D

0.2791 1.67 33 0.0522 1.6924 0.1044 2.0345

t = 1.67, p-value = .0522. There is not enough evidence to infer that when the times between movies increase so do sales.

565


16.139 a 195,0 190,0 185,0

Height

180,0 175,0 170,0 165,0 160,0 155,0 150,0 4,0

5,0

6,0

7,0

8,0

9,0

10,0

Index Finger Length

b.

H0 :   0 H1 :   0

t = 7.60, p-value = 0. There is enough evidence to infer that index finger length and height are linearly related.

566


c.

Lower prediction limit = 150.5, Upper prediction limit = 182.5

16.140

H0 :   0 H1 :   0

A 1 Correlation 2 3 Hours and GPA 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

B

-0.5748 -9.88 198 0 1.6526 0 1.9720

t = –9.88, p–value = 0. There is enough evidence of a linear relationship between time spent at part time jobs and grade point average.

16.141

H0 :   0 H1 :   0

567


A 1 Correlation 2 3 Times and Amount 4 Pearson Coefficient of Correlation 5 t Stat 6 df 7 P(T<=t) one tail 8 t Critical one tail 9 P(T<=t) two tail 10 t Critical two tail

B

0.7976 29.51 498 0 1.6479 0 1.9647

t = 29.51, p–value = 0. There is overwhelming evidence of a linear relationship between listening times and amounts spent on music.

16.142 The intervals between values must be constant, which is arguable.

16.143 H0:β1 = 0 H1:β1 > 0

t = 5.34, p-value = 0. There is enough evidence to infer that older people are more likely to be conservative. 16.144 H0:β1 = 0 H1:β1 > 0

t = 2.30, p-value = .0218/2 = .0109. There is enough evidence to infer that as income increases people are more likely to be conservative. 16.145 H0:β1 = 0 H1:β1 < 0

t = -5.05, p-value = 0. There is enough evidence to infer that as education increases people are more likely to be liberal.

568


16.146 H0:β1 = 0 H1:β1 > 0

t = 15.67, p-value = 0. 16.147 H0:β1 = 0 H1:β1 > 0

t = 12.43, p-value = 0. 16.148 H0:β1 = 0 H1:β1 > 0

t = 15.51, p-value = 0. 16.149 H0:β1 = 0 H1:β1 > 0

t = 15.60, p-value = 0.

16.150 In general liberals want government to act and conservatives want no government action. However, the coefficients of determination indicate weak relationships.

Case 16.1 a

569


A B C D E F 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9896 5 R Square 0.9794 6 Adjusted R Square 0.9787 7 Standard Error 355.6 8 Observations 32 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 1 180,066,717 180,066,717 1424 0.0000 13 Residual 30 3,793,289 126,443 14 Total 31 183,860,006 15 16 Coefficients Standard Error t Stat P-value 17 Intercept 16.23 114.70 0.14 0.8884 18 A-Park 0.693 0.0184 37.74 0.0000

The regression equation is ŷ = 16.23 + .693x. This equation was used to predict museum attendance when it was closed (observations 33 to 179). The sum of the predictions is 785,009.

b

570


A B C D E F 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.9909 4 Multiple R 0.9819 5 R Square 0.9811 6 Adjusted R Square 572.9 7 Standard Error 8 Observations 26 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 1 426,295,375 426,295,375 1299 0.0000 13 Residual 24 7,875,875 328,161 14 Total 25 434,171,250 15 16 Coefficients Standard Error t Stat P-value 17 Intercept 459.5 295.4 1.56 0.1330 18 A-Park 0.970 0.0269 36.04 0.0000

The regression equation is ŷ = 459.5 + .970x. This equation was used to predict museum attendance when it was closed (observations 33 to 179). The sum of the predictions is 1,162,994. c The predicted lost revenue should be based on the regression using the first 32 weeks. Multiply 785,009 by the price of tickets and subtract fixed costs to produce the amount the insurance company should pay the museum.

Case 16.2 Regression using the best 6 OACs:

A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.4883 5 R Square 0.2385 6 Adjusted R Square 0.2363 7 Standard Error 0.8295 8 Observations 363 9 10 ANOVA 11 df SS 12 Regression 1 77.78 13 Residual 361 248.39 14 Total 362 326.17 15 16 Coefficients Standard Error 17 Intercept -5.35 1.31 18 Best-6 0.15 0.01

571

D

E

MS 77.78 0.69

F Significance F 113.04 0.0000

t Stat P-value -4.08 0.0001 10.63 0.0000

F


t = 10.63, p–value = 0. There is evidence of a linear relationship between the average of the best 6 OACs and university GPA. R 2 = .2385, s = .8295

Regression using the best 4 OACs plus English and calculus: A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.5924 5 R Square 0.3509 6 Adjusted R Square 0.3491 7 Standard Error 0.7658 8 Observations 363 9 10 ANOVA 11 df SS 12 Regression 1 114.46 13 Residual 361 211.71 14 Total 362 326.17 15 16 Coefficients Standard Error 17 Intercept -3.32 0.85 18 B4+E+C 0.14 0.01

D

E

MS 114.46 0.59

F Significance F 195.17 0.0000

t Stat -3.89 13.97

P-value 0.0001 0.0000

t = 13.97, p–value = 0; there is evidence of a linear relationship between the average of the best 4 OACs plus English and calculus and university GPA. R 2 = .3509, s = .7658.

The second model fits better (higher coefficient of determination and lower standard error of estimate) and as such is likely to be a better predictor of university GPA.

572

F


Chapter 17 17.1

a ŷ  51.39  .700 x1  .679 x 2  .378 x3 b The standard error of estimate is s  = 40.24. It is an estimate of the standard deviation of the error variable. c The coefficient of determination is R 2 = .2425; 24.25% of the variation in prices is explained by the model. d The coefficient of determination adjusted for degrees of freedom is .2019. It differs from R 2 because it includes an adjustment for the number of independent variables. e

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 5.97, p-value = .0013. There is enough evidence to conclude that the model is valid. f b1 = .700; for each addition thousand square feet the price on average increases by .700 thousand dollars provided that the other variables remain constant.

b2 = .679; for each addition tree the price on average increases by .679 thousand dollars provided that the other variables remain constant.

b3 = –.378; for each addition foot from the lake the price on average decreases by .378 thousand dollars provided that the other variables remain constant. g

H0: βi = 0 H1: βi ≠ 0 593


Lot size: t = 1.25, p-value = .2156 Trees: t = 2.96, p-value = .0045 Distance: t = –1.94, p-value = .0577 Only for the number of trees is there enough evidence to infer a linear relationship with price. h

We predict that the lot in question will sell for between $35,500 and $172,240 i

Estimated average price lies between $39,290 and $90,300.

594


17.2

a ŷ  13.01  .194 x1  1.11x 2 b The standard error of estimate is s  = 3.75. It is an estimate of the standard deviation of the error variable. c The coefficient of determination is R 2 = .7629; 76.29% of the variation in final exam marks is explained by the model. d

H0: β1 = β2 = 0 H1: At least one  i is not equal to zero

F = 43.43, p-value = 0. There is enough evidence to conclude that the model is valid. e b1 = .194; for each addition mark on assignments the final exam mark on average increases by .194 provided that the other variable remains constant.

b2 = 1.112; for each addition midterm mark the final exam mark on average increases by 1.112 provided that the other variable remains constant. f

H 0 : 1  0 H1 : 1  0

t = .97, p-value = .3417. There is not enough evidence to infer that assignment marks and final exam marks are linearly related. g

H0 : 2  0 H1 :  2  0

t = 9.12, p-value = 0. There is sufficient evidence to infer that midterm marks and final exam marks are linearly related. 595


h.

Pat’s final exam mark is predicted to lie between 23 and 39 i. Pat’s predicted final grade: LCL = 12 + 14 + 23 = 49, UCL = 12 + 14 + 39 = 65

17.3a

b The standard error of estimate is s  = 40.13. It is an estimate of the standard deviation of the error variable. c The coefficient of determination is R 2 = .8935; 89.35% of the variation in monthly sales of drywall is explained by the model.

596


d

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 39.86, p-value = 0. There is enough evidence to conclude that the model is valid. e b1 = 4.76; for each addition building permit monthly sales on average increase by 4.76 hundred sheets provided that the other variables remain constant.

b2 = 16.99; for each addition one point increase in mortgage rates monthly sales on average increase by 16.99 hundred sheets provided that the other variables remain constant.

b3 = –10.53; for each one percentage point increase in the apartment vacancy rate monthly sales decrease on average by 10.53 hundred sheets provided that the other variables remain constant.

b4 = 1.31; for each one percentage point increase in the office vacancy rate monthly sales increase on average by 1.31 hundred sheets provided that the other variables remain constant. f

H 0 : i  0 H1 :  i  0

Permits: t = 12.06, p-value = 0 Mortgage: t = 1.12, p-value = .2764 A Vacancy: t = –1.65, p-value = .1161 O Vacancy: t = .47, p-value = .6446 Only the number of building permits is linearly related to monthly sales. g

Next month’s drywall sales are predicted to lie between 16,710 and 35,290 sheets.

597


17.4a

b b1 = .666; for each additional minor league home run the number of major league home runs increases on average by .666 provided that the other variables remain constant.

b2 = .136; for each additional year of age the number of major league home runs increases on average by .14 provided that the other variables remain constant.

b3 = 1.18; for each additional year as a professional the number of major league home runs increases on average by 1.18 provided that the other variables remain constant. c s  = 6.99 and R 2 = .3511; the model's fit is not very good. d

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 22.01, p-value = 0. There is enough evidence to conclude that the model is valid. e

H 0 : i  0 H1 :  i  0

Minor league home runs: t = 7.64, p-value = 0 Age: t = .26, p-value = .7961 Years professional: t = 1.75, p-value = .0819 At the 5% significance level only the number of minor league home runs is linearly related to the number of major league home runs.

598


f

We predict that the player will hit between 9.86 (rounded to 10) and 38.76 (rounded to 39) home runs. g

It is estimated that the average player will hit between 14.66 and 24.47 home runs.

599


17.5

a The regression equation is ŷ = 6.06 –.00781x 1 + .603x 2 –.0702x 3 b s  = 1.92, R 2  .7020, F = 36.12, p-value = 0. The model is valid and the fit is reasonably good. c

H 0 : i  0 H1 :  i  0

Age: t = –.118, p-value = .9069 Years: t = 6.25, p-value = 0 Pay: t = –1.34, p-value = .1864 Only the number of years with the company is linearly related to severance pay d.

600


95% prediction interval: Lower prediction limit = 5.64, upper prediction limit = 13.50. The offer of 5 weeks severance pay falls below the prediction interval and thus Bill is correct.

17.6

b The coefficient of determination is R 2 = .2882; 28.82% of the variation in university GPAs is explained by the model. c

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 12.96, p-value = 0. There is enough evidence to conclude that the model is valid. d

H 0 : i  0 H1 :  i  0

High school GPA: t = 6.06,, p-value = 0 SAT: t = .942, p-value = .3485 Activities: t = .722, p-value = .4720 At the 5% significance level only the high school GPA is linearly related to the university GPA

601


e

We predict that the student's GPA will fall between 4.45 and 12.00 (12 is the maximum). f

The mean GPA is estimated to lie between 6.90 and 8.22.

602


17.7

a The regression equation is ŷ = 12.31 + .570x 1 + 3.32x 2 + .732x 3 b The coefficient of determination is R 2 = .1953; 19.53% of the variation in sales is explained by the model. The coefficient of determination adjusted for degrees of freedom is .0803. The model fits poorly. c The standard error of estimate is s  = 2.59. It is an estimate of the standard deviation of the error variable d

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 1.70, p-value = .1979. There is not enough evidence to conclude that the model is valid. e

H 0 : i  0 H1 :  i  0

Direct: t = .331, p-value = .7437 Newspaper: t = 2.16, p-value = .0427 Television: t = .374, p-value = .7123 Only expenditures on newspaper advertising is linearly related to sales.

603


f&g

f We predict that sales will fall between $12,270 and $24,150. g We estimate that mean sales will fall between $15,700 and $20,730. h The interval in part f predicts one week’s gross sales, whereas the interval in part h estimates the mean weekly gross sales.

17.8 a

b

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 29.80, p-value = 0. There is enough evidence to conclude that the model is valid.

604


c b1= .0019; for each additional square foot the amount of garbage increases on average by .0019 pounds holding the other variables constant. b2 = 1.10; for each additional child in the home the amount of garbage increases on average by 1.10 pounds holding the other variables constant. b3= 1.04; for each additional adult at home during the day the amount of garbage increases on average by 1.04 holding the other variables constant. d

H 0 : i  0 H1 :  i  0

House size : t = 3.21, p-value = .0014 Number of children: t = 7.84 p-value = 0 Number of adults at home: t = 4.48, p-value = 0 All three independent variable are linearly related to the amount of garbage.

17.9a

a ŷ = 35.68 + .247x 1 + .245x 2 + .133x 3 b

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 6.66, p-value = .0011. There is enough evidence to conclude that the model is valid. c b1 = .247; for each one percentage point increase in the proportion of teachers with mathematics degrees the test score increases on average by .247 provided the other variables are constant.

605


b2 = .245; for each one year increase in mean age test score increases on average by .245 provided the other variables are constant .

b3 = .135; for each one thousand dollar increase in salary test score increases on average by .135 provided the other variables are constant.

H 0 : i  0 H1 :  i  0 Proportion of teachers with at least one mathematics degree: t = 3.54, p-value = .0011 Age: t = 1.32, p-value = .1945 Income: t = .87, p-value = .3889. The proportion of teachers with at least one mathematics degree is linearly related to test scores. The other two variables appear to be unrelated.

d The school's test score is predicted to fall between 49.02 and 81.02.

606


17.10a

b

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 67.97, p-value = 0. There is enough evidence to conclude that the model is valid. c b1 = .451; for each one year increase in the mother's age the customer's age increases on average by .451 provided the other variables are constant (which may not be possible because of the multicollinearity).

b2 = .411; for each one year increase in the father's age the customer's age increases on average by .411 provided the other variables are constant.

b3 = .0166; for each one year increase in the grandmothers' mean age the customer's age increases on average by .0166 provided the other variables are constant.

b 4 = .0869; for each one year increase in the grandfathers' mean age the customer's age increases on average by .0869 provided the other variables are constant.

H 0 : i  0 H1 :  i  0 Mothers: t = 8.27, p-value = 0 Fathers: t = 8.26, p-value = 0 Grandmothers: t = .250, p-value .8028 Grandfathers: t = 1.32, p-value = .1890

607


The ages of mothers and fathers are linearly related to the ages of their children. The other two variables are not. d

The man is predicted to live to an age between 65.54 and 77.31 e

The mean longevity is estimated to fall between 68.75 and 74.66.

608


17.11

Assessing the Model:

s  = 7.01 and R 2 = .7209; the model fits well. Testing the validity of the model: H0: β1 = β2 = 0 H1: At least one  i is not equal to zero F = 60.70, p-value = 0. There is enough evidence to conclude that the model is valid. Drawing inferences about the independent variables:

H 0 : i  0 H1 :  i  0 Evaluations: t = .598, p-value = .5529 Articles: t = 8.08, p-value = 0. The number of articles a professor publishes is linearly related to salary. Teaching evaluations are not.

609


17.12

a ŷ = –28.43 + .604x 1 + .374x 2 b s  = 7.07 and R 2 = .8072; the model fits well. c b1 = .604; for each one additional box, the amount of time to unload increases on average by .604 minutes provided the weight is constant.

b2 = .374; for each additional hundred pounds the amount of time to unload increases on average by .374 minutes provided the number of boxes is constant.

H 0 : i  0 H1 :  i  0 Boxes: t = 10.85, p-value = 0 Weight: t = 4.42, p-value = .0001 Both variables are linearly related to time to unload.

610


d&e

d It is predicted that the truck will be unloaded in a time between 35.16 and 66.24 minutes. e The mean time to unload the trucks is estimated to lie between 44.43 and 56.96 minutes.

17.13

b

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 18.17, p-value = 0. There is enough evidence to conclude that the model is valid.

611


e

H 0 : i  0 H1 :  i < 0 (for beliefs 1 and 4) H1 :  i > 0 for beliefs 2 and 3)

Belief 1: t = –3.26, p-value = .0016/2 = .0008 Belief 2: t = 1.16, p-value = .2501/2 = .1251 Belief 3: t = .417, p-value = .6780/2 = .3390 Belief 4: t = –2.69, p-value = .0085/2 = .0043 There is enough evidence to support beliefs 1 and 4. There is not enough evidence to support beliefs 2 and 3.

17.14 a

b

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 24.48, p-value = 1.64E-11 .≈ 0. There is sufficient evidence to conclude that the model is valid. c

H 0 : i  0 H1 :  i  0

Undergraduate GPA: t = .524, p-value = .6017 GMAT: t = 8.16, p-value = 0

612


Work experience: t = 3.00, p-value = .0036 Both the GMAT and work experience are linearly related to MBA GPA

17.15

ŷ = 3.02 + .0289 AGE + .2461 EDUC + 3.75X10-6 INCOME b.

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 41.92, p-value = 0. There is sufficient evidence to conclude that the model is valid. c.

H0: βi = 0 H1: βi ≠ 0

Variable

t-statistic

p-value

AGE

6.93

5.99E-12 ≈ 0

EDUC

8.07

1.36E-15 ≈ 0

INCOME

1.72

.0858

AGE and EDUC are linearly related to DEFINITE.

613


17.16

ŷ = 6.365 + .135 DAYS1 + .036 DAYS2 + .060 DAYS3 + .107 DAYS4 + .142 DAYS5 + .134 DAYS6 b.

H0: β1 = β2 = β3 = β4 = β5 = β6 = 0 H1: At least one  i is not equal to zero

F = 11.72, p-value = 0. There is enough evidence to infer that the model is valid. c.

H0: βi = 0 H1: βi ≠ 0

Variable

t-statistic

p-value

DAYS1

3.33

.0009

DAYS2

.81

.4183

DAYS3

1.41

.1582

DAYS4

3.00

.0027

DAYS5

3.05

.0024

DAYS6

3.71

.0002

Only DAYS2 and DAYS3 are not linearly related to DEFINITE.

614


7.17

b.

H0: β1 = β2 = β3 = β4 = β5 = β6 = 0 H1: At least one  i is not equal to zero

F = 17.14, p-value = 3.03E-13 ≈ 0. There is enough evidence to conclude that the model is valid. c.

H 0 : i  0 H1 :  i  0

Variable

t-statistic

p-value

Number

-6.07

0

Nearest

2.60

.0108

Office space

5.80

0

Enrollment

1.59

.1159

Income

2.96

.0039

Distance

-1.26

.2107

The intercept is b 0  38.14. This is the average operating margin when all the independent variables are zero.

615


The relationship between operating margin and the number of motel and hotel rooms within 3 miles is described by b1  −.0076. Changing the units we can interpret b1 to say that for each additional 1,000 rooms, the margin decreases by 7.6%. The coefficient b 2  1.65 specifies that for each additional mile that the nearest competitor is to a La Quinta Inn, the average operating margin increases by 1.65%, assuming the constancy of the other independent variables. The relationship between office space and operating margin is expressed by b 3  .020. For every extra 100,000 square feet of office space, the operating margin increases on average by 2.0%.

The relationship between operating margin and college and university enrollment is described by

b 4  .21, which we interpret to mean that for each additional thousand students the average operating margin increases by .21% when the other variables are constant. The relationship between operating margin and median household income is described by b 5  .41. For each additional thousand dollar increase in median household income, the average operating margin increases by .41%, holding all other variables constant. This statistic suggests that motels in more affluent communities have higher operating margins.

The last variable in the model is distance to the downtown core. Its relationship with operating margin is described by b 6  −.23. This tells us that for each additional mile to the downtown center, the operating margin decreases on average by .23%, keeping the other independent variables constant. It may be that people prefer to stay at motels that are closer to town centers. e. & f.

616


e. Lower prediction limit n= 25.40, Upper prediction limit = 48.79 f. LCL = 32.97, UCL = 41.21

17.18

b.

H0: β1 = β2 = 0 H1: At least one  i is not equal to zero

F = 284.0, p-value = 0. There is sufficient evidence to conclude that the model is valid.

c

H 0 : i  0 H1 :  i  0

PAEDUC: t = 10.16, p-value = 0 MAEDUC: t = 8.49, p-value = 0 Both variables are linearly related to EDUC. d. b1: For each additional year of the father’s education the son or daughter’s years of education increases on average by .212 years holding the other variable constant. b2: For each additional year of the mother’s education the son or daughter’s years of education increases on average by .194 years holding the other variable constant.

17.19

617


b.

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 11.51, p-value = 0. There is enough evidence to conclude that the model is valid. c.

H 0 : i  0 H1 :  i  0

AGE: t = 3.62, p-value = 0 RINCOME: t = 3.33, p-value = .0001 EDUC: t = .49, p-value = .6214 HRS1: t = 1.44, p-value = .1515 Only AGE and RINCOME are linearly related to EQWLTH d. R2 = .0503; 5.03% of the variation in EQWLTH is explained by the model.

618


17.20

b.

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 16.41, p-value = 0. There is enough evidence to conclude that the model is valid. c.

H 0 : i  0 H1 :  i  0

AGE: t = 2.74, p-value = .0062 EDUC: t = -6.14, p-value = 0 HRS1: t = -3.92, p-value = 0 CHILDS: t = .11, p-value = .9108 AGE, EDUC, and HRS1 are linearly related to TVHOURS d. R2 = .0618; 6.18% of the variation in TVHOURS is explained by the model.

619


17.21

b.

H0: β1 = β2 = β3 = β4 = 0 H1: At least one  i is not equal to zero

F = 2.27, p-value = .0060. There is not enough evidence to conclude that the model is valid. c.

H 0 : i  0 H1 :  i  0

AGE: t = 2.15, p-value = .0316 EDUC: t = .43, p-value = .6703 RINCOME: t = .89, p-value = .3716 HRS1: t = .65, p-value = .5189 Only AGE is linearly related to HELPPOOR d. R2 = .0105; 1.05% of the variation in HELPPOOR is explained by the model.

620


17.22

ŷ = 56,326 - 883.7 AGE - 4503 EDUC b.

H0: β1 = β2 = 0 H1: At least one  i is not equal to zero

F = 140.30, p-value = 0. There is enough evidence to infer that the model is valid. c.

H0: β1 = 0 H1: β1 > 0

t = 8.69, p-value = 0. There is enough evidence to infer that more education leads to more income. d. R2 = .1895; 18.95% of the variation in income is explained by the model.

17.23

ŷ = 7612 -37.88 AGE – 72.39 EDUC + .0190 INCOME 621


b.

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 54.73, p-value = 0. There is enough evidence to infer that the model is valid. c.

AGE: t = -5.77, p-value = 0. EDUC: t = -.180, p-value = .0722 INCOME: t = 8.74, p-value = 0.

AGE and INCOME are linearly related to FOODHOME.

17.24

ŷ = 2206 -21.95 AGE + 4.52 EDUC + .0140 INCOME b.

H0: β1 = β2 = β3 = 0 H1: At least one  i is not equal to zero

F = 62.20, p-value = 0. There is enough evidence to infer that the model is valid. c.

AGE: t = -4.95, p-value = 0. EDUC: t = .17, p-value = .8680 INCOME: t = 9.50, p-value = 0.

AGE and INCOME are linearly related to FOODAWAY.

622


17.25 a

Frequency

Histogram 40 20 0 -2

-1

0

1

2

3

-3

The normality requirement is satisfied. b Plot of Residuals vs predicted

Residuals

100 50 0 40

60

80

100

120

140

-50 -100 Predicted

The variance of the error variable appears to be constant.

17.26

A 1 2 Lot size 3 Trees 4 Distance

B Lot size

C Trees

1 0.2857 -0.1895

1 0.0794

D Distance

1

There does not appear to be a multicollinearity problem. The t–tests are valid.

623


17.27b

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

2

3

Standard Residuals

The normality requirement has not been violated. c

Residuals

Plot of Residuals vs Predicted 10 8 6 4 2 0 -2 20 -4 -6 -8 -10

30

40

50

Predicted

The variance of the error variable appears to be constant. d

A 1 2 Assignment 3 Midterm

B C Assignment Midterm 1 0.1037 1

The lack of multicollinearity means that the t–tests were valid.

624

60


17.28a

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable may be normally distributed. b Plot of Residuals vs Predicted 100

Residuals

50 0 0

100

200

300

400

500

-50 -100 Predicted

The variance of the error variable is constant. c

1 2 3 4 5

A

B Permits

Permits Mortgage A Vacancy O Vacancy

1 0.0047 -0.1505 -0.1027

C D Mortgage A Vacancy 1 -0.0399 -0.0332

1 0.0652

Multicollinearity is not a problem.

17.29a

A B 1 Minor HR 2 Minor HR 1 3 Age 0.0354 4 Years Pro -0.0392

C Age

D Years Pro

1 0.7355

1

625

E O Vacancy

1


Age and years as a professional are highly correlated. The correlations of the other combinations are small. b The t–tests may not be valid.

17.30 a

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable is approximately normally distributed. b Plot of Residuals vs Predicted 6

Residuals

4 2 0 -2

0

5

10

15

20

-4 -6 Predicted

The variance of the error variable is constant. c

A 1 2 Age 3 Years 4 Pay

B Age

C Years

1 0.8080 0.1725

1 0.2610

D Pay

1

The correlation between age and years is high indicating that multicollinearity is a problem.

626


17.31

Frequency

Histogram 50 0 -3

-2

-1

0

1

2

Standard Residuals

Residuals

Plot of Residuals vs Predicted 5 4 3 2 1 0 -1 3 -2 -3 -4 -5

5

7

9

Predicted

The error variable is approximately normally distributed and the variance is constant.

17.32a

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

Standard Residuals

627

2

3


Plot of Residuals vs Predicted 6

Residuals

4 2 0 -2 15

17

19

21

23

-4 -6 Predicted

The error variable is approximately normal. However, the variance is not constant. b

A 1 2 Direct 3 Newspaper 4 Television

B Direct

C D Newspaper Television

1 -0.1376 -0.1246

1 0.1468

1

Multicollinearity is not a problem. c There is one observation whose standardized residual exceeds 2.0 that should be checked

17.33

Frequency

Histogram 200 100

0 -3

-2

-1

0

1

Standard Residuals

The error appears to be normally distributed.

628

2

3


15

Plot of Residuals vs Predicted

10

5 0 8

10

12

14

16

18

20

22

-5

-10 -15

There are indications that the error grows smaller as the predicted value increases. However, overall the requirement of constant variance may be valid.

17.34 a

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

2

3

Standard Residuals

The normality requirement is satisfied. Plot of Residuals vs Predicted

Residuals

20 10 0 -10 50

55

60

65

70

-20 -30 Predicted

The variance of the error variable is constant.

629

75


c

A B 1 Math Degree 2 Math Degree 1 3 Age 0.0766 4 Income 0.0994

C Age

D Income

1 0.5698

1

The correlation between age and income is high. Multicollinearity is a problem.

17.35a

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

More

Standard Residuals

The normality requirement is satisfied. b Plot of Residuals vs Predicted 10 8 6

Residuals

4 2 0 -2 60

` 65

70

75

80

-4 -6 -8 -10 Predicted

The variance of the error variable is constant. c

630

85


A 1 2 3 4 5

B Mother

Mother Father Gmothers Gfathers

C Father

1 0.2766 0.4343 0.3910

1 0.2409 0.3752

D E Gmothers Gfathers

1 -0.0077

1

The correlations are large enough to cause problems with the t–tests.

17.36

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standard Residuals

The error variable appears to be normal. The error variable's variance appears to be constant. The required conditions are satisfied.

17.37

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

Standard Residuals

631

2

3


Plot of Residuals vs Predicted

Residuals

20 10 0 20

40

60

80

100

-10 -20 Predicted

c The error variable appears to be normally distributed. The variance of the errors appears to be constant.

17.38a

Frequency

Histogram 40 20 0 -3

-2

-1

0

1

2

3

Standard Residuals

Plot of Residuals vs Predicted

Residuals

10 5 0 -5

-5

0

5

10

-10 Predicted

The errors appear to be normally distributed. The variance of the errors is not constant.

632


b

Lottery

Education

Age

1 -0.6202 0.1767 -0.0230 -0.5891

1 -0.1782 0.1073 0.7339

1 0.1072 -0.0418

Lottery Education Age Children Income

Children

Income

1 0.0801

1

There is a strong correlation between income and education. The t–tests of these two coefficients may be distorted.

Frequency

17.39

25 20 15 10 5 0 -2

-1,5

-1

-0,5

0

0,5

1

1,5

2

Residuals The histogram of the residuals is bell shaped.

2

1,5 1

Residuals

0,5 0 -0,5

6

7

8

9

-1 -1,5

-2 -2,5

Predicted

There is no evidence of heteroscedasticity. The requirements are satisfied.

633

10

11


17.40

Frequency

30 20 10 0 -10

-7

-4

-1

2

5

8

11

14

Residuals

The histograms is somewhat bimodal. There residuals may not be normally distributed.

15 10

Residuals

5 0

30

35

40

45

-5 -10 -15

Predicted

There is no evidence of heteroscedasticity.

17.41 Histogram of standardized residuals

634

50

55

60


600 Frequency

500 400 300 200 100 0 -3 -2,5 -2 -1,5 -1 -0,5 0

0,5

1

1,5

2

2,5

3

Std Res The error term appears to be negatively skewed.

Plot of residuals vs predicted 6 4

Residuals

2 0 0

2

4

6

8

10

-2 -4 -6 -8 -10

Predicted There is some form of heteroscedasticity.

17.42 Histogram of standardized residuals

Histogram Frequency

400 300 200 100 0 -3 -2,5 -2 -1,5 -1 -0,5 0 0,5 1 1,5 2 2,5 3 3,5 Std Res

635

12


The error term appears to be negatively skewed.

Plot of residuals versus predicted 6 4

Residuals

2 0 0

2

4

6

8

10

12

-2

-4 -6 -8 -10

Predicted

There is some form of heteroscedasticity.

17.43a

b. The correlations between RINCOME and AGE, EDUC, and HRS1 may be large enough to produce a multicollinearity effect.

636


c. AGE (r = .0855, p-value = .0121) and RINCOME (r = .0695, p-value = .0416) are linearly related to HELPPOOR. In the analysis in Exercise 17.21 only AGE was significant.

17.44

The correlation between PAEDUC and MAEDUC is .6934. Multicollinearity is a problem in Exercise 17.18.

17.45a.

The correlations between RINCOME AGE, EDUC and HRS1 are large enough to indicate the presence of multicollinearity. b. All four independent variables are linearly related to EQWLTH. In the analysis in Exercise 17.19 only RINCOME and AGE were linearly related to EQWLTH.

637


17.46a.

The correlations between AGE and CHILDS and EDUC and CHILDS indicate that multicollinearity may be a problem. b. AGE (r = .0883, p-value = .0051), EDUC (r = -.1971, p-value = 0), HRS1 (r = -.1322, p-value = 0) are linearly related to TVHOURS. These are the same independent variables that were shown to be linearly related to TVHOURS in the regression analysis.

17.47 d L = 1.12, d U = 1.66. There is evidence of positive first–order autocorrelation. 17.48 d L = 1.16, d U = 1.59, 4 – d U = 2.41, 4 – d L = 2.84. There is evidence of negative first– order autocorrelation. 17.49 d L = .95, d U = 1.89, 4 – d U = 2.11, 4 – d L = 3.05. There is evidence of first–order autocorrelation.

17.50 d L = 1.46, d U = 1.63. There is evidence of positive first–order autocorrelation. 17.51 d L = 1.41, d U = 1.64, 4 – d U = 2.36, 4 – d L = 2.59. The test is inconclusive 17.52 4 – d U = 4 – 1.73 = 2.27, 4 – d L = 4 – 1.19 = 2.81. There is no evidence of negative first– order autocorrelation.

638


17.53 a The regression equation is ŷ = 303.3 + 14.94 x1 + 10.52 x 2 b Plot of Residuals vs Time 1000

Y

500 0 0

20

40

60

80

100

120

-500 -1000 Tim e

The graph indicates that autocorrelation exists. c

A B C 1 Durbin-Watson Statistic 2 3 d = 0.7749 d L = 1.63, d U = 1.72, 4 – d U = 2.37, 4 – d L = 2.28. There is evidence of autocorrelation. d The model is y   0 + 1 x1 +  2 x 2 + 3t +  The regression equation is ŷ = 10.00 + 6.78 x1 + 9.37 x 2 + 9.64t e Plot of Residuals vs Time 600 400 200

Y

0 -200

0

20

40

60

80

-400 -600 -800 Tim e

A B C 1 Durbin-Watson Statistic 2 3 d = 2.1237

639

100

120


There is no evidence of autocorrelation. First model: s  = 348.7 and R 2 = .2825. Second model: s  = 208.1 and R 2 = .7471 The second model fits better.

17.54 a The regression equation is ŷ = 2260 + .423x b Plot of Residuals vs Time 1500 1000

Y

500 0 -500 0 -1000 -1500

10

20

30

40

50

60

-2000 -2500 Tim e

There appears to be a strong autocorrelation. c

A B C 1 Durbin-Watson Statistic 2 3 d = 0.7859 d L  1.50, d U  1.59, 4  d U  2.41, 4  d L   2.50. There is evidence of first–order autocorrelation. d The model is y   0 + 1 x +  2 t +  The regression equation is ŷ = 446.2 + 1.10x + 38.92t e

640


Plot of Residuals vs Time 1000

Sales

500 0 0

10

20

30

40

50

60

-500 -1000 -1500 Tim e

A B C 1 Durbin-Watson Statistic 2 3 d = 2.2631 There is no evidence of autocorrelation. First model: s  = 709.7 and R 2 = .0146. Second model: s  = 413.7 and R 2 = .6718. The second model fits better.

17.55

A B C 1 Durbin-Watson Statistic 2 3 d = 1.755 d L = 1.01, d U = 1.78. There is no evidence of positive first–order autocorrelation.

17.56

A B C 1 Durbin-Watson Statistic 2 3 d = 2.2003 d = 2.2003; d L = 1.30, d U = 1.46, 4 – d U = 2.70, 4 – d L = 2.54. There is no evidence of first– order autocorrelation.

17.57 a ŷ = 898.0 + 11.33x b

641


Frequency

Histogram 10 5 0 -3

-2

-1

0

1

2

3

Standard Residuals

Plot of Residuals vs Predicted 200

Residuals

150 100 50 0 -50900

950

1000

1050

1100

-100 -150 Predicted

The normality requirement is satisfied. However, the constant variance requirement is not.

A B C 1 Durbin-Watson Statistic 2 3 d = 1.0062 d L = 1.24, d U = 1.43, 4 – d U = 2.76, 4 – d L = 2.57. There is evidence of first–order autocorrelation. c The problem is that the errors are not independent. We add a time variable (week number) to the model. Thus, the new model is y =  0 + 1 x +  2 t +  . The regression equation is ŷ = 960.6 + 13.88x – 7.69t

Frequency

Histogram 10 5 0 -3

-2

-1

0

1

Standard Residuals

642

2

3


Plot of Residuals vs Years 100

Tires

50 0 800

900

1000

1100

1200

-50 -100 -150 Years

A B C 1 Durbin-Watson Statistic 2 3 d = 1.9012 d L = 1.15, d U = 1.54, 4 – d U = 2.85, 4 – d L = 2.46. There is no evidence of first–order autocorrelation. d s  = 48.55 and R 2 = .7040 Snowfall: b1  13 .88; for each additional inch of snowfall tire sales increase on average by 13.88 (holding the time period constant). t = 5.862, p-value = 0. Week: b2  7.687 ; weekly sales decrease on average by 7.687 (holding snowfall constant). t = –4.579, p-value = .0002

643


17.58a

A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.6894 5 R Square 0.4752 6 Adjusted R Square 0.4363 7 Standard Error 63.08 8 Observations 30 9 10 ANOVA 11 df SS 12 Regression 2 97283 13 Residual 27 107428 14 Total 29 204711 15 16 Coefficients Standard Error 17 Intercept 164.01 35.9 18 Fetilizer 0.140 0.081 19 Water 0.0313 0.0067

D

E

MS 48641 3979

F Significance F 12.23 0.0002

t Stat P-value 4.57 9.60E-05 1.72 0.0974 4.64 8.08E-05

a ŷ  164 .01  .140 x 1  .0313 x 2 For each additional unit of fertilizer crop yield increases on average by .140 (holding the amount of water constant). For each additional unit of water crop yield increases on average by .0313 (holding the fertilizer constant). b

H 0 : 1  0 H1 : 1  0

t = 1.72, p-value = .0974. There is not enough evidence to conclude that there is a linear relationship between crop yield and amount of fertilizer. c

H 0 : 2  0 H1 : 2  0

t = 4.64, p-value = .0001. There is enough evidence to conclude that there is a linear relationship between crop yield and amount of water. d s  = 63.08 and R 2 = .4752; the model fits moderately well.

644

F


e

Frequency

Histogram 20 10

0 -3

-2

-1

0

1

2

3

Standard Residuals

Plot of Residuals vs Predicted 150 100

Residuals

50

0 0

100

200

300

400

500

-50 -100 -150 Predicted

The errors appear to be normal, but the plot of residuals vs predicted aeems to indicate a problem. f

A 1 Prediction Interval 2 3 4 5 Predicted value 6 7 Prediction Interval 8 Lower limit 9 Upper limit 10 11 Interval Estimate of Expected Value 12 Lower limit 13 Upper limit 14

B

Yield 209.3

69.2 349.3

155.7 262.8

We predict that the crop yield will fall between 69.2 and 349.3.

645


17.59

A B C D E F 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.6882 4 Multiple R 0.4736 5 R Square 0.4134 6 Adjusted R Square 2,644 7 Standard Error 8 Observations 40 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 4 220,130,124 55,032,531 7.87 0.0001 13 Residual 35 244,690,939 6,991,170 14 Total 39 464,821,063 15 16 Coefficients Standard Error t Stat P-value 17 Intercept 1,433 2,093 0.68 0.4980 18 Size -14.55 20.70 -0.70 0.4866 19 Apartments 113.0 24.01 4.70 0.0000 20 Age -50.10 98.81 -0.51 0.6153 21 Floors -223.8 171.1 -1.31 0.1994 b

H 0 : 1  2  3  4  0 H1 : At least one  i is not equal to zero

F = 7.87, p-value = .0001. There is enough evidence to conclude that the model is valid. c The regression equation for Exercise 16.12 is ŷ = 4040 + 44.97x. The addition of the new variables changes the coefficients of the regression line in Exercise 17.12. 17.60a ŷ  29 .60  .309 x 1  1.11x 2

646


A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.7825 4 Multiple R 0.6123 5 R Square 0.5835 6 Adjusted R Square 2.16 7 Standard Error 8 Observations 30 9 10 ANOVA 11 df SS 12 Regression 2 199.65 13 Residual 27 126.44 14 Total 29 326.09 15 16 Coefficients Standard Error 17 Intercept 29.60 2.08 18 Vacancy -0.309 0.067 19 Unemployment -1.11 0.24

D

E

MS 99.82 4.68

F Significance F 21.32 0.0000

t Stat P-value 14.22 0.0000 -4.58 0.0001 -4.73 0.0001

2 b R  .6123 ; 61.23 % of the variation in rents is explained by the independent variables.

H 0 : 1   2  0

c

H1 : At least one  i is not equal to zero F = 21.32, p-value = 0. There is enough evidence to conclude that the model is valid.

H 0 : i  0

d

H1 :  i  0 Vacancy rate: t = –4.58, p-value = .0001 Unemployment rate: t = –4.73, p-value = .0001 Both vacancy and unemployment rates are linearly related to rents. e

Frequency

Histogram 20 10 0 -3

-2

-1

0

1

Standard Residuals

647

2

3

F


Plot of Residuals vs Predicted 6

Residuals

4 2 0 -2

10

12

14

16

18

20

22

-4 Predicted

The error is approximately normally distributed with a constant variance. f

A B C 1 Durbin-Watson Statistic 2 3 d = 2.0687 d L = 1.28, d U = 1.57, 4 – d U = 2.72, 4 – d L = 2.43. There is no evidence of first–order autocorrelation. g

A B C 1 Prediction Interval 2 Rent 3 4 18.72 5 Predicted value 6 7 Prediction Interval 14.18 8 Lower limit 23.27 9 Upper limit 10 11 Interval Estimate of Expected Value 17.76 12 Lower limit 19.68 13 Upper limit

D

The city's office rent is predicted to lie between $14.18 and $23.27.

648


Case 17.1 A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.1228 4 Multiple R 0.0151 5 R Square 0.0131 6 Adjusted R Square 8.30 7 Standard Error 8 Observations 2029 9 10 ANOVA 11 df SS 12 Regression 4 2,137 13 Residual 2024 139,561 14 Total 2028 141,698 15 16 Coefficients Standard Error 17 Intercept -1.15 2.20 18 SAT 0.0051 0.0013 19 MBA 0.674 0.376 20 Age -0.141 0.042 21 Tenure 0.082 0.176

D

E

F

MS 534.29 68.95

F

Significance F 0.0000

7.75

t Stat P-value -0.52 0.6012 3.96 0.0001 1.79 0.0730 -3.31 0.0009 0.47 0.6412

The model is valid (F = 7.75, p-value = 0) but the model does not fit well (R 2 = .0151; only 1.51% of the variation in returns is explained by the model).

Interpreting the coefficients in this sample: For each additional one–point increase in the SAT score, returns increase on average by .0051 provided the other variables remain constant. The returns of mutual funds managed by MBAs are on average .674 larger than the returns of mutual funds managed by people without an MBA For each additional one–year increase in age of the manager , returns decrease on average by .141 provided the other variables remain constant. For each additional one–year increase in the manager’s job tenure, returns increase on average by .082 provided the other variables remain constant.

Testing the coefficients: SAT: t = 3.96, p-value = .0001 MBA: t = 1.79, p-value = .0730 Age: t = –3.31, p-value = .0009 Tenure: t = .47, p-value = .6412

649


There is overwhelming evidence to infer that SAT scores of the undergraduate university and age of the manager are linearly related to returns. There is weak evidence that MBAs and non–MBAs have different mean returns. There is not enough evidence to conclude that job tenure is linearly related to returns.

Case 17.2 Analysis of Betas

A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.3597 5 R Square 0.1294 6 Adjusted R Square 0.1277 7 Standard Error 0.2245 8 Observations 2029 9 10 ANOVA 11 df SS 12 Regression 4 15.15 13 Residual 2024 101.97 14 Total 2028 117.12 15 16 Coefficients Standard Error 17 Intercept 0.152 0.059 18 SAT 0.00050 0.000035 19 MBA 0.0366 0.0102 20 Age 0.0088 0.0011 21 Tenure -0.0352 0.0047

D

E

F

MS 3.79 0.050

F Significance F 75.20 0.0000

t Stat P-value 2.56 0.0107 14.55 0.0000 3.60 0.0003 7.66 0.0000 -7.42 0.0000

The model is valid (F = 75.20, p-value = 0) with R 2 = .1294; only 12.94% of the variation in betas is explained by the model.

Interpreting the coefficients in this sample: For each additional one–point increase in the SAT score, betas increase on average by .00050 provided the other variables remain constant. The betas of mutual funds managed by MBAs are on average .0366 larger than the betas of mutual funds managed by people without an MBA For each additional one–year increase in age of the manager, betas increase on average by .0088 provided the other variables remain constant. For each additional one–year increase in the manager’s job tenure, betas decrease on average by .0352 provided the other variables remain constant.

650


Testing the coefficients: SAT: t = 14.55, p-value = 0 MBA: t = 3.60, p-value = .0003 Age: t = 7.66, p-value = 0 Tenure: t = –7.42, p-value = 0 There is overwhelming evidence to infer that all four independent variables are linearly related to mutual fund betas.

Analysis of MERs

A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.2697 4 Multiple R 0.0728 5 R Square 0.0705 6 Adjusted R Square 0.6847 7 Standard Error 8 Observations 2029 9 10 ANOVA 11 df SS 12 Regression 5 74.42 13 Residual 2023 948.48 14 Total 2028 1022.90 15 16 Coefficients Standard Error 17 Intercept 2.89 0.183 18 SAT -0.00055 0.00011 19 MBA -0.082 0.0310 20 Age 0.0133 0.0035 21 Tenure 0.0375 0.0145 22 Log Assets -0.209 0.0229

D

E

MS 14.88 0.47

F Significance F 31.74 0.0000

t Stat 15.73 -5.21 -2.65 3.80 2.59 -9.13

F

P-value 0.0000 0.0000 0.0081 0.0001 0.0097 0.0000

The model is valid (F = 31.74, p-value = 0) with R 2 = .0728; only 7.28% of the variation in MERs is explained by the model.

Interpreting the coefficients in this sample: For each additional one–point increase in the SAT score, MERs decrease on average by .00055 provided the other variables remain constant. The MERs of mutual funds managed by MBAs are on average .082 smaller than the MERs of mutual funds managed by people without an MBA For each additional one–year increase in age of the manager, MERs increase on average by .0133 provided the other variables remain constant.

651


For each additional one–year increase in the manager’s job tenure, MERs increase on average by .0375 provided the other variables remain constant. For each additional one–point increase in the log of the assets, MERs decrease on average by .209 provided the other variables remain constant.

Testing the coefficients: SAT: t = –5.21, p-value = 0 MBA: t = –2.65, p-value = .0081 Age: t = 3.80, p-value = .0001 Tenure: t = 2.59, p-value = .0097 Log Assets: t = –9.13, p-value = 0 There is overwhelming evidence to infer that all five independent variables are linearly related to mutual fund MERs.

652


Chapter 18 18.1 a

b

18.2 a

671


b

2

18.3 a Sales =  0 + 1 Space +  2 Space +  b A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.6378 4 Multiple R 0.4068 5 R Square 0.3528 6 Adjusted R Square 41.15 7 Standard Error 8 Observations 25 9 10 ANOVA 11 df SS 12 Regression 2 25,540 13 Residual 22 37,248 14 Total 24 62,788 15 16 Coefficients Standard Error 17 Intercept -108.99 97.24 18 Space 33.09 8.59 19 Space-sq -0.666 0.177

D

E

F

MS 12,770 1,693

F

Significance F 0.0032

7.54

t Stat P-value -1.12 0.2744 3.85 0.0009 -3.75 0.0011

s  = 41.15 and R 2 = .4068. The model's fit is relatively poor. F = 7.54, p-value = .0032. However, there is enough evidence to support the validity of the model.

18.4a

First–order model: a Demand =  0 + 1 Price+  2

Second–order model: a Demand =  0 + 1 Price +  2 Price + 

672


First–order model: A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9249 5 R Square 0.8553 6 Adjusted R Square 0.8473 7 Standard Error 13.29 8 Observations 20 9 10 ANOVA 11 df SS 12 Regression 1 18,798 13 Residual 18 3,179 14 Total 19 21,977 15 16 Coefficients Standard Error 17 Intercept 453.6 15.18 18 Price -68.91 6.68

D

E

MS 18,798 176.6

F Significance F 106.44 0.0000

F

t Stat P-value 29.87 0.0000 -10.32 0.0000

Second–order model: A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.9862 4 Multiple R 0.9726 5 R Square 0.9693 6 Adjusted R Square 5.96 7 Standard Error 8 Observations 20 9 10 ANOVA 11 df SS 12 Regression 2 21,374 13 Residual 17 603 14 Total 19 21,977 15 16 Coefficients Standard Error 17 Intercept 766.9 37.40 18 Price -359.1 34.19 19 Price-sq 64.55 7.58

D

E

MS 10,687 35.49

F Significance F 301.15 0.0000

t Stat P-value 20.50 0.0000 -10.50 0.0000 8.52 0.0000

c The second order model fits better because its standard error of estimate is 5.96, whereas that of the first–order models is 13.29 d ŷ .= 766.9 –359.1(2.95) + 64.55(2.95) 2 = 269.3

673

F


18.5a

First–order model: a Time =  0 + 1 Day+  2

Second–order model: a Time =  0 + 1 Day +  2 Day +  b First–order model A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.9222 4 Multiple R 0.8504 5 R Square 0.8317 6 Adjusted R Square 1.79 7 Standard Error 8 Observations 10 9 10 ANOVA 11 df SS 12 Regression 1 145.34 13 Residual 8 25.56 14 Total 9 170.90 15 16 Coefficients Standard Error 17 Intercept 41.40 1.22 18 Day -1.33 0.197

D

E

F

MS 145.34 3.20

F Significance F 45.48 0.0001

t Stat P-value 33.90 0.0000 -6.74 0.0001

F = 45.48, p-value = 0. The model is valid. Second–order model A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.9408 4 Multiple R 0.8852 5 R Square 0.8524 6 Adjusted R Square 1.67 7 Standard Error 8 Observations 10 9 10 ANOVA 11 df SS 12 Regression 2 151.28 13 Residual 7 19.62 14 Total 9 170.90 15 16 Coefficients Standard Error 17 Intercept 43.73 1.97 18 Day -2.49 0.822 19 Dat-sq 0.106 0.073

D

E

MS 75.64 2.80

F Significance F 26.98 0.0005

t Stat P-value 22.21 0.0000 -3.03 0.0191 1.46 0.1889

F = 26.98, p-value = .0005. The model is valid. c The second–order model is only slightly better because its standard error of estimate is smaller.

674

F


18.6a MBA GPA=  0 + 1 UnderGPA +  2 GMAT +  3 Work +  4 UnderGPA  GMAT +  b A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.6836 4 Multiple R 0.4674 5 R Square 0.4420 6 Adjusted R Square 0.790 7 Standard Error 8 Observations 89 9 10 ANOVA 11 df SS 12 Regression 4 45.97 13 Residual 84 52.40 14 Total 88 98.37 15 16 Coefficients Standard Error 17 Intercept -11.11 14.97 18 UnderGPA 1.19 1.46 19 GMAT 0.0311 0.0255 20 Work 0.0956 0.0312 21 UGPA-GMAT -0.0019 0.0025

D

E

F

MS 11.49 0.62

F Significance F 18.43 0.0000

t Stat P-value -0.74 0.4601 0.82 0.4159 1.22 0.2265 3.06 0.0030 -0.78 0.4392

F = 18.43, p-value = 0; s  = .790 and R 2 = .4674. The model is valid, but the fit is relatively poor. c MBA example s  = .788 and R 2 = .4635. There is little difference between the fits of the two models.

18.7 a (Excel output shown below) b

H 0 : 1   2   3  0 H1 : At least on  i is not equal to 0

F = 80.65, p-value = 0. There is enough evidence to infer that the model is valid.

675


A B C D E F 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9330 5 R Square 0.8705 6 Adjusted R Square 0.8597 7 Standard Error 4745 8 Observations 40 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 3 5,446,857,189 1,815,619,063 80.65 0.0000 13 Residual 36 810,445,659 22,512,379 14 Total 39 6,257,302,848 15 16 Coefficients Standard Error t Stat P-value 17 Intercept -82,044 48,530 -1.69 0.0996 18 Home % 98,443 97,463 1.01 0.3192 19 Visiting % 106,779 98,313 1.09 0.2846 20 Home-Visit 53,204 196,610 0.27 0.7882

18.8a A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9255 5 R Square 0.8566 6 Adjusted R Square 0.8362 7 Standard Error 5.20 8 Observations 25 9 10 ANOVA 11 df SS 3 3398.7 12 Regression 21 568.8 13 Residual 14 Total 24 3967.4 15 16 Coefficients Standard Error 17 Intercept 260.7 162.3 18 Temperature -3.32 2.09 19 Currency -164.3 667.1 20 Temp-Curr 3.64 8.54

676

D

E

MS 1132.9 27.08

F Significance F 41.83 0.0000

t Stat P-value 1.61 0.1230 -1.59 0.1270 -0.25 0.8078 0.43 0.6741

F


b C B A 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9312 5 R Square 0.8671 6 Adjusted R Square 0.8322 7 Standard Error 5.27 25 8 Observations 9 10 ANOVA SS 11 df 3440.3 5 12 Regression 527.1 19 13 Residual 3967.4 24 14 Total 15 16 Coefficients Standard Error 283.8 274.8 17 Intercept 6.88 -1.72 18 Temperature 888.5 -828.6 19 Currency 0.0475 -0.0024 20 Temp-sq 1718.5 2054.0 21 Curr-sq 10.57 -0.870 22 Temp-Curr

F

D

E

MS 688.1 27.74

Significance F F 0.0000 24.80

P-value t Stat 0.3449 0.97 0.8053 -0.25 0.3627 -0.93 0.9608 -0.05 0.2467 1.20 0.9353 -0.08

c Both models fit equally well. The standard errors of estimate and coefficients of determination are quite similar.

18.9a A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.3788 4 Multiple R 0.1435 5 R Square 0.1167 6 Adjusted R Square 1.58 7 Standard Error 8 Observations 100 9 10 ANOVA 11 df SS 12 Regression 3 40.38 13 Residual 96 241.06 14 Total 99 281.44 15 16 Coefficients Standard Error 17 Intercept -4.86 1.83 18 Faceoff 0.121 0.0366 19 PM-diff 0.135 0.399 20 Face-PM -0.0009 0.0080

677

D

E

F

MS 13.46 2.51

F

Significance F 0.0019

5.36

t Stat P-value -2.66 0.0092 3.31 0.0013 0.34 0.7360 -0.12 0.9086


b

H 0 : 1   2   3  0 H1 : At least on  i is not equal to 0

F = 5.36, p-value = .0019. There is enough evidence to infer that the model is valid. c

H 0 : 3  0 H1 : 3  0

t = –.12, p-value = .9086. There is not enough evidence to infer that there is an interaction effect between face–offs won and penalty minutes differential.

18.10a Yield =  0 + 1 Pressure +  2 Temperature +  3 Pressure

2

2

+  4 Temperature +  5 Pressure Temperature +  b A B C D E F 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.8290 4 Multiple R 0.6872 5 R Square 0.6661 6 Adjusted R Square 512 7 Standard Error 8 Observations 80 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 5 42,657,846 8,531,569 32.52 0.0000 13 Residual 74 19,413,277 262,342 14 Total 79 62,071,123 15 16 Coefficients Standard Error t Stat P-value 17 Intercept 74462 7526 9.89 0.0000 18 Pressure 14.40 5.92 2.43 0.0174 19 Temperature -613.3 59.95 -10.23 0.0000 20 Press-sq -0.0159 0.0032 -5.04 0.0000 21 Temp-sq 1.23 0.12 9.86 0.0000 22 Press-temp 0.0381 0.0174 2.19 0.0316 2

c s = 512 and R = .6872. The model's fit is good. 18.11 The number of indicator variables is m – 1 = 5 – 1 = 4.

678


18.12 a

I 1 = 1 if Catholic I 1 = 0 otherwise

I 2 = 1 if Protestant I 2 = 0 otherwise b

I 1 = 1 if 8:00 A.M. to 4:00 P.M. I 1 = 0 otherwise

I 2 = 1 if 4:00 P.M. to midnight

I 2 = 0 otherwise c

I 1 = 1 if Jack Jones I 1 = 0 otherwise

I 2 = 1 if Mary Brown I 2 = 0 otherwise I 3 = 1 if George Fosse I 3 = 0 otherwise 18.13 a Macintosh

18.14

b IBM

c other

I1  1 if B.A. = 0 otherwise

I 2  1 if B.B.A. = 0 otherwise

I 3  1 if B.Sc. or B.Eng. = 0 otherwise

679


I1: t = -1.54, p-value = .1269 I2: t = 2.93, p-value = .0043 I3: t = .166, p-value = .8684 Only I2 is statistically significant. However, this allows us to answer the question affirmatively. 18.15a

A B C 1 Prediction Interval 2 MBA GPA 3 4 10.11 5 Predicted value 6 7 Prediction Interval 8.55 8 Lower limit 11.67 9 Upper limit 10 11 Interval Estimate of Expected Value 9.53 12 Lower limit 10.68 13 Upper limit

D

Prediction: MBA GPA will lie between 8.55 and 11.67

680


b

A B C 1 Prediction Interval 2 MBA GPA 3 4 9.73 5 Predicted value 6 7 Prediction Interval 8.15 8 Lower limit 11.31 9 Upper limit 10 11 Interval Estimate of Expected Value 9.10 12 Lower limit 10.36 13 Upper limit

D

Prediction: MBA GPA will lie between 8.15 and 11.31

18.16a A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.8973 4 Multiple R 0.8051 5 R Square 0.7947 6 Adjusted R Square 2.32 7 Standard Error 8 Observations 100 9 10 ANOVA 11 df SS 12 Regression 5 2096.3 13 Residual 94 507.5 14 Total 99 2603.8 15 16 Coefficients Standard Error 17 Intercept 23.57 5.98 18 Mother 0.306 0.0542 19 Father 0.303 0.0476 20 Gmothers 0.0316 0.0577 21 Gfathers 0.0778 0.0573 22 Smoker -3.72 0.669

D

E

MS 419.26 5.40

F Significance F 77.66 0.0000

t Stat P-value 3.94 0.0002 5.65 0.0000 6.37 0.0000 0.55 0.5853 1.36 0.1777 -5.56 0.0000

b Exercise 18.10: ŷ = 3.24 + .451Mother + .411Father + .0166Gmothers + .0869Gfathers There are large differences to all the coefficients. c

H 0 : 5  0 H1 : 5  0

t = –5.56, p-value = 0. There is enough evidence to infer that smoking affects longevity. 681

F


18.17a A B C D 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.8368 4 Multiple R 0.7002 5 R Square 0.6659 6 Adjusted R Square 810.8 7 Standard Error 8 Observations 40 9 10 ANOVA 11 df SS MS 12 Regression 4 53,729,535 13,432,384 13 Residual 35 23,007,438 657,355 14 Total 39 76,736,973 15 16 Coefficients Standard Error t Stat 17 Intercept 3490 469.2 7.44 18 Yest Att 0.369 0.078 4.73 19 I1 1623 492.5 3.30 20 I2 733.5 394.4 1.86 21 I3 -765.5 484.7 -1.58

b

E

F Significance F 20.43 0.0000

P-value 0.0000 0.0000 0.0023 0.0713 0.1232

H 0 : 1   2   3   4  0 H1 : At least on  i is not equal to 0

F = 20.43, p-value = 0. There is enough evidence to infer that the model is valid. c

H 0 : i  0 H1 :  i  0

I 2 : t = 1.86, p-value = .0713 I 3 : t = –1.58, p-value = .1232 Weather is not a factor in attendance. d

H0 : 2  0 H1 :  2 > 0

t = 3.30, p-value = .0023/2 = .0012. There is sufficient evidence to infer that weekend attendance is larger than weekday attendance.

682

F


18.18a

A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.5602 4 Multiple R 0.3138 5 R Square 0.2897 6 Adjusted R Square 5.84 7 Standard Error 8 Observations 60 9 10 ANOVA 11 df SS 12 Regression 2 887.9 13 Residual 57 1941.7 14 Total 59 2829.6 15 16 Coefficients Standard Error 17 Intercept 7.02 3.24 18 Length 0.250 0.056 19 Type -1.35 0.947 b

D

E

MS 443.95 34.06

F Significance F 13.03 0.0000

t Stat P-value 2.17 0.0344 4.46 0.0000 -1.43 0.1589

H0 : 2  0 H1 :  2  0

t = –1.43, p-value = .1589. There is not enough evidence to infer that the type of commercial affects memory test scores. c Let

I1 = 1 if humorous I1 = 0 otherwise I 2 = 1 if musical I 2 = 0 otherwise See Excel output below. d

H 0 : i  0 H1 :  i  0

I1: t = 1.61, p-value = .1130 I2: t = 3.01, p-value = .0039 There is enough evidence to infer that there is a difference in memory test scores between watchers of humorous and serious commercials. e The variable type of commercial in parts (a) and (b) is nominal. It is usually meaningless to conduct a regression analysis with such variables without converting them to indicator variables.

683

F


A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.6231 5 R Square 0.3882 6 Adjusted R Square 0.3554 7 Standard Error 5.56 8 Observations 60 9 10 ANOVA 11 df SS 12 Regression 3 1099 13 Residual 56 1731 14 Total 59 2830 15 16 Coefficients Standard Error 17 Intercept 2.53 2.15 18 Length 0.223 0.054 19 I1 2.91 1.81 20 I2 5.50 1.83

D

E

MS 366.17 30.91

F Significance F 11.85 0.0000

F

t Stat P-value 1.18 0.2445 4.10 0.0001 1.61 0.1130 3.01 0.0039

18.19 a A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9233 5 R Square 0.8525 6 Adjusted R Square 0.8429 7 Standard Error 6.25 8 Observations 50 9 10 ANOVA 11 df SS 12 Regression 3 10384 13 Residual 46 1796 14 Total 49 12181 15 16 Coefficients Standard Error 17 Intercept -41.42 7.00 18 Boxes 0.644 0.050 19 Weight 0.349 0.075 20 Codes 4.54 1.21

b Let

I1 = 1 if morning I1 = 0 otherwise I 2 = 1 if early afternoon I 2 = 0 otherwise 684

D

E

MS 3461.40 39.05

F Significance F 88.64 0.0000

t Stat P-value -5.92 0.0000 12.79 0.0000 4.65 0.0000 3.76 0.0005

F


A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.9727 4 Multiple R 0.9461 5 R Square 0.9414 6 Adjusted R Square 3.82 7 Standard Error 8 Observations 50 9 10 ANOVA 11 df SS 12 Regression 4 11525 13 Residual 45 655.9 14 Total 49 12181 15 16 Coefficients Standard Error 17 Intercept -29.72 3.73 18 Boxes 0.618 0.031 19 Weight 0.346 0.046 20 I1 -6.76 1.50 21 I2 6.48 1.45

D

MS 2881.16 14.58

E

F Significance F 197.66 0.0000

t Stat P-value -7.97 0.0000 19.99 0.0000 7.54 0.0000 -4.51 0.0000 4.47 0.0001

c Model 1: s  = 6.25 and R 2 = .8525. Model 2: s  = 3.82 and R 2 = .9461. The second model fits better. d

H 0 : i  0 H1 :  i  0

I1: t = –4.51, p-value = 0. There is enough evidence to infer that the average time to unload in the morning is different from that in the late afternoon. I2: t = 4.47, p-value = .0001. There is enough evidence to infer that the average time to unload in the early afternoon is different from that in the late afternoon.

18.20a Let

I1 = 1 if no scorecard I1 = 0 otherwise I 2 = 1 if scorecard overturned more than 10% of the time I 2 = 0 otherwise

685

F


b A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.7299 5 R Square 0.5327 6 Adjusted R Square 0.5181 7 Standard Error 4.20 8 Observations 100 9 10 ANOVA 11 df SS 12 Regression 3 1933 13 Residual 96 1696 14 Total 99 3629 15 16 Coefficients Standard Error 17 Intercept 4.65 2.06 18 Loan Size 0.00012 0.00015 19 I1 4.08 1.14 20 I2 10.18 1.01

D

E

F

MS 644.46 17.67

F Significance F 36.48 0.0000

t Stat P-value 2.26 0.0260 0.83 0.4084 3.57 0.0006 10.08 0.0000

c s  = 4.20 and R 2 = .5327. The model's fit is mediocre. d

1 2 3 4 5

A

B Pct Bad

C Loan Size

D I1

Pct Bad Loan Size I1 I2

1 0.1099 -0.1653 0.6835

1 -0.0346 0.0737

1 -0.5471

E I2

1

There is a high correlation between I1 and I 2 that may distort the t–tests. e b1 =.00012; in this sample for each additional dollar lent the default rate increases by .00012 provided the other variables remain the same.

b 2 = 4.08; In this sample banks that don't use scorecards on average have default rates 4.08 percentage points higher than banks that overturn their scorecards less than 10% of the time.

b 3 = 10.18; In this sample banks that overturn their scorecards more than 10% of the time on average have default rates 10.18 percentage points higher than banks that overturn their scorecards less than 10% of the time.

686


f

A B C 1 Prediction Interval 2 Pct Bad 3 4 9.94 5 Predicted value 6 7 Prediction Interval 1.39 8 Lower limit 18.49 9 Upper limit 10 11 Interval Estimate of Expected Value 8.08 12 Lower limit 11.81 13 Upper limit

D

We predict that the bank's default rate will fall between 1.39 and 18.49%.

18.21 a Let

I1 = 1 if welding machine I1 = 0 otherwise I 2 = 1 if lathe I 2 = 0 otherwise A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.7706 5 R Square 0.5938 6 Adjusted R Square 0.5720 7 Standard Error 48.59 8 Observations 60 9 10 ANOVA 11 df SS 12 Regression 3 193,271 13 Residual 56 132,223 14 Total 59 325,494 15 16 Coefficients Standard Error 17 Intercept 119.3 35.00 18 Age 2.54 0.402 19 I1 -11.76 19.70 20 I2 -199.4 30.71

687

D

E

MS 64,424 2,361

F Significance F 27.29 0.0000

t Stat P-value 3.41 0.0012 6.31 0.0000 -0.60 0.5531 -6.49 0.0000

F


b b1 = 2.54; in this sample for each additional month repair costs increase on average by $2.54 provided that the other variable remains constant.

b 2 = –11.76; in this sample welding machines cost on average $11.76 less to repair than stamping machines for the same age of machine.

b 3 = –199.4; in this sample lathes cost on average $199.40 less to repair than stamping machines for the same age of machine. c

H 0 : 2  0 H1 :  2 < 0

t = –.60, p-value .5531/2 = .2766. There is no evidence to infer that welding machines cost less to repair than stamping machines.

18.22

a. The coefficient of determination in Exercise 16.107 was .3270. In this model the coefficient of determination is .6385. This model is better.

688


b

Lower prediction limit = 150.5, upper prediction limit = 174.1 c

Lower prediction limit = 162.6, upper prediction limit = 186.3

d No, because the width of the prediction intervals are far too wide.

689


18.23a

A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.7296 5 R Square 0.5323 6 Adjusted R Square 0.5075 7 Standard Error 2.36 8 Observations 100 9 10 ANOVA 11 df SS 12 Regression 5 593.9 13 Residual 94 521.7 14 Total 99 1115.6 15 16 Coefficients Standard Error 17 Intercept 10.26 1.17 18 Wage -0.00020 0.000036 19 Pct PT -0.107 0.029 20 Pct U 0.060 0.012 21 Av Shift 1.56 0.50 22 UM Rel -2.64 0.492 b

D

E

F

MS 118.78 5.55

F Significance F 21.40 3.08E-14

t Stat P-value 8.76 8.12E-14 -5.69 1.43E-07 -3.62 0.0005 4.83 5.38E-06 3.11 0.0025 -5.36 5.99E-07

H0 : 4  0 H1 :  4  0

t = 3.11, p-value = .0025. There is enough evidence to infer that the availability of shiftwork affects absenteeism. c

H 0 : 5  0 H1 : 5 < 0

t = –5.36, p-value = (5.99E-07) /2 = 3.00×10-7= virtually 0. There is enough evidence to infer that in organizations where the union–management relationship is good absenteeism is lower.

690


18.24a Let I1= 1 if respondent works for him or herself (0 if not)

b. b4 = 8429; after removing the effect of age, education, and weekly hours of work people who work for themselves earn on average $8429 more than people who work for someone else. c.

H0:β4 = 0 H1:β4 > 0

t = 3.07, p-value = .0022/2 = .0011. There is enough evidence to infer that people who work for themselves have larger incomes after removing the effect of age, education, and weekly hours of work.

18.25 Let

I1 = 1 if Democrat (PARTYID3 = 1) = 0 if not I2 = 1 if Republican (PARTYID3 - 3 = 0, if not

691


a.

H0:β5 = 0 H1:β5 < 0

t = -6.77, p-value = 0. There is sufficient evidence to infer that Democrats are more likely than Independents to believe that government should reduce income differences after removing the effects of age, income, education, and weekly hours. b.

H0:β6 = 0 H1:β6 > 0

t = 7.06, p-value = 0. There is sufficient evidence to infer that Republicans are more likely than Independents to believe that government should take no action to reduce income differences after removing the effects of age, income, education, and weekly hours.

18.26

I1 = 1 if Liberal (POLVIEWS3 = 1) = 0 if not I2 = 1 if Conservative (POLVIEWS3 = 3 = 0, if not

692


a.

H0:β5 = 0 H1:β5 < 0

t = -5.79, p-value = 0. There is sufficient evidence to infer that liberals are more likely than moderates to believe that government should reduce income differences after removing the effects of age, income, education, and weekly hours. b.

H0:β6 = 0 H1:β6 > 0

t = 6.47, p-value = 0. There is sufficient evidence to infer that conservatives are more likely than moderates to believe that government should take no action to reduce income differences after removing the effects of age, income, education, and weekly hours.

693


18.27

I1 = 1 if Male = 0 if not

H0:β3 = 0 H1:β3 ≠ 0 t = .735, p-value = .4624. There is not enough evidence to infer that men and women differ in the amount of television per day after removing the effects of age and education.

18.28 Let

I1 = 1 if Liberal (POLVIEWS3 = 1) = 0 if not I2 = 1 if Conservative (POLVIEWS3 = 3) = 0, if not

694


a.

H0:β5 = 0 H1:β5 < 0

t = -4.59, p-value = 0. There is sufficient evidence to infer that liberals are more likely than moderates to believe that government help poor people after removing the effects of age, income, education, and weekly hours. b.

H0:β6 = 0 H1:β6 > 0

t = 4.59, p-value = 0. There is sufficient evidence to infer that conservatives are more likely than moderates to believe that people should help themselves after removing the effects of age, income, education, and weekly hours.

18.29

Let I1= 1 if respondent works for government (0 if not)

695


H0:β4 = 0 H1:β4 ≠ 0 t = -2.16, p-value = .0312. There is enough evidence to infer that there are differences in mean income between people who work for the government and people who work for private employers after removing the effects of age, education, and weekly hours of work.

18.30

I1= 1, if White =0, if not I2 =1, if black = 0, if not

696


Compare whites and others H0:β3 = 0 H1:β3 ≠ 0 t = -1.12, p-value = .2639

Compare blacks and others H0:β4 = 0 H1:β4 ≠ 0 t = 4.00, p-value = 0. There is enough evidence to infer that there are differences between the blacks and others after removing the effects of age and education.

18.31

I1 = 1 if Democrat (PARTYID3 = 1) = 0 if not I2 = 1 if Republican (PARTYID3 - 3 = 0, if not

697


a.

H0:β5 = 0 H1:β5 < 0

t = -5.83, p-value = 0. There is sufficient evidence to infer that Democrats are more likely than Independents to believe that government should help poor people after removing the effects of age, income, education, and weekly hours. b.

H0:β6 = 0 H1:β6 > 0

t = 5.94, p-value = 0. There is sufficient evidence to infer that Republicans are more likely than Independents to believe that people should help themselves after removing the effects of age, income, education, and weekly hours.

18.32

I1 = 1 if born in the U.S. (0 if not)

698


H0:β4 = 0 H1:β4 ≠ 0 t = -1.35, p-value = .1768. There is not enough evidence to infer that there are differences in income between Americans born in the United States and those born elsewhere after removing the effects of age, education , and weekly hours of work.

18.33

I1 = 1 if UNION = 1, 2, or 3 I1 = 0 if UNION = 4 (neither belong)

H0:β4 = 0 H1:β4 ≠ 0

699


t = .356, p-value = .7216. There is not enough evidence to infer that union membership affects income after removing the effects of age, education, and hours of work.

18.34 A B C D E F 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9737 5 R Square 0.9482 6 Adjusted R Square 0.9454 7 Standard Error 3015 8 Observations 100 9 10 ANOVA 11 df SS MS F Significance F 12 Regression 5 15,636,303,318 3,127,260,664 344.04 0.0000 13 Residual 94 854,451,113 9,089,905 14 Total 99 16,490,754,431 15 16 Coefficients Standard Error t Stat P-value 17 Intercept -5916 3141 -1.88 0.0627 18 Years 1022 48.93 20.88 0.0000 19 PhD 725.7 961.5 0.75 0.4523 20 Evaluation 3729 619.8 6.02 0.0000 21 Articles 439.1 80.7 5.44 0.0000 22 Gender 1090 632.0 1.72 0.0879

a

H 0 : 1   2   3   4   5  0 H1 : At least on  i is not equal to 0

F = 344.04, p-value = 0. There is enough evidence to infer that the model is valid. b

H 0 : 5  0 H1 : 5 > 0

t = 1.72, p-value = .0879/2 = .0440. There is evidence that male professors are better paid than female professors with the same qualifications.

700


18.35 A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.8311 4 Multiple R 0.6907 5 R Square 0.5670 6 Adjusted R Square 1.86 7 Standard Error 8 Observations 8 9 10 ANOVA 11 df SS 12 Regression 2 38.47 13 Residual 5 17.23 14 Total 7 55.70 15 16 Coefficients Standard Error 17 Intercept 2.01 4.02 18 Score 3.25 1.00 19 Gender -0.039 1.35

D

E

F

MS 19.24 3.45

F

Significance F 0.0532

5.58

t Stat P-value 0.50 0.6385 3.25 0.0227 -0.03 0.9782

In this case male–dominated jobs are paid on average $.039 (3.9 cents) less than female– dominated jobs after adjusting for the value of each job.

18.36 All weights = .2 A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.7623 4 Multiple R 0.5812 5 R Square 0.4136 6 Adjusted R Square 2.16 7 Standard Error 8 Observations 8 9 10 ANOVA 11 df SS 12 Regression 2 32.37 13 Residual 5 23.33 14 Total 7 55.70 15 16 Coefficients Standard Error 17 Intercept 4.70 4.07 18 Score 2.57 1.01 19 Gender 0.26 1.56

701

D

E

F

MS 16.19 4.67

F

Significance F 0.1135

3.47

t Stat P-value 1.15 0.3011 2.55 0.0514 0.16 0.8761


In this case male–dominated jobs are paid on average $.26 (26 cents) more than female–dominated jobs after adjusting for the value of each job.

18.37 The strength of this approach lies in regression analysis. This statistical technique allows us to determine whether gender is a factor in determining salaries. However, the conclusion is very much dependent upon the subjective assignment of weights. Change the value of the weights and a totally different conclusion is achieved.

18.38a.

b. In this regression analysis the variables DAYS2 and DAYS3 are not included in the equation.

18.39a

702


b. The variables Enrollment and Distance are excluded. 18.40

18.41

703


18.42

2

18.43a Mileage = 0 + 1 Speed +  2 Speed +  b

704


A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.8428 4 Multiple R 0.7102 5 R Square 0.6979 6 Adjusted R Square 3.86 7 Standard Error 8 Observations 50 9 10 ANOVA 11 df SS 12 Regression 2 1719.9 13 Residual 47 701.64 14 Total 49 2421.5 15 16 Coefficients Standard Error 17 Intercept 9.34 1.71 18 Speed 0.802 0.077 19 Speed-sq -0.0079 0.00073

D

E

MS 859.94 14.93

F Significance F 57.60 0.0000

F

t Stat P-value 5.47 0.0000 10.39 0.0000 -10.73 0.0000

c s  = 3.86 and R 2 = .7102. The model fits moderately well.

18.44a Apply a first–order model with interaction. b A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.8623 5 R Square 0.7436 6 Adjusted R Square 0.7299 7 Standard Error 1.27 8 Observations 60 9 10 ANOVA 11 df SS 12 Regression 3 260.2 13 Residual 56 89.72 14 Total 59 349.9 15 16 Coefficients Standard Error 17 Intercept 640.8 53.80 18 Cars -64.17 5.27 19 Speed -10.63 0.897 20 Cars-Speed 1.08 0.088

c:

H 0 : 1   2   3  0 H1 : At least on  i is not equal to 0

705

D

E

MS 86.74 1.60

F Significance F 54.14 0.0000

t Stat P-value 11.91 0.0000 -12.19 0.0000 -11.85 0.0000 12.26 0.0000

F


F = 54.14, p-value = 0. There is enough evidence to infer that the model is valid.

18.45a A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.8668 5 R Square 0.7514 6 Adjusted R Square 0.7284 7 Standard Error 1.27 8 Observations 60 9 10 ANOVA 11 df SS 12 Regression 5 262.95 13 Residual 54 86.99 14 Total 59 349.93 15 16 Coefficients Standard Error 17 Intercept 404.5 327.0 18 Cars -66.57 6.54 19 Speed -2.35 10.54 20 Cars-sq 0.107 0.097 21 Speed-sq -0.070 0.085 22 Cars-Speed 1.08 0.096

D

E

MS 52.59 1.61

F Significance F 32.65 0.0000

t Stat P-value 1.24 0.2214 -10.19 0.0000 -0.22 0.8246 1.10 0.2741 -0.82 0.4180 11.21 0.0000

b F = 32.65, p-value = 0. There is enough evidence to infer that the model is valid.

18.46 a Let

I1 = 1 if ad was in newspaper I1 = 0 otherwise I 2 = 1 if ad was on radio I 2 = 0 otherwise

b

706

F


A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.6946 5 R Square 0.4824 6 Adjusted R Square 0.4501 7 Standard Error 44.87 8 Observations 52 9 10 ANOVA 11 df SS 12 Regression 3 90057 13 Residual 48 96627 14 Total 51 186684 15 16 Coefficients Standard Error 17 Intercept 282.6 17.46 18 Ads 25.23 3.98 19 I1 -23.36 15.83 20 I2 -46.59 16.44

b

D

E

MS 30019 2013

F Significance F 14.91 0.0000

t Stat P-value 16.19 0.0000 6.34 0.0000 -1.48 0.1467 -2.83 0.0067

H 0 : 1   2   3  0 H1 : At least on  i is not equal to 0

F = 14.91, p-value = 0. There is enough evidence to infer that the model is valid. c

H 0 : i  0 H1 : i  0

I–1: t = –1.48, p-value = .1467 I–2: t = –2.83, p-value = .0067 There is enough evidence to infer that the advertising medium makes a difference.

18.47 (See Excel output below) b

H 0 : 3  0 H1 : 3 < 0

t = –8.61, p-value = 0. There is enough evidence to infer that a team that fires its manager within 12 months wins less frequently than other teams.

707

F


A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.9347 5 R Square 0.8736 6 Adjusted R Square 0.8654 7 Standard Error 0.0183 8 Observations 50 9 10 ANOVA 11 df SS 12 Regression 3 0.107 13 Residual 46 0.0154 14 Total 49 0.122 15 16 Coefficients Standard Error 17 Intercept 0.357 0.0592 18 BA -0.401 0.236 19 ERA 0.0764 0.00478 20 Fired -0.0509 0.00591

D

MS 0.0355 0.00034

E

F

F Significance F 106.01 0.0000

t Stat 6.03 -1.70 15.98 -8.61

P-value 0.0000 0.0964 0.0000 0.0000

D

E

MS 87,646 7,740

F Significance F 11.32 0.0000

2

18.48a Units = 0 + 1 Years +  2 Years +  b A B C 1 SUMMARY OUTPUT 2 Regression Statistics 3 0.4351 4 Multiple R 0.1893 5 R Square 0.1726 6 Adjusted R Square 87.98 7 Standard Error 8 Observations 100 9 10 ANOVA 11 df SS 12 Regression 2 175,291 13 Residual 97 750,764 14 Total 99 926,056 15 16 Coefficients Standard Error 17 Intercept 331.2 17.55 18 Years 21.45 5.50 19 Years-sq -0.848 0.325

c s  = 87.98 and R 2 = .1893. The model fits poorly.

708

t Stat P-value 18.87 0.0000 3.90 0.0002 -2.61 0.0105

F


2

18.49a Depletion = 0 + 1 Temperature +  2 PH–level + 3 PH–level +  4 I 4 +  5 I 5 +  where

I1 = 1 if mainly cloudy I1 = 0 otherwise I 2 = 1 if sunny I 2 = 0 otherwise b A B C 1 SUMMARY OUTPUT 2 3 Regression Statistics 4 Multiple R 0.8085 5 R Square 0.6537 6 Adjusted R Square 0.6452 7 Standard Error 4.14 8 Observations 210 9 10 ANOVA 11 df SS 12 Regression 5 6596 13 Residual 204 3495 14 Total 209 10091 15 16 Coefficients Standard Error 17 Intercept 1003 55.12 18 Temperature 0.194 0.029 19 PH Level -265.6 14.75 20 PH-sq 17.76 0.983 21 I1 -1.07 0.700 22 I2 1.16 0.700

c

D

E

MS 1319 17.13

F Significance F 77.00 0.0000

t Stat P-value 18.19 0.0000 6.78 0.0000 -18.01 0.0000 18.07 0.0000 -1.53 0.1282 1.65 0.0997

H 0 : 1   2   3   4   5  0 H1 : At least on  i is not equal to 0

F = 77.00, p-value = 0. There is enough evidence to infer that the model is valid. d

H 0 : 1  0 H1 : 1 > 0

t = 6.78, p-value = 0. There is enough evidence to infer that higher temperatures deplete chlorine more quickly. e

H 0 : 3  0 H1 : 3 > 0

t = 18.07, p-value = 0. There is enough evidence to infer that there is a quadratic relationship between chlorine depletion and PH level. 709

F


f

H 0 : i  0 H1 :  i  0

I1 : t = –1.53, p-value = .1282. There is not enough evidence to infer that chlorine depletion differs between mainly cloudy days and partly sunny days.

I 2 : t = 1.65, p-value = .0997. There is not enough evidence to infer that chlorine depletion differs between sunny days and partly sunny days. Weather is not a factor in chlorine depletion.

710


Chapter 19 19.1

H0: The two population locations are the same H1: The location of population 1 is different from the location of population 2

E (T )  T 

n1 (n1  n 2  1) 15(15  15  1)   232 .5 2 2 n1n 2 (n1  n 2  1) (15)(15)(15  15  1)   24 .11 12 12

a z

T  E(T) 250  232 .5 = = .73, p-value = 2P(Z > .73) = 2(1 – .7673) = .4654. 24 .11 T

b z

T  E(T) 275  232 .5 = = 1.76, p-value 2P(Z > 1.76) = 2(1 – .9608) = .0784. 24 .11 T

c The value of the test statistic increases and the p-value decreases.

19.2

H 0 : The two population locations are the same H1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z   z .05 = 1.645

E (T ) 

n 1 (n 1  n 2  1) 30 (30  40  1)   1,065 2 2

T 

n 1 n 2 (n 1  n 2  1)  12

a z

(30 )( 40 )(30  40  1)  84 .26 12

T  E(T) 1,205  1,065 = = 1.66, p-value P(Z > 1.66) = 1 – .9515 = .0485. There is enough 84 .26 T

evidence to infer that the location of population 1 is to the right of the location of population 2. b z

T  E(T) 1,065  1,065 = = 0, p-value = P(Z > 0) = .5. There is not enough evidence to infer 84 .26 T

that the location of population 1 is to the right of the location of population 2. c The value of the test statistic decreases and the p-value increases.

19.3

H 0 : The two population locations are the same H1 : The location of population 1 is to the left of the location of population 2

Rejection region: T  TL = 19

707


Sample 1 75 60 73 66 81

Rank Sample 2 Rank 5 90 9 1 72 3 4 103 10 2 82 8 7 78 6 T1 = 19 T2 = 36 There is enough evidence to infer that the location of population 1 is to the left of the location of population 2.

19.4

H 0 : The two population locations are the same H1 : The location of population 1 is different from the location of population 2

Rejection region: T  TU = 127 or T  TL = 83 Sample 1 Rank Sample 2 Rank 15 4.0 8 2.0 7 1.0 27 18.0 22 14.0 17 7.0 20 .5 25 16.0 32 20.0 20 11.5 18 9.5 16 5.0 26 17.0 21 13.0 17 7.0 17 7.0 23 15.0 10 3.0 30 19.0 18 9.5 T1 = 118 T2 = 92 There is not enough evidence to infer that the location of population 1 is different from the location of population 2.

19.5

H 0 : The two population locations are the same H1 : The location of population 1 is to the left of the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 Rank Sum Observations 3 4 New 623.5 25 651.5 25 5 Leading 6 z Stat -0.2716 7 P(Z<=z) one-tail 0.3929 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.7858 1.96 10 z Critical two-tail

E

a z = –.27, p-value = .3929. There is not enough evidence to infer that the new beer is less highly rated than the leading brand. b The printout is identical to that of part a. c All codes that preserve the order produce the same results. 708


19.6

H 0 : The two population locations are the same H1 : The location of population 1 is different from the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 Business 4004 40 5 Economy 8086 115 6 z Stat 3.6149 7 P(Z<=z) one-tail 0.0002 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.0004 10 z Critical two-tail 1.96

E

a z = 3.61, p-value = .0004. There is enough evidence to infer that the business and economy class differ in their degree of satisfaction. b The printout is identical to that of part a. c All codes that preserve the order produce the same results.

19.7

H 0 : The two population locations are the same H1 : The location of population 1 is to the right of the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 New 276.5 15 5 Aspirin 188.5 15 6 z Stat 1.825 7 P(Z<=z) one-tail 0.034 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.068 10 z Critical two-tail 1.96

E

a z = 1.83, p-value = .0340. There is enough evidence to infer that the new painkiller is more effective than aspirin. b The results are identical because the codes in this exercise and in Example 19.2 are ranked identically.

19.8

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.05 = 1.645

709


E(T) 

T 

z

n1 (n1  n 2  1) 82(82  75  1)   6478 2 2 n1n 2 (n1  n 2  1) (82 )(75)(82  75  1)   284 .6 12 12

T  E(T) 6,807  6,478 = = 1.16, p-value P(Z > 1.16) = 1 – .8770 = .1230. There is not 284 .6 T

enough evidence to infer that members of the Mathematics department rate nonparametric techniques as more important than do members of other departments.

19.9

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.05  1.645

E (T )  T 

z

n1 (n1  n 2  1) 30 (30  30  1)   915 2 2 n1n 2 (n1  n 2  1) (30 )(30 )(30  30  1)   67 .6 12 12

T  E(T) 797  915 =  1.75 , p-value P(Z < –1.75) = .0401. There is enough evidence to 67 .6 T

infer that companies that provide exercise programs should be given discounts.

19.10

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.05  1.645

E (T ) 

n1 (n1  n 2  1) 125 (125  125  1)   15,687 .5 2 2

T 

n1n 2 (n1  n 2  1)  12

z

(125 )(125 )(125  125  1)  571 .7 12

T  E(T) 14 ,873  15,687 .5  1.42 , p-value P(Z < – 1.42) =.0778. There is not enough = 571 .7 T

evidence to infer that women are doing less housework today than last year.

19.11

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.05  1.645

710


E (T )  T 

z

n1 (n1  n 2  1) 100 (100  100  1)   10,050 2 2 n1n 2 (n1  n 2  1) (100 )(100 )(100  100  1)   409 .3 12 12

T  E(T) 10 ,691  10 ,050 =  1.57 , p-value P(Z > 1.57) = 1 – .9418 = .0582. There is not 409 .3 T

enough evidence to conclude that public support has decreased between this year and last year.

19.12

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region: z  z / 2  z.025  1.96 or z  z  / 2  z .025  1.96

E (T )  T 

z

n1 (n1  n 2  1) 50 (50  50  1)   2525 2 2 n1n 2 (n1  n 2  1) (50 )(50 )(50  50  1)   145 .1 12 12

T  E(T) 2810  2525 =  1.964 , p-value = 2P(Z > 1.964), which is slightly less than 2P(Z > 145 .1 T

1.96) = 2(1 – .9750) = .0500. There is enough evidence to infer that men and women experience different levels of stomach upset.

19.13

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.05  1.645

E (T )  T 

z

n1 (n1  n 2  1) 15(15  25  1)   307 .5 2 2 n1n 2 (n1  n 2  1) (15)( 25)(15  25  1)   35 .8 12 12

T  E(T) 383 .5  307 .55  2.12, p-value = P(Z > 2.12) = 1 – .9830 = .0170. There is = 35 .8 T

enough evidence to infer that Tastee is superior.

19.14

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.05  1.645

711


E (T )  T 

z

n1 (n1  n 2  1) 20 (20  20  1)   410 , 2 2 n1n 2 (n1  n 2  1) (20 )( 20 )( 20  20  1)   37 .0 12 12

T  E(T) 439 .5  410 =  .80 , p-value = P(Z > .80) = 1 – .7881 = .2119. There is not enough 37 .0 T

evidence to infer that women perceive another woman wearing a size 6 dress as more professional than one wearing a size 14 dress. 19.15a. H0: The two population locations are the same H1: The location of population 1 is to the left of the location of population 2 Rejection region: z  z  z.05  1.645

E (T )  T 

z

n1 (n1  n 2  1) 125 (125  125  1)   15,687 .5 2 2 n1n 2 (n1  n 2  1) (125 )(125 )(125  125  1)   571 .7 12 12

T  E(T) 13,078  15,687 .5  4.56 , p-value = P(Z < –4.56) = 0. There is enough evidence = 571 .7 T

to infer that changing the name of prunes to dried plums will increase the likelihood that shoppers will buy.

19.16

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region: z  z / 2  z.025  1.96 or z  z  / 2  z .025  1.96

E (T )  T 

z

n1 (n1  n 2  1) 182 (182  163  1)   31,486 2 2 n1n 2 (n1  n 2  1) (182 )(163 )(182  163  1)   924 .9 12 12

T  E(T) 32,225 .5  31,486  .80 , p-value = 2P(Z > .80) = 2(1 – .7881) = .4238. There is = 924 .9 T

not enough evidence to infer that the night and day shifts rate the service differently. 19.17

H0: The two population locations are the same H1: The location of population 1 (males) is to the left of the location of population 2(females)

712


z = -2.53, p-value = .0116. There is enough evidence to infer that men prefer jobs with higher incomes than do women. 19.18

H0: The two population locations are the same H1: The location of population 1(government worker) is to the left of the location of population 2(private-sector worker)

z = -.179, p-value = .4290. There is not enough evidence to infer that government workers show a greater preference for job security than do private-sector workers. 19.19

H0: The two population locations are the same H1: The location of population 1 (self-employed) is to the left of the location of population 2(work for someone else)

713


z = -3.77, p-value = .0001. There is enough evidence to infer that people who work for themselves prefer shorter work hours with lots of free time than do workers who work for someone else. 19.20

H0: The two population locations are the same H1: The location of population 1 is different from the location of population 2

z = 2.90, p-value = .0036. There is enough evidence to infer that men and women differ in their preference for work that is important and gives a feeling of accomplishment. 19.21

H0: The two population locations are the same

714


H1: The location of population 1 is different from the location of population 2

z = -.345, p-value = .7300. There is not enough evidence to conclude that those who would stop working and those who would continue working differ in their preference for high income. 19.22

H0: The two population locations are the same H1: The location of population 1 (continue working) is to the left of the location of population 2(stop working)

z = -2.18, p-value = .0147. There is enough evidence to conclude that those who would continue working have a higher preference for work they consider important and gives them a feeling of accomplishment. 19.23

H0: The two population locations are the same H1: The location of population 1 is different from the location of population 2

715


z = .493, p-value = .6220. There is not enough evidence to conclude that men and women differ in their preference for jobs where there is a chance for advancement? 19.24

H0: The two population locations are the same H1: The location of population 1 is different from the location of population 2

z = 3.73, p-value = .0002. There is enough evidence to conclude that Democrats and Republicans differ in their views about the federal income tax that they have to pay.

19.25

H0: The two population locations are the same H1: The location of population 1 is different from the location of population 2

716


z = 1.56, p-value = .1198. There is not enough evidence to conclude that people who are selfemployed differ from people who work for someone else differ in their views about the federal income tax that they have to pay.

19.26

H0: The two population locations are the same H1: The location of population 1 (men) is to the right of the location of population 2 (women)

z = .707, p-value = .2399. There is not enough evidence to infer that women are more likely than men to lose their jobs in the next 12 months. 19.27

H0: The two population locations are the same H1: The location of population 1 (men) is to the left of the location of population 2 (women)

717


z = -.017, p-value = .4934. There is not enough evidence to infer that men consider themselves to be healthier than women. 19.28

H0: The two population locations are the same H1: The location of population 1 (2012) is to the left of the location of population 2 (2014)

z = -.452, p-value = .3257. There is not enough evidence to infer that Americans were healthier in 2012 than in 2014. 19.29

H0: The two population locations are the same H1: The location of population 1 (2012) is to the left of the location of population 2 (2014)

718


z = -1.23, p-value = .1089. There is not enough evidence to infer that Americans were more worried about their chances of losing their jobs in 2012 than in 2014. 19.30

H0: The two population locations are the same H1: The location of population 1 (2012) is to the right of the location of population 2 (2014)

z = .630, p-value = .2642. There is not enough evidence to conclude that Americans were more optimistic about their children’s standard of living in 2014 than they were in 2012. 19.31a. Assets are required to be normally distributed. b. The histograms are not bell shaped. c.

H0: The two population locations are the same H1: The location of population 1 (Some college) is to the left of the location of population 2 (College)

719


z = -5.21, p-value = 0. There is enough evidence to infer that heads of households with college degrees have more assets than do heads of households who have some college. 19.32a. Income is required to be normally distributed. The histograms are not bell shaped. The Wilcoxon rank sum test should be used. b.

H0: The two population locations are the same H1: The location of population 1(Work for someone else) is to the right of the location of population 2 (Self-employed0

z = 3.07, p-value = .0011. There is enough evidence to conclude that heads of households who work for someone else have higher incomes than self-employed heads of households. 19.33a. Incomes are required to be normally distributed. b. The histograms are not bell shaped. c.

H0: The two population locations are the same H1: The location of population 1 (Male) is to the right of the location of population 2 (Female)

720


z = 9.91, p-value = 0. There is enough evidence to conclude that male heads of households have higher incomes than female heads of households. 19.34a. Debt is required to be normally distributed. The histograms are not bell shaped. The Wilcoxon rank sum test should be used. b.

H0: The two population locations are the same H1: The location of population 1 (Some college) is to the left of the location of population 2 (College)

z = -4.65, p-value = 0. There is evidence to conclude that heads of households with some college have less debt than heads of heads of households who have graduated college.

19.35

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region: z  z / 2  z.025  1.96 or z  z / 2  z.025  1.96 z=

x  .5n .5 n

=

15  .5(45 ) .5 45

= –2.24, p-value = 2P(Z < – 2.24) = 2(.0125) = .0250. There is enough

evidence to infer that the population locations differ.

721


19.36

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z   z   z.10  1.28 z=

x  .5n .5 n

=

28  .5(69 ) .5 69

= –1.57, p-value = P(Z < –1.57) = .0582. There is enough evidence to

infer that the location of population 1 is to the left of the location of population 2.

19.37

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of right of the location of population 2

Rejection region: z  z  z.05  1.645 z=

x  .5n .5 n

=

18  .5(30 ) .5 30

= 1.10, p-value = P(Z > 1.10) = 1 – .8643 = .1357. There is not enough

evidence to infer that the location of population 1 is to the right of the location of population 2.

19.38

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Pair 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

A 5 3 4 2 3 4 3 5 4 3 4 5 4 5 3 2

B 3 2 4 3 3 1 3 4 2 5 1 2 2 3 2 2

Sign of Difference + + 0 – 0 + 0 + + – + + + + + 0

Rejection region: z  z  z.05  1.645 x = 10, n = 12, z 

x  .5n .5 n

=

10  .5(12 ) .5 12

= 2.31, p-value = P(Z > 2.31) = 1 – .9896 = .0104

There is enough evidence to infer that the population 1 is located to the right of population 2.

722


19.39

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region: z  z  / 2  z .025  1.96 or z  z / 2  z.025  1.96

E (T ) 

z

n (n  1) 55(56 )   770 ; T  4 4

n (n  1)( 2n  1)  24

55(56 )(111)  119 .35 24

T  E(T) 660  770 = = –.92, p-value = 2P(z < –.92) = 2(.1788) = .3576. There is not enough 119 .35 T

evidence to infer that the population locations differ.

19.40

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.01  2.33

E (T ) 

z

n (n  1) 108 (109 )   2943 ; T  4 4

n (n  1)( 2n  1) 108 (109 )( 217 )   326 .25 24 24

T  E(T) 3457  2943 = = 1.58, p-value = P(Z > 1.58) = 1 – .9429 = .0571. There is not 326 .25 T

enough evidence to conclude that population 1 is located to the right of the location of population 2.

19.41

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2 Rejection region: T  TU = 19 or T  TL = 2

Pair 1 2 3 4 5 6

|Difference| Ranks 4 5.5 2 3.5 2 3.5 1 1.5 4 5.5 1 1.5 _____________________________ T  = 19.5 T  = 1.5 T = 19.5. There is enough evidence to infer that the population locations differ.

19.42

Sample 1 9 12 13 8 7 10

Sample 2 5 10 11 9 3 9

Difference 4 2 2 –1 4 1

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Pair 1

Rejection region: T  TU = 39 or T  TL = 6 Sample 1 Sample 2 Difference |Difference| 18.2 18.2 0 0 723

Ranks


2 3 4 5 6 7 8 9 10 11 12

14.1 24.5 11.9 9.5 12.1 10.9 16.7 19.6 8.4 21.7 23.4

14.1 23.6 12.1 9.5 11.3 9.7 17.6 19.4 8.1 21.9 21.6

0 .9 –.2 0 .8 1.2 –.9 .2 .3 –.2 1.8

0 .9 6.5 .2 2 0 .8 5 1.2 8 .9 6.5 .2 2 .3 4 .2 2 1.8 9 _____________________________  T  = 34.5 T = 10.5 T = 34.5. There is not enough evidence to conclude that the population locations differ.

19.43

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

New - Leading 46 30 24 1.84 0.0332 1.6449 0.0664 1.96

a z = 1.84, p-value = .0332. There is enough evidence to indicate that the new beer is more highly rated than the leading brand. b The printout is identical to that of part a. c All codes that preserve the order produce the same results.

19.44

H 0 : H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

724


A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Brand A - Brand B 21 15 14 1.00 0.1587 1.6449 0.3174 1.96

a z = 1.00, p-value = .1587. There is no evidence to infer that Brand A is preferred. b The printout is identical to that of part a. c All codes that preserve the order produce the same results.

19.45

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

European - American 17 6 2 2.29 0.0109 1.6449 0.0218 1.96

a z = 2.29, p-value = .0109. There is enough evidence to infer that the European car is perceived to be more comfortable. b The results are identical. All codes that preserve the order produce the same results.

19.46

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

725


A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Sample 1 - Sample 2 51 74 0 -2.06 0.0198 1.6449 0.0396 1.96

a. z = –2.06, p-value = .0396. There is enough evidence to infer that the population locations differ. b

A B C D 1 Wilcoxon Signed Rank Sum Test 2 3 Difference Sample 1 - Sample 2 4 5 T+ 3726.5 6 T4148.5 125 7 Observations (for test) -0.52 8 z Stat 0.3016 9 P(Z<=z) one-tail 1.6449 10 z Critical one-tail 0.6032 11 P(Z<=z) two-tail 1.96 12 z Critical two-tail

E

z = –.52, p-value = .6032. There is not enough evidence to infer that the population locations differ. c The sign test ignores the magnitudes of the paired differences whereas the Wilcoxon signed rank sum test does not.

19.47

H 0 : The two population locations are the same

H1 : The location of population 1 is different from the location of population 2

726


A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Sample 1 - Sample 2 19 29 22 -1.44 0.0745 1.6449 0.149 1.96

a z = –1.44, p-value .1490. There is not enough evidence to infer that the population locations differ. b

A B C D 1 Wilcoxon Signed Rank Sum Test 2 Sample 1 - Sample 2 3 Difference 4 304 5 T+ 872 6 T48 7 Observations (for test) -2.91 8 z Stat 0.0018 9 P(Z<=z) one-tail 1.6449 10 z Critical one-tail 0.0036 11 P(Z<=z) two-tail 1.96 12 z Critical two-tail z = –2.91, p-value = .0036. There is enough evidence to conclude that the population locations differ. c The sign test ignores the magnitudes of the paired differences whereas the Wilcoxon signed rank sum test does not.

19.48

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.05  1.645

E (T ) 

n (n  1) 72 (72  1)   1314 ; T  4 4

n (n  1)( 2n  1)  24

72 (72  1)( 2[72 ]  1)  178 .2 24

T  E(T) 378 .5  1314 = = –5.25, p-value = P(Z < –5.25) = 0. There is enough evidence to 178 .2 T infer that the drug is effective. z

727


19.49

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.01  2.33

E (T ) 

z

n (n  1) 40 (40  1)  410 ; T   4 4

n (n  1)( 2n  1)  24

40 (40  1)( 2[40 ]  1)  74 .4 24

T  E(T) 62  410 =  4.68, p-value = P(z < –4.68) = 0. There is enough evidence to infer 74 .4 T

that women are doing less housework now than last year.

19.50

H 0 : The two population locations are the same

H 1 : The location of population 1 is to the right of the location of population 2 Rejection region: z  z  z.05  1.645

x  .5n

60  .5(98 )

 2.22, p-value = P(Z > 2.22) = 1 – .9868 = .0132. There is enough .5 n .5 98 evidence to conclude that concern about a gasoline shortage exceeded concern about an electricity shortage.

z=

19.51

=

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region T ≤ TL = 73 or T ≥ TU = 203 T = 40.5. There is enough evidence of a difference between machines.

19.52

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: T ≤ TL = 110 T = 111. There is not enough evidence to infer that the swimming department has higher gross sales.

19.53

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.10  1.28

x  .5n

30  .5(38)

 3.57 , p-value = P(Z > 3.57) = 0. There is enough evidence to conclude .5 38 .5 n that the European brand is preferred. z=

=

728


19.54

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.01  2.33

x  .5n

5  .5(20 )

 2.24, p-value = P(Z < –2.24) = .0125. There is not enough evidence to .5 n .5 20 conclude that children feel less pain. z=

19.55

=

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

Rejection region T ≤ TL = 90 or T ≥ TU = 235 T = 190. There is not enough evidence of a difference in salary offers between men and women

19.56

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Rejection region: z  z  z.05  1.645

x  .5n

32  .5(53)

 1.51, p-value = P(Z > 1.51) = 1 – .9345 = .0655. There is not enough .5 n .5 53 evidence to infer that preference should be given to students for high school 1.

z=

19.57

=

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

Rejection region: z  z  z.05  1.645

E (T ) 

z

n (n  1) 39 (39  1)  390 ; T   4 4

n (n  1)( 2n  1)  24

39 (39  1)( 2[39 ]  1)  71 .7 24

T  E(T) 48  390  4.77 , p-value = P(Z < –4.77) = 0. There is enough evidence to = 71 .7 T

support the belief. 19.58

H0: The two population locations are the same H1: The location of population 1 PARSOL) is to the right of the location of population 2 (KIDDSOL)

729


z = 5.17, p-value = 0. There is enough evidence to conclude that Americans are more optimistic about their children than themselves. 19.59

H0: The two population locations are the same H1: The location of population 1 (DEGREE) is different from the location of population 2 (SPDEG)

z = 1.27 p-value = .2032. There is not enough evidence to conclude that married couples have different amounts of education.

19.60

H 0 : The locations of all 3 populations are the same. H 1 : At least two population locations differ.

Rejection region: H  2 , k 1  .205,2 = 5.99

730


 12 H  n (n  1)

 12  984 2 1502 2 1430 2  Tj2     3(88  1) = 1.56, p    3(n  1)   36 29  n j   88(88  1)  23

value = .4584. There is not enough evidence to conclude that the population locations differ.

19.61

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1  .201,3 = 11.3

 12  T j2   1207 2 1088 2 1310 2 1445 2  12    3(100  1)   3(n  1)      H  25   nj  100 ( 100 1 ) 25 25 25  n (n  1)      = 3.28, p-value = .3504. There is not enough evidence to conclude that the population locations

differ.

19.62

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1  .210, 2 = 4.61  12 H  n (n  1)

  3741 2 1610 2 4945 2  Tj2  12    3(143  1) = 6.30, p-value   3(n  1)     n j  29 67  143 (143  1)  47

= .0429. There is enough evidence to conclude that the population locations differ.

19.63

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1  .205, 2 = 5.99 1 Rank 27 8 33 12.5 18 4.5 29 10 41 15.5 52 17 75 18 _ T 1 = 85.5  12 H  n (n  1)

.

2 37 12 17 22 30

Rank 14 1.5 3 7 11

3 19 12 33 41 28 18

T 2 = 36.5 Tj2 

Rank 6 1.5 12.5 15.5 9 4.5 T 3 = 49

 12  85 .5 36 .5 49 2       3(n  1)     3(18  1) = 3.03, p-value = n j  5 6  18 (18  1)  7 2

2

.2195. There is no evidence to conclude that at least two population locations differ.

19.64

H 0 : The locations of all 3 populations are the same.

731


H1 : At least two population locations differ. Rejection region: H  2 , k 1  .205, 2 = 5.99 1 25 15 20 22 23

Rank 10.5 1 3 6 8.5 T 1 = 29

 12 H  n (n  1)

.

2 Rank 19 2 21 4 23 8.5 22 6 28 13.5 T 2 = 34

3 27 25 22 29 28

Rank 12 10.5 6 15 13.5 T 3 = 57

 12  29 2 34 2 57 2  Tj2    3(15  1) = 4.46, p-value = .1075.      3(n  1)   n j  5 5  15(15  1)  5

There is not enough evidence to conclude that at least two population locations differ.

19.65

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

A B C 1 Kruskal-Wallis Test 2 3 Group Rank Sum Observations 4 Printer 1 4889.5 50 5 Printer 2 5350 50 6 Printer 3 4864.5 50 7 Printer 4 4996 50 8 9 H Stat 0.899 10 df 3 11 p-value 0.8257 12 chi-squared Critical 7.8147

D

a H = .899, p-value = .8257. There is not enough evidence to conclude that differences exist between the ratings of the four printings. b The printout is identical to that of part a. c All codes that preserve the order produce the same results.

19.66

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .210,3 = 6.25 Block 1 2 3 4

1 Rank 10 2 8 2 13 2 9 1.5

2 12 10 14 9

Treatment Rank 3 3 3 1.5

3 15 11 16 12

Rank 4 4 4 3 732

4 9 6 11 13

Rank 1 1 1 4


5

7

1 T 1 = 8.5

8

2 T 2 = 12.5

14

4 T 3 = 19

10 3 T 4 = 10

k     12 12 Fr   Tj2   3b(k  1)   (8.5 2  12 .5 2  19 2  10 2   3(5)(5) = 7.74,  b(k )( k  1) j1    (5)( 4)(5)  

p-value = .0517. There is enough evidence to infer that at least two population locations differ.

19.67

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .205, 2 = 5.99 Block 1 2 3 4 5

1 7.3 8.2 5.7 6.1 5.9

Rank 2 3 1 1 1 T1 = 8

Treatment 2 Rank 6.9 1 7.0 1 6.0 2 6.5 2 6.1 2 T2 = 8

3 Rank 8.4 3 7.3 2 8.1 3 9.1 3 8.0 3 T 3 = 14

k     12 12 Fr   Tj2   3b(k  1)   (8 2  8 2  14 2   3(5)( 4) = 4.8, p-value = .0907.  b(k )( k  1) j1    (5)(3)( 4)  

There is not enough evidence to infer that at least two population locations differ.

19.68

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

A B 1 Friedman Test 2 3 Group 4 Brand A 5 Brand B 6 Brand C 7 Brand D 8 9 Fr Stat 10 df 11 p-value 12 chi-squared Critical

C

Rank Sum 65 65 85 85 8.00 3 0.0460 7.8147

a Fr = 8.00, p-value = .0460. There is enough evidence to infer that differences exist between the ratings of the four brands of coffee. b Printout is identical to that of part a. c Different codes produce identical results provided the codes are in order.

733


19.69

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

A B 1 Friedman Test 2 3 Group 4 Manager 1 5 Manager 2 6 Manager 3 7 Manager 4 8 9 Fr Stat 10 df 11 p-value 12 chi-squared Critical

C

Rank Sum 21 10 24.5 24.5 10.613 3 0.014 7.8147

a Fr = 10.613, p-value = .0140. There is enough evidence to infer that differences exist between the ratings of the four managers. b The results are identical because the codes are in order.

19.70

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205, 2 = 5.99 k  12  767 .5 2 917 2 1165 2  Tj2  12   3(75  1) = 6.81,    3(n  1) = H   75(75  1)  25 25 25   n (n  1) j1 n j   

p-value = .0333. There is enough evidence to infer that there are differences in student satisfaction between the teaching methods.

19.71

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205,3 = 7.81

734


k  12 Tj2    3(n  1) = H  n (n  1) j1 n j   

 17 ,116 .5 2 16,816 .5 2 17 ,277 2 29,391 2  12   3(401  1) = 6.65, p-value = .0838.     401(401  1)  80 90 77 154  There is not enough evidence to infer that there are differences between the four groups of GMAT scores.

19.72

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .205,3 = 7.81 Judge 1 2 3 4 5 6 7 8 9 10

1 3 2 4 3 2 4 3 2 4 2

Rank 1.5 1 3 2 1 2 1.5 1 2.5 1 T1 = 16.5

Orange Juice Brand 2 Rank 3 Rank 5 4 4 3 3 2 5 4 4 3 3 1 4 3 5 4 4 3.5 4 3.5 5 3.5 5 3.5 3 1.5 4 3.5 3 3 3 3 3 1 5 4 4 3 5 4 T2 = 27.5 T3 = 33.5

4 3 4 4 2 3 3 4 3 4 3

Rank 1.5 3 3 1 2 1 3.5 3 2.5 2 T4 = 22.5

k     12 12 (16 .52  27 .52  33 .52  22 .52   3(10 )( 5) = Tj2   3b(k  1)   Fr    b(k )( k  1) j1    (10 )( 4)( 5)  

9.42, p-value = .0242. There is enough evidence to infer that differences in sensory perception exist between the four brands of orange juice. 19.73a The randomized block experiment of the analysis of variance and the Friedman test should be considered. The analysis of variance requires the number of pedestrians to be normally distributed. b

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .205, 2 = 5.99

735


k     12 12 Fr   Tj2   3b(k  1)   (46 2  72 2  62 2   3(30 )( 4) = 11.47, p-value =  b(k )( k  1) j1   (30 )(3)( 4)   

.0032. There is enough evidence to infer that there are differences in the number of people passing between the three locations.

19.74

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr   2 ,k 1 =  .205, 2 = 5.99 k     12 12 Fr   Tj2   3b(k  1)   (28 .5 2  22 .5 2  21 2   3(12 )( 4) = 2.63, ( 12 )( 3 )( 4 )  b ( k )( k 1 )     j1  

p-value = .2691. There is not enough evidence to infer that there are differences in delivery times between the three couriers. 19.75a The randomized block experimental design of the analysis of variance and the Friedman test. b

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .205, 2 = 5.99 Job Advertised Receptionist Systems analyst Junior secretary Computer programmer Legal secretary Office manager

1 14 8 25 12 7 5

Rank 2 2 3 2 2 2 T1 = 13

Newspaper 2 Rank 17 3 9 3 20 1 15 3 10 3 9 3 T2 = 16

3 12 6 23 10 5 4

Rank 1 1 2 1 1 1 T3 = 7

k    12  12 Fr   Tj2   3b(k  1)   (13 2  16 2  7 2   3(6)( 4) = 7.00, p-value = ( 6 )( 3 )( 4 )  b(k )( k  1) j1     

.0302. There is enough evidence to infer that differences exist between the newspapers.

19.76

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205,3 = 7.81

736


k  12 Tj2    3(n  1) = H  n (n  1) j1 n j   

 2195 2 1650 .5 2 2830 2 2102 .5 2  12    3(132  1)    132 (132  1)  33 34 34 31  = 14.04, p-value = .0029. There is enough evidence to conclude that there are differences in grading standards between the four high schools.

19.77

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr  2 , k 1 = .205,3 = 7.81 k     12 12 Tj2   3b(k  1)   Fr   (59 .5 2  63 .5 2  64 2  63 2   3(25)(5) = .300,  b(k )( k  1) j1    (25)( 4)(5)  

p-value = .9600. There is not enough evidence to infer that differences exist between the four drugs.

19.78

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: Fr   2 ,k 1 =  .205, 2 = 5.99 k     12 12 Fr   Tj2   3b(k  1)   (33 2  39 .5 2  47 .5 2   3(20 )( 4) = 5.28,  b(k )( k  1) j1   (20 )(3)( 4)   

p-value = .0715. There is not enough evidence to infer that there are differences in the ratings of the three recipes.

19.79 H 0 : The locations of all 3 populations are the same.

H1 : At least two population locations differ. Rejection region: H  2 , k 1 =  .205,2 = 5.99 k  12 Tj2   13850 .52 14909 .52 16390 2  12    3(300  1)   3(n  1) = H   100  100 300 (300  1)  100  n (n  1) j1 n j   

=4.32, p-value = .1151. There is not enough evidence to infer that differences exist between the three shifts. 19.80a The one-way analysis of variance and the Kruskal-Wallis test should be considered. 737


b

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205,3 = 7.81 k  12  4180 2 5262 2 5653 2 5005 2  Tj2  12    3(200  1)   3(n  1) = H    200 (200  1)  50 50 50 50   n (n  1) j1 n j   

= 6.96, p-value = .0733. There is not enough evidence to infer that differences exist between the speeds at which the four brands perform. 19.81

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

Rejection region: H   2,k 1 =  .205, 2 = 5.99 k  12 Tj2   1565 2 1358 .52 1171 .52  12    3(90  1) = 3.78, p  3(n  1) = H   30 30  90 (90  1)  30  n (n  1) j1 n j   

value = .1507. There is not enough evidence to infer that Democrat’s ratings of their chances changed over the 3–month period.

19.82

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205,3 = 7.81 k  12 Tj2    3(n  1) = H  n (n  1) j1 n j   

 21,246 2 19,784 2 20,976 18,194 2  12    3(400  1) = 4.34, p-value = .2269.    400 (400  1)  100 100 100 100  There is not enough evidence to infer that differences in believability exist between the four ads.

19.83

H 0 : The locations of all 4 populations are the same. H1 : At least two population locations differ.

Rejection region: H  2 , k 1 =  .205,3 = 7.81

738


k  12 Tj2    3(n  1) = H  n (n  1) j1 n j   

 28,304 2 21,285 2 21,796 2 20,421 2  12    3(428  1) = 4.64, p-value = .1999.    428 (428  1)  123 109 102 94  There is not enough evidence to infer that there are differences in support between the four levels of students. 19.84

H0: The locations of all 5 populations are the same. H1: At least two population locations differ.

Rejection region: H  2 , k 1 =  .205, 4 = 9.49 k  12 Tj2    3(n  1) H  n (n  1) j1 n j   

=

 638 .5 2 1233 .5 2 1814 .5 2 3159 .5 2 2065 2  12    3(133  1) = 18.73, p-value =     133 (133  1)  18 14 26 42 33 

.0009. There is enough evidence to infer that differences in perceived ease of use between the five brands of scanners. 19.85

H0: The locations of all 5 populations are the same. H1: At least two population locations differ.

H = 1637, p-value = 0. There is enough evidence to conclude that there are differences in job satisfaction between the five groups of American adults.

739


19.86

H0: The locations of all 3 populations are the same. H1: At least two population locations differ.

H = 22.27, p-value = 0. There is enough evidence to infer that differences in perception exist between the three races. 19.87

H0: The locations of all 5 populations are the same. H1: At least two population locations differ.

H = 7.11, p-value = .1302. There is not enough evidence to infer that there are differences between the five categories with respect to health. 19.88

H0: The locations of all 5 populations are the same. H1: At least two population locations differ.

740


H = 11.55, p-value = .0210. There is enough evidence to conclude that there are differences in perceived likelihood of losing their jobs between the five degree categories. 19.89

H0: The locations of all 3 populations are the same. H1: At least two population locations differ.

H = 12.04, p-value = .0024. There is enough evidence to infer that there are differences in perceived likelihood of losing their jobs between the three races. 19.90

H0: The locations of all 3 populations are the same. H1: At least two population locations differ.

741


H = 14.39, p-value = .0008. There is enough evidence to conclude that there are differences in their views about the federal income tax they have to pay between Democrats, Independents, and Republicans. 19.91

H0: The locations of all 3 populations are the same. H1: At least two population locations differ.

H = 37.99, p-value = 0. There is enough evidence to conclude that there are differences in their views about the federal income tax they have to pay between liberals, moderates, and conservatives. 19.92

H0: The locations of all 3 populations are the same. H1: At least two population locations differ.

742


H = 1.91, p-value = .3858. There is not enough evidence to infer that the races differ with respect to health. 19.93

H0: The locations of all 5 populations are the same. H1: At least two population locations differ.

H = 122.8, p-value = 0. There is enough evidence to infer that there are differences between the five educational attainment groups with respect to the newspaper readership. 19.94 Rejection region: z  z  / 2  z .025  1.96 or z  z  / 2  z .025  1.96 z  rS n  1  (.23) 50  1  1.61, p-value = 2P(Z > 1.61) = 2(1 – .9463) = .1074 There is not

enough evidence to reject the null hypothesis.

19.95

H 0 : S  0 H1 :  S  0 743


Rejection region: rS  .497 rS  .15. There is not enough evidence to conclude that there is a positive relationship between the two variables.

19.96

H 0 : S  0 H1 : S  0

Mathematics 4 2 5 4 2 2 1 Totals n

n

a i = 28

i 1

a2 30.25 9 49 30.25 9 9 1 137.5

a Economics b 5.5 5 6.5 3 2 1.5 7 3 4 5.5 5 6.5 3 3 4 3 3 4 1 2 1.5 28 28

n

n

b i = 28

i 1

b2 42.25 2.25 16 42.25 16 16 2.25 137.0

a i2 = 137.5

i 1

ab 35.75 4.5 28 35.75 12 12 1.5 129.5

n

b i2 = 137.0

 a b = 129.5 i

i

i 1

i 1

n n   ai bi   n (28 )( 28  1  1   129 .5   2.917 s ab  a i b i  i 1 i 1  = 7  7  1  n n  1  i 1     

 

2   n      ai  n     2  1  a i2   i 1   = 1 137 .5  (28)   4.250 , s a  s a2  4.250  2.062 s a2  n n  1  i 1  7  1  7       

2   n     bi   n     2  1  b i2   i 1   = 1 137 .0  (28)   4.167 , s b  s 2b  4.167  2.041 s 2b  n  1  i 1 n  7  1  7       

rS 

s ab 2.917   .6931 s b s b (2.062 )( 2.041)

H 0 : S  0 H1 :  S  0 Rejection region: rS  .786 or rS  .786 There is not enough evidence to infer a relationship between the grades in the two courses.

744


19.97

H 0 : S  0 H1 : S  0

Number a 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 Totals 36 n

a i = 36

i 1

Rating 4 5 3 3 3 2 3 1

b 7 8 4.5 4.5 4.5 2 4.5 1 36

n

n

b i = 36

i 1

a2 1 4 9 16 25 36 49 64 204

b2 49 64 20.25 20.25 20.25 4 20.25 1 199

n

a i2 = 204

i 1

 i 1

ab 7 16 13.5 18 22.5 12 31.5 8 128.5 n

b i2 = 199

 a b = 128.5 i

i

i 1

n n   ai bi   n (36 )(36  1  1    4.786 128 .5  s ab  a i b i  i 1 i 1  = 8  1  8  n  1  i 1 n     

 

2   n      ai  n     2  1  a i2   i 1   = 1 204  (36 )   6.000 , s a  s a2  6.000  2.450 s a2  n n  1  i 1  8  1  8       

2   n     bi   n     2  1  b i2   i 1   = 1 199  (36 )   5.286 , s b  s 2b  5.286  2.299 s 2b  n  1  i 1 n  8  1  8       

rS 

s ab  4.786   .8497 s b s b (2.450 )( 2.299 )

H 0 : S  0 H1 :  S  0 Rejection region: rS  .643 or rS  .643 There is enough evidence to infer a relationship between the number of commercials and the rating.

19.98

H 0 : S  0

745


H1 : S  0 Stock 1 a –7 4.5 –4 6.5 –7 4.5 –3 8 2 10.5 –10 2.5 –10 2.5 5 12 1 9 –4 6.5 2 10.5 6 13 –13 1 Totals 91 n

n

Stock 2 6 6 –4 9 3 –3 7 –3 4 7 9 5 –7

a i = 91

n

b i = 91

i 1

i 1

n

a i2 = 817

i 1

a2 20.25 42.25 20.25 64 110.25 6.25 6.25 144 81 42.25 110.25 169 1 817

b 8.5 8.5 2 12.5 5 3.5 10.5 3.5 6 10.5 12.5 7 1 91

 i 1

b2 72.25 72.25 4 156.25 25 12.25 110.25 12.25 36 110.25 156.25 49 1 817

ab 38.25 55.25 9 100 52.5 8.75 26.25 42 54 68.25 131.25 91 1 677.5

n

b i2 = 817

 a b = 677.5 i

i

i 1

n n   ai bi   n (91)(91  1  1   677 .5   3.375 s ab  a i b i  i 1 i 1  =   13  13  1  n n  1  i 1     

 

2     n    ai  n     2  1  a i2   i 1   = 1 817  (91)   15 .00, sa  sa2  15 .00  3.873 s a2  n n  1  i 1  13  1  13       

2   n      bi  n     2  1  b i2   i 1   = 1 817  (91)   15 .00, s b  s 2b  15 .00  3.873 s 2b  n  1  i 1 n  13  1  13       

rS 

sab 3.375   .2250 s bs b (3.873 )(3.873 )

H 0 : S  0 H1 :  S  0 Rejection region: rS  .566 or rS  .566 There is not enough evidence to infer a relationship between the two stock returns.

19.99

H 0 : S  0

746


H1 : S  0 Experience 1 17 20 9 2 13 9 23 7 10 12 24 8 20 21 19 1 22 20 11 18 14 21 21 Totals

a 1.5 13 17 6.5 3 11 6.5 23 4 8 10 24 5 17 20 15 1.5 22 17 9 14 12 20 20 300

i 1

a i = 300

b i = 300

b2 ab 4 3 256 208 256 272 462.25 139.75 36 18 256 176 121 71.5 462.25 494.5 36 24 462.25 172 462.25 215 36 144 36 30 462.25 365.5 121 220 36 90 4 3 121 242 256 272 121 99 462.25 301 256 192 121 220 4 40 4,850.5 4,012.25 n

n

a i2 = 4,895

i 1

i 1

a2 2.25 169 289 42.25 9 121 42.25 529 16 64 100 576 25 289 400 225 2.25 484 289 81 196 144 400 400 4,895

b 2 16 16 21.5 6 16 11 21.5 6 21.5 21.5 6 6 21.5 11 6 2 11 16 11 21.5 16 11 2 300

n

n

n

Rating 1 4 4 5 2 4 3 5 2 5 5 2 2 5 3 2 1 3 4 3 5 4 3 1

b i2 = 4,850.5

 a b = 4,012.25 i

i

i 1

i 1

n n   a bi   n i (300 )(300 )  1  1   4,012 .25  s ab  a i b i  i 1 i 1  =   11 .40   24 24 1  n n  1  i 1       

 

2     n   ai   n     2  1  a i2   i 1   = 1 4,895  (300 )   49 .78, sa  sa2  49 .78  7.056 s a2  n n  1  i 1  24  1  24       

2   n     bi   n    2  1   b i2   i 1   = 1 4,850 .5  (300 )   47 .85, s b  s 2b  47 .85  6.917 s 2b  n  1  i 1 n  24  1  24       

747


rS 

sab 11 .40   .2336 s bs b (7.056 )( 6.917 )

Rejection region: rS  .409 or rS  −.409 There is not enough evidence to infer a relationship between experience and quality.

19.100

H 0 : S  0 H1 :  S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Price and Odometer Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

-0.0201 -0.20 0.4206 1.6449 0.8412 1.96

z = –.20, p-value = .8412. There is not enough evidence to infer that odometer reading and price are related.

19.101

H 0 : S  0 H1 :  S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Rating and Grade Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

-0.2628 -2.93 0.0017 1.6449 0.0034 1.96

z = –2.93, p-value = .0034. There is enough evidence to support the theory. 19.102

H 0 : S  0 H1 :  S  0

748


1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Age and Heartburn Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.0302 0.54 0.2931 1.6449 0.5862 1.96

z = .54, p-value = .2931. There is not enough evidence to conclude that age and severity of heartburn are positively related.

19.103

H 0 : S  0 H1 : S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Test and Length Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.5460 4.19 0 1.6449 0 1.96

z = 4.19, p-value = 0. There is enough evidence to conclude that the longer the commercial the higher the test score will be.

19.104

H 0 : S  0 H1 : S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Floor and Price Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.553 3.87 0.0001 1.6449 0.0002 1.96

749


z = 3.87, p-value = .0001. There is sufficient evidence to conclude that price and floor number are positively related.

19.105

H 0 : S  0 H1 : S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Cigarettes and Taste Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

-0.3026 -5.05 0 1.6449 0 1.96

z = –5.05, p-value = 0. There is sufficient evidence to conclude that the more a person smokes the less taste sensation he or she has. 19.106 H0: ρs= 0 H1: ρs> 0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Wager and Enjoyment Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.3912 5.52 0 1.6449 0 1.96

z = 5.52, p-value = 0. There is enough evidence to infer that the greater the wager the more enjoyable the game is. 19.107 H0: ρs= 0 H1: ρs > 0

750


z = 2.45, p-value = .0144. There is enough evidence to conclude that as one gets older the probability of losing one’s job decreases. 19.108 H0: ρs= 0 H1: ρs < 0

z = -10.24, p-value = 0. There is enough evidence to infer that more educated people read newspapers more often.

19.109 H0: ρs= 0 H1: ρs < 0

751


z = -6.18, p-value = 0. There is enough evidence to infer that more satisfying jobs have higher incomes.

19.110 H0: ρs= 0 H1: ρs < 0

z = -6.59, p-value = 0. There is enough evidence to conclude that jobs that are most secure are also most satisfying.

19.111 H0: ρs= 0 H1: ρs < 0

752


z = -13.09, p-value = 0. There is enough evidence to conclude that more educated people perceive themselves as healthier.

19.112 H0: ρs= 0 H1: ρs < 0

z = -.435, p-value = .3318. There is not enough evidence to conclude that more educated people are more likely to believe that their standard of living is better than that of their parents.

19.113 H0: ρs= 0 H1: ρs < 0

753


z = -10.77, p-value = 0. There is enough evidence to conclude that younger people read newspapers less frequently than do older people. 19.114 H0: ρs= 0 H1: ρs > 0

z = 3.04, p-value = .0012. There is enough evidence to conclude that the longer one works, the probability of losing one’s job decreases. 19.115 H0: ρs= 0 H1: ρs ≠ 0

754


z = 1.10, p-value = .2698. There is not enough evidence to conclude that age affects one’s belief concerning the federal income tax one must pay.

19.116 H0: ρs= 0 H1: ρs < 0

z = -4.86, p-value = 0. There is enough evidence to conclude that higher income individuals are healthier.

19.117 H0: ρs= 0 H1: ρs> 0

z = 3.90, p-value = 0. There is enough evidence to conclude that richer Americans are pessimistic about their children’s chances of having a better standard of living.

19.118

H 0 : S  0 H1 : S  0 755


1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Education and Income Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.5742 9.64 0 1.6449 0 1.96

z = 9.64, p-value = 0. There is sufficient evidence to conclude that more education and higher incomes are linked.

19.119

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 Section 1 15297.5 113 5 Section 2 14592.5 131 6 z Stat 2.65 7 P(Z<=z) one-tail 0.0041 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.0082 10 z Critical two-tail 1.96

E

z = 2.65, p-value = .0082. There is enough evidence to infer that the two teaching methods differ

19.120

H 0 : The two population locations are the same H 1 : The location of population 1 is left of the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 New 207.5 15 5 Existing 257.5 15 6 z Stat -1.04 7 P(Z<=z) one-tail 0.1499 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.2998 10 z Critical two-tail 1.96

E

z = –1.04, p-value = .1499. There is not enough evidence to infer that the new method is better. 19.121

H 0 : The locations of all 4 populations are the same. 756


H1 : At least two population locations differ.

A B 1 Friedman Test 2 3 Group 4 Typeface 1 5 Typeface 2 6 Typeface 3 7 Typeface 4 8 9 Fr Stat 10 df 11 p-value 12 chi-squared Critical

C

Rank Sum 50.5 38 66 45.5 12.615 3 0.0055 7.8147

Fr = 12.615, p-value = .0055. There is enough evidence to conclude that there are differences between typefaces.

19.122

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Drug A - Drug B 2 18 10 -3.58 0.0002 1.6449 0.0004 1.96

z = –3.58, p-value = .0002. There is enough evidence to conclude that drug B is more effective. 19.123

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

757


A B C D 1 Wilcoxon Signed Rank Sum Test 2 3 Difference Drug A - Drug B 4 5 T+ 36 6 T342 7 Observations (for test) 27 8 z Stat -3.68 9 P(Z<=z) one-tail 0.0001 10 z Critical one-tail 1.6449 11 P(Z<=z) two-tail 0.0002 12 z Critical two-tail 1.96 Note that the number of non-zero observations is less than 30, invalidating the use of z statistic. T = 36, rejection region: T ≤ 120. There is enough evidence to conclude that drug B is more effective. 19.124a The one-way analysis of variance and the Kruskal-Wallis test should be considered. If the data are normal apply the analysis of variance, otherwise use the Kruskal-Wallis test. b

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 827 25 4 Binding 1 1110 25 5 Binding 2 913 25 6 Binding 3 7 3.55 8 H Stat 2 9 df 0.1699 10 p-value 5.9915 11 chi-squared Critical

D

H = 3.55, p-value = .1699. There is not enough evidence to infer that there are differences between bindings.

19.125

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

758


A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 New Material 2747 50 5 Old Material 2303 50 6 z Stat 1.53 7 P(Z<=z) one-tail 0.063 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.126 10 z Critical two-tail 1.96

E

z = 1.53, p-value = .0630. There is not enough evidence to conclude that the new material takes longer to burst into flames.

19.126

H 0 : The locations of all 7 populations are the same. H1 : At least two population locations differ.

A B C 1 Kruskal-Wallis Test 2 3 Group Rank Sum Observations 4 Sunday 10060 63 5 Monday 2977 26 6 Tuesday 2932.5 29 7 Wednesday 3834.5 31 8 Thursday 4060.5 30 9 Friday 6045 42 10 Saturday 6405.5 48 11 12 H Stat 14.87 13 df 6 14 p-value 0.0213 15 chi-squared Critical 12.5916

D

H = 14.87, p-value = .0213. There is enough evidence to infer that there are differences in the perceptions of speed of service between the days of the week.

19.127

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

759


A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

F

Commercial 1 - Commercial 2 15 21 24 -1.00 0.1587 1.6449 0.3174 1.96

z = –1.00, p-value = .3174. There is not enough evidence to infer differences in believability between the two commercials.

19.128

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B C D 1 Wilcoxon Signed Rank Sum Test 2 3 Difference Men - Women 4 5 T+ 324 6 T204 32 7 Observations (for test) 1.12 8 z Stat 0.1309 9 P(Z<=z) one-tail 1.6449 10 z Critical one-tail 0.2618 11 P(Z<=z) two-tail 1.96 12 z Critical two-tail z = 1.12, p-value = .1309. There is not enough evidence to conclude that men lose a greater percentage of their hearing than women.

19.129

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

760


A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 This Year 37525.5 200 5 10 Years Ago 42674.5 200 6 z Stat -2.23 7 P(Z<=z) one-tail 0.013 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.026 10 z Critical two-tail 1.96

E

z = –2.23, p-value = .0130. There is enough evidence to infer that people perceive newspapers as doing a better job 10 years ago than today.

19.130

H 0 : S  0 H1 :  S  0

1 2 3 4 5 6 7 8 9

A B C Spearman Rank Correlation Reference and GPA Spearman Rank Correlation z Stat P(Z<=z) one tail z Critical one tail P(Z<=z) two tail z Critical two tail

D

0.0775 1.05 0.148 1.6449 0.296 1.96

z = 1.05, p-value = .2960. There is not enough evidence to infer that the letter of reference and MBA GPA are related.

19.131

H 0 : The two population locations are the same H 1 : The location of population 1 is different from the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 Males 10336.5 100 5 Females 9763.5 100 6 z Stat 0.70 7 P(Z<=z) one-tail 0.24 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.484 10 z Critical two-tail 1.96

E

z = .70, p-value = .4840. There is not enough evidence to conclude that businesswomen and business men differ in the number of business trips taken per year. 761


19.132

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Before - After 19 5 16 2.86 0.0021 1.6449 0.0042 1.96

z = 2.86, p-value = .0021. There is enough evidence to infer that the midterm test negatively influences student opinion.

19.133

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 Rank Sum Observations 3 4 2 Years Ago 10786.5 100 5 ThisYear 9313.5 100 6 z Stat 1.80 7 P(Z<=z) one-tail 0.036 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.072 10 z Critical two-tail 1.96

E

z = 1.80, p-value = .0360. There is enough evidence to indicate that the citizens of Stratford should be concerned.

19.134

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

762


A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 Low 9055 100 5 High 11045 100 6 z Stat -2.43 7 P(Z<=z) one-tail 0.0075 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.015 10 z Critical two-tail 1.96

E

z = –2.43, p-value = .0075. There is enough evidence to conclude that boys with high levels of lead are more aggressive than boys with low levels.

19.135a H 0 : The two population locations are the same

H 1 : The location of population 1 is to the right of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

F

G

Female Professor - Male Professor 45 7 48 5.27 0 1.6449 0 1.96

z = 5.27, p-value = 0. There is enough evidence to infer that female students rate female professors higher than they rate male professors. b

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

F

Female Professor - Male Professor 21 31 48 -1.39 0.0828 1.6449 0.1656 1.96 763

G


z = –1.39, p-value = .0828. There is not enough evidence to infer that male students rate male professors higher than they rate female professors.

19.136

H 0 : The locations of all 3 populations are the same. H1 : At least two population locations differ.

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 4 Unattractive 16844.5 134 5 Neutral 13313 68 6 Attractive 26122.5 133 7 8 H Stat 42.59 9 df 2 10 p-value 0 11 chi-squared Critical 5.9915

D

H = 42.59, p-value = 0. There is enough evidence to conclude that incomes of lawyers are affected by physical attractiveness.

19.137

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

A B C D 1 Wilcoxon Rank Sum Test 2 Rank Sum Observations 3 4 Telecommuters 10934.5 100 5 Office 9165.5 100 6 z Stat 2.16 7 P(Z<=z) one-tail 0.0153 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.0306 10 z Critical two-tail 1.96

E

z = 2.16, p-value = .0153. There is enough evidence to conclude that telecommuters are more satisfied with their jobs.

19.138

H 0 : The two population locations are the same H 1 : The location of population 1 is to the left of the location of population 2

764


A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 3 Hours Before 22553.5 180 5 Closing 42426.5 180 6 z Stat -10.06 7 P(Z<=z) one-tail 0 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0 10 z Critical two-tail 1.96

E

z = –10.06, p-value = 0. There is enough evidence to conclude that alcohol impairs judgment. 19.139

H 0 : The two population locations are the same H 1 : The location of population 1 is to the right of the location of population 2

Day 1versus Before

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Before - Day 1 67 0 5 8.19 0 1.6449 0 1.96

z = 8.19, p-value = 0 Day 2 versus Before

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

Before - Day 2 59 0 13 7.68 0 1.6449 0 1.96

z = 7.68, p-value = 0 Day 3 versus Before 765

E


A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Before - Day 3 35 0 37 5.92 0 1.6449 0 1.96

z = 5.92, p-value = 0 There is enough evidence to infer that exercisers who abstain from physical activity are less happy than when they are exercising.

H 0 : The two population locations are the same

b

H 1 : The location of population 1 (Day 2) is to the left of the location of population 2 (Day 3)

A B 1 Sign Test 2 3 Difference 4 5 Positive Differences 6 Negative Differences 7 Zero Differences 8 z Stat 9 P(Z<=z) one-tail 10 z Critical one-tail 11 P(Z<=z) two-tail 12 z Critical two-tail

C

D

E

Day 2 - Day 3 3 37 32 -5.38 0 1.6449 0 1.96

z = –5.38, p-value = 0. There is enough evidence to conclude that by the third day their moods were improving. c 1. Exercisers who abstained were adjusting to their inactivity by the third day. 2. By the third day exercisers realized that they were closer to the end of their inactivity.

Case 19.1 a Wilcoxon rank sum tests

H 0 : The two population locations are the same. H 1 : The location of population 1 is to the left of the location of population 2. 766


Quality of work

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 No 875.5 26 5 Yes 8169.5 108 6 z Stat -4.95 7 P(Z<=z) one-tail 0 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0 10 z Critical two-tail 1.96

E

z = –4.95, p-value = 0. There is overwhelming evidence to infer that customers who say they will return assess quality of work higher than customers who do not plan to return. Fairness of price

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 No 1477 26 5 Yes 7568 108 6 z Stat -1.56 7 P(Z<=z) one-tail 0.0589 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.1178 10 z Critical two-tail 1.96

E

z = = –1.56, p-value = .0589. There is not enough evidence to infer that customers who say they will return assess fairness of price higher than customers who do not plan to return. Explanation of work and guarantee

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 No 1656 26 7389 108 5 Yes -0.56 6 z Stat 7 P(Z<=z) one-tail 0.2888 8 z Critical one-tail 1.6449 0.5776 9 P(Z<=z) two-tail 10 z Critical two-tail 1.96

E

z = –.56, p-value = .2888. There is no evidence to infer that customers who say they will return assess explanation of work and guarantee higher than customers who do not plan to return.

767


Checkout process

A B C D 1 Wilcoxon Rank Sum Test 2 3 Rank Sum Observations 4 No 1461 26 5 Yes 7584 108 6 z Stat -1.65 7 P(Z<=z) one-tail 0.049 8 z Critical one-tail 1.6449 9 P(Z<=z) two-tail 0.098 10 z Critical two-tail 1.96

E

z = –1.65, p-value = .0490. There is evidence to infer that customers who say they will return assess the checkout process higher than customers who do not plan to return. b Kruskal-Wallis tests

H 0 : The location of all 3 populations is the same H 1 : At least two population locations differ Quality of work

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 2495.5 33 4 Positive 1018 21 5 Negative 80 6 No commen 5531.5 7 6.63 8 H Stat 2 9 df 0.0364 10 p-value 5.9915 11 chi-squared Critical

D

H = 6.63, p-value = .0364. There is evidence to conclude that customers who make positive, negative, and no comment differ in their assessment of quality of work performed.

768


Fairness of price

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 2381.5 33 4 Positive 1143.5 21 5 Negative 5520 80 6 No commen 7 2.97 8 H Stat 2 9 df 0.2268 10 p-value 5.9915 11 chi-squared Critical

D

H = 2.97, p-value = .2268. There is no evidence to conclude that customers who make positive, negative, and no comment differ in their assessment of fairness of price Explanation of work and guarantee

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 2291 33 4 Positive 1335 21 5 Negative 5419 80 6 No commen 7 0.299 8 H Stat 2 9 df 0.8611 10 p-value 5.9915 11 chi-squared Critical

D

H = .299, p-value = .8611. There is no evidence to conclude that customers who make positive, negative, and no comment differ in their assessment of explanation of work and guarantee. Checkout process

A B C 1 Kruskal-Wallis Test 2 Rank Sum Observations 3 Group 1933.5 33 4 Positive 1200 21 5 Negative 80 6 No commen 5911.5 7 5.40 8 H Stat 2 9 df 0.0672 10 p-value 5.9915 11 chi-squared Critical

D

769


H = 5.40, p-value = .0672. There is not enough evidence to conclude that customers who make positive, negative, and no comment differ in their assessment of the checkout process.

770


Chapter 20 20.1 Time series

Moving average

48 41 37 32 36 31 43 52 60 48 41 30

(48+41+37)/3 = 42.00 (41+37+32)/3 = 36.67 (37+32+36)/3 = 35.00 (32+36+31)/3 = 33.00 (36+31+43)/3 = 36.67 (31+43+52)/3 = 42.00 (43+52+60)/3 = 51.67 (52+60+48)/3 = 53.33 (60+48+41)/3 = 49.67 (48+41+30)/3 = 39.67

20.2 Time series

Moving average

48 41 37 32 36 31 43 52 60 48 41 30

(48 +41+37+32+36)/5 = 38.8 (41+37+32+36+31)/5 = 35.4 (37+32+36+31+43)/5 = 35.8 (32+36+31+43+52)/5 = 38.8 (36+31+43+52+60)/5 = 44.4 (31+43+52+60+48)/5 = 46.8 (43+52+60+48+41)/5 = 48.8 (52+60+48+41+30)/5 = 46.2

20.3 70

60

Series1

50

Series2 40 30

Series3

20 10 0 1

2

3

4

5

6

7

8

793

9

10

11

12


20.4 Time series

Moving average

16 22 19 24 30 26 24 29 21 23 19 15

(16+22+19)/3 = 19.00 (22+19+24)/3 = 21.67 (19+24+30)/3 = 24.33 (24+30+26)/3 = 26.67 (30+26+24)/3 = 26.67 (26+24+29)/3 = 26.33 (24+29+21)/3 = 24.67 (29+21+23)/3 = 24.33 (21+23+19)/3 = 21.00 (23+19+15)/3 = 19.00

20.5 Time series 16 22 19 24 30 26 24 29 21 23 19 15

Moving average

(16+22+19+24+30)/5 = 22.2 (22+19+24+30+26)/5 = 24.2 (19+24+30+26+24)/5 = 24.6 (24+30+26+24+29)/5 = 26.6 (30+26+24+29+21)/5 = 26.0 (26+24+29+21+23)/5 = 24.6 (24+29+21+23+19)/5 = 23.2 (29+21+23+19+15)/5 = 21.4

20.6 35

30

Series1

25

Series2 20 15

Series3

10 5 0 1

2

3

4

5

6

7

8

794

9

10

11

12


20.7 Time series

Exponentially smoothed time series

12 18 16 24 17 16 25 21 23 14

12 .1(18) +. 9(12) = 12.60 .1(16) +. 9(12.60) = 12.94 .1(24) +. 9(12.94) = 14.05 .1(17) +. 9(14.05) = 14.34 .1(16) +. 9(14.34) = 14.51 .1(25) +. 9(14.51) = 15.56 .1(21) +. 9(15.56) = 16.10 .1(23) + .9(16.10) = 16.79 .1(14) + .9(16.79) = 16.51

20.8 Time series

Exponentially smoothed time series

12 18 16 24 17 16 25 21 23 14

12.00 .8(18) +. 2(12) = 16.80 .8(16) +. 2(16.80) = 16.16 .8(24) +. 2(16.16) = 22.43 .8(17) +. 2(22.43) = 18.09 .8(16) +. 2(18.09) = 16.42 .8(25) +. 2(16.42) = 23.28 .8(21) +. 2(23.28) = 21.46 .8(23) + .2(21.46) = 22.69 .8(14) + .2(22.69) = 15.74

20.9 30 25

Series1

20 Series2 15 Series3 10

5 0 1

2

3

4

5

6

7

There appears to be a gradual upward trend.

795

8

9

10


20.10 Time series

Exponentially smoothed time series

38 43 42 45 46 48 50 49 46 45

38.00 .1(43) +. 9(38) = 38.50 .1(42) +. 9(38.50) = 38.85 .1(45) +. 9(38.85) = 39.47 .1(46) +. 9(39.47) = 40.12 .1(48) +. 9(40.12) = 40.91 .1(50) +. 9(40.91) = 41.82 .1(49) +. 9(41.82) = 42.53 .1(46) + .9(42.53) = 42.88 .1(45) + .9(42.88) = 43.09

There appears to be a gradual upward trend.

20.11 Time series

Exponentially smoothed time series

38 43 42 45 46 48 50 49 46 45

38 .8(43) +. 2(38) = 42.00 .8(42) +. 2(42.00) = 42.00 .8(45) +. 2(42.00) = 44.40 .8(46) +. 2(44.40) = 45.68 .8(48) +. 2(45.68) = 47.54 .8(50) +. 2(47.54) = 49.51 .8(49) +. 2(49.51) = 49.10 .8(46) + .2(49.10) = 46.62 .8(45) + .2(46.62) = 45.32

20.12 55

Series1 50

Series2

45

40

Series3

35

30 1

2

3

4

5

6

7

There is a trend component.

796

8

9

10


20.13 & 20.14 Sales

3-Day moving average

5-Day moving average

43 45 22 25 31 51 41 37 22 25 40 57 30 33 37 64 58 33 38 25

(43+45+22)/3 = 36.67 (45+22+25)/3 = 30.67 (22+25+31)/3 = 26.00 (25+31+51)/3 = 35.67 (31+51+41)/3 = 41.00 (51+41+37)/3 = 43.00 (41+37+22)/3 = 33.33 (37+22+25)/3 = 28.00 (22+25+40)/3 = 29.00 (25+40+57)/3 = 40.67 (40+57+30)/3 = 42.33 (57+30+33)/3 = 40.00 (30+33+37)/3 = 33.33 (33+37+64)/3 = 44.67 (37+64+58)/3 = 53.00 (64+58+33)/3 = 51.67 (58+33+38)/3 = 43.00 (33+38+25)/3 = 32.00

(43+45+22+25+31)/5 = 33.20 (45+22+25+31+51)/5 = 34.80 (22+25+31+51+41)/5 = 34.00 (25+31+51+41+37)/5 = 37.00 (31+51+41+37+22)/5 = 36.40 (51+41+37+22+25)/5 = 35.20 (41+37+22+25+40)/5 = 33.00 (37+22+25+40+57)/5 = 36.20 (22+25+40+57+30)/5 = 34.80 (25+40+57+30+33)/5 = 37.00 (40+57+30+33+37)/5 = 39.40 (57+30+33+37+64)/5 = 44.20 (30+33+37+64+58)/5 = 44.40 (33+37+64+58+33)/5 = 45.00 (37+64+58+33+38)/5 = 46.00 (64+58+33+38+25)/5 = 43.60

70 Series1

60 50

Series2

40 30

Series3

20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

c There appears to be a seasonal (weekly) pattern.

797


20.15a Sales 18

4-quarter moving average

Centered moving average

22 (18+22+27+31)/4 = 24.50 27

(24.50+28.25)/2 = 26.375 (22+27+31+33)/4 = 28.25

31

(28.25+27.75)/2 = 28.000 (27+31+33+20)/4 = 27.75

33

(27.75+30.50)/2 = 29.125 (31+33+20+38)/4 = 30.50

20

(30.50+29.25)/2 = 29.875 (33+20+38+26)/4 = 29.25

38

(29.25+27.25)/2 = 28.250 (20+38+26+25)/4 = 27.25

26

(27.25+31.25)/2 = 29.250 (38+26+25+36)/4 = 31.25

25

(31.25+32.75)/2 = 32.000 (26+25+36+44)/4 = 32.75

36

(32.75+33.50)/2 = 33.125 (25+36+44+29)/4 = 33.50

44

(33.50+37.50)/2 = 35.500 (36+44+29+41)/4 = 37.50

29

(37.50+36.75)/2 = 37.125 (44+29+41+33)/4 = 36.75

41

(36.75+38.75)/2 = 37.750 (29+41+33+52)/4 = 38.75

33

(38.75+42.75)/2 = 40.750 (41+33+52+25)/4 = 42.75

52 45 b 60 50 Series1

40 30

Series2

20

10 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16

c There appears to be a gradual trend of increasing sales.

798


20.16 & 20.17 Sales

Exponentially smoothed w = .4

Exponentially smoothed w = .8

18 22 27 31 33 20 38 26 25 36 44 29 41 33 52 45

18.00 .4(22)+.6(18) = 19.60 .4(27)+.6(19.6) = 22.56 .4(31)+.6(22.56) = 25.94 .4(33)+.6(25.94) = 28.76 .4(20)+.6(28.76) = 25.26 .4(38)+.6(25.26) = 30.35 .4(26)+.6(30.35) = 28.61 .4(25)+.6(28.61) = 27.17 .4(36)+.6(27.17) = 30.70 .4(44)+.6(30.70) = 36.02 .4(29)+.6(36.02) = 33.21 .4(41)+.6(33.21) = 36.33 .4(33)+.6(36.33) = 35.00 .4(52)+.6(35.00) = 41.80 .4(45)+.6(41.80) = 43.08

18.00 .8(22)+.2(18) = 21.20 .8(27)+.2(21.2) = 25.84 .8(31)+.2(25.84) = 29.97 .8(33)+.2(29.97) = 32.39 .8(20)+.2(32.39) = 22.48 .8(38)+.2(22.48) = 34.90 .8(26)+.2(34.90) = 27.78 .8(25)+.2(27.78) = 25.56 .8(36)+.2(25.56) = 33.91 .8(44)+.2(33.91) = 41.98 .8(29)+.2(41.98) = 31.60 .8(41)+.2(31.60) = 39.12 .8(33)+.2(39.12) = 34.22 .8(52)+.2(34.22) = 48.44 .8(45)+.2(48.44) = 45.69

20.16 b 60 50

Series1

40 30

Series2

20

10 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16

799


20.17 b 60 50 Series1

40 30

Series2

20

10 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16

20.18

Line chart 25

Time series

20 15 10

5 0 1

2

3

4

5 Period

The quadratic model would appear to be the best model.

800

6

7

8


20.19

Line chart 60

Time series

50

40 30 20 10 0 1

2

3

4

5

6

Period

The linear trend model appears to be best.

20.20 ŷ  4.96  2.38 t

(R 2  .81)

ŷ  3.14  2.48 t  .54 t 2

(R 2  .98)

The quadratic trend line fits better.

20.21 ŷ  63 .87  3.94 t

(R 2  .94 )

ŷ  57 .2  .61t  .30 t 2

(R 2  .98)

The quadratic trend line fits slightly better.

801

7

8

9

10


20.22 Week

Day

Period t

y

y / yˆ

1

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

12 18 16 25 31 11 17 19 24 27 14 16 16 28 25 17 21 20 24 32

17.2 17.5 17.9 18.3 18.6 19.0 19.4 19.7 20.1 20.5 20.8 21.2 21.6 21.9 22.3 22.7 23.0 23.4 23.8 24.1

0.699 1.027 0.894 1.369 1.664 0.579 0.878 0.963 1.194 1.320 0.672 0.755 0.742 1.277 1.122 0.750 0.912 0.855 1.010 1.327

Monday .699 .579 .672 .750 .675

Tuesday 1.027 .878 .755 .912 .893

Day Wednesday .894 .963 .742 .855 .864

Thursday 1.369 1.194 1.277 1.010 1.213

Friday 1.664 1.320 1.122 1.327 1.358

5.003

.675

.892

.864

1.212

1.357

5.000

2

3

4

Week 1 2 3 4 Average Seasonal Index

802

Total


20.23 Year

Quarter

y

y / yˆ

1

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

55 44 46 39 41 38 37 30 43 39 39 35 36 32 30 25 50 25 24 22

46.6 45.6 44.5 43.5 42.4 41.3 40.3 39.2 38.2 37.1 36.0 35.0 33.9 32.9 31.8 30.7 29.7 28.6 27.6 26.5

1.179 0.965 1.033 0.897 0.967 0.919 0.919 0.765 1.127 1.051 1.082 1.001 1.061 0.974 0.943 0.813 1.685 0.874 0.871 0.830

2

3

4

5

Year 1 2 3 4 5 Average Seasonal Index

1 1.179 0.967 1.127 1.061 1.685 1.204

2 0.965 0.919 1.051 0.974 0.874 0.957

Quarter 3 1.033 0.919 1.082 0.943 0.871 0.970

1.207

0.959

0.972

803

4 0.897 0.765 1.001 0.813 0.830 0.861

Total

3.991

0.863

4.000


20.24 Year

Quarter

Period t

y

y / yˆ

2001

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

52 67 85 54 57 75 90 61 60 77 94 63 66 82 98 67

62.9 64.1 65.3 66.5 67.7 68.8 70.0 71.2 72.4 73.6 74.7 75.9 77.1 78.3 79.5 80.6

0.827 1.045 1.302 0.812 0.842 1.090 1.286 0.857 0.829 1.046 1.258 0.830 0.856 1.047 1.233 0.831

2002

2003

2004

Year 2001 2002 2003 2004 Average Seasonal Index

1 .827 .842 .829 .856 .838

2 1.045 1.090 1.046 1.047 1.057

Quarter 3 1.302 1.286 1.258 1.233 1.270

4 .812 .857 .830 .831 .833

3.998

.839

1.058

1.270

.833

4.000

Total

20.25a

Scatter Diagram 18 000 17 500

y = 253,78x - 491595

17 000 Enrollment

16 500 16 000

15 500

15 000 14 500 14 000 13 500 1992

1994

1996

1998

2000 Year

b ŷ  491,595  253 .78 Year

804

2002

2004

2006

2008


20.26a

Balance of Trade

Scatter Diagram 0 -100 -200 -300 -400 -500 -600 -700 -800 -900 1975

y = -25,039x + 49661

1980

1985

1990

1995

2000

2005

2010

20

25

30

Year b ŷ = 49,661 – 25.039Year

20.27

Scatter Diagram 400

y = 7.416x + 143.0

350

Subscribers

300 250 200 150 100 50 0 0

5

10

15 Quarter

ŷ = 143 + 7.42 t

805


Year

Quarter

Period t

y

ŷ =143+7.42t

y / yˆ

1

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

184 173 160 189 191 185 184 200 205 192 200 229 236 219 211 272 280 261 275 322 331 301 306 351

150.42 157.84 165.26 172.68 180.1 187.52 194.94 202.36 209.78 217.2 224.62 232.04 239.46 246.88 254.3 261.72 269.14 276.56 283.98 291.4 298.82 306.24 313.66 321.08

1.223 1.096 0.968 1.095 1.061 0.987 0.944 0.988 0.977 0.884 0.890 0.987 0.986 0.887 0.830 1.039 1.040 0.944 0.968 1.105 1.108 0.983 0.976 1.093

1 1.223 1.060 .977 .986 1.040 1.108 1.066

2 1.096 .987 .884 .887 .944 .983 .964

Quarter 3 .968 .944 .890 .830 .968 .976 .929

4 1.095 .988 .987 1.039 1.105 1.093 1.051

4.010

1.063

.962

.927

1.048

4.000

2

3

4

5

6

Year 1 2 3 4 5 6 Average Seasonal Index

20.28 Regression line: ŷ = 145 + 1.66 t

806

Total


Week

Day

Period t

y

ŷ =145+1.66t

y / yˆ

1

1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

240 85 93 106 125 188 314 221 80 75 121 110 202 386 235 86 74 100 117 205 402 219 91 102 89 105 192 377

146.66 148.32 149.98 151.64 153.3 154.96 156.62 158.28 159.94 161.6 163.26 164.92 166.58 168.24 169.9 171.56 173.22 174.88 176.54 178.2 179.86 181.52 183.18 184.84 186.5 188.16 189.82 191.48

1.636 0.573 0.620 0.699 0.815 1.213 2.005 1.396 0.500 0.464 0.741 0.667 1.213 2.294 1.383 0.501 0.427 0.572 0.663 1.150 2.235 1.206 0.497 0.552 0.477 0.558 1.011 1.969

2

3

4

Week 1 2 3 4 Average Seasonal Index

Sunday 1.636 1.396 1.383 1.206 1.405

Monday .573 .500 .501 .497 .518

Tuesday .620 .464 .427 .552 .516

Day Wednesday .699 .741 .572 .477 .657

Thursday .815 .667 .663 .558 .676

Friday 1.213 1.213 1.150 1.011 1.187

Saturday 2.005 2.294 2.235 1.969 2.126

7.085

1.404

.517

.515

.621

.675

1.145

2.123

7.000

20.29 Regression line: ŷ = 90.4 + 2.02 t

807

Total


Year

Quarter

t

y

ŷ =90.4 + 2.02t

y / yˆ

1

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

106 92 65 121 115 100 73 135 114 105 79 140 121 111 82 163

92.42 94.44 96.46 98.48 100.5 102.52 104.54 106.56 108.58 110.6 112.62 114.64 116.66 118.68 120.7 122.72

1.147 0.974 0.674 1.229 1.144 0.975 0.698 1.267 1.050 0.949 0.701 1.221 1.037 0.935 0.679 1.328

1 1.147 1.144 1.050 1.037 1.095

2 .974 .975 .949 .935 .959

3 .674 .698 .701 .679 .688

4 1.229 1.267 1.221 1.328 1.261

4.003

1.094

.958

.688

1.260

4.000

2

3

4

Quarter Year 1 2 3 4 Average Seasonal Index

Total

20.30

MAD 

166  173  179  186  195  192  214  211  220  223

5 7  7  3  3  3 23    4.60 5 5 SSE  (166  173 ) 2  (179  186 ) 2  (195  192 ) 2  ( 214  211 ) 2  ( 220  223 ) 2  49  49  9  9  9  125

20.31

Model 1:

MAD 

6.0  7.5  6.6  6.3  7.3  5.4  9.4  8.2 4

1.5  .3  1.9  1.2  4 4.9   1.225 4

808


Model 2:

MAD 

6.0  6.3  6.6  6.7  7.3  7.1  9.4  7.5

4 .3  .1  .2  1.9 2.5    .625 4 4

Model 2 is more accurate. Model 1: SSE = (6.0 - 7.5)2 + (6.6 - 6.3)2 + (7.3 - 5.4)2 + (9.4 - 8.2)2 = 2.25 + .09 + 3.61 + 1.44 = 7.39 Model 2: SSE = (6.0 - 6.3)2 + (6.6 - 6.7)2 + 7.3 - 7.1)2 + (9.4 - 7.5)2 = .09 + .01 + .04 + 3.61 = 3.75 Model 2 is more accurate.

20.32

57  63  60  72  70  86  75  71  70  60 MAD 

5 6  12  16  4  10 48    9.6 5 5

SSE = (57 - 63)2 + (60 - 72)2 + (70 - 86)2 + (75 - 71)2 +(70-60)2 = 36 + 144 + 256 + 16 + 100 = 552

20.33

Technique 1:

MAD  

19  21  24  27  28  29  32  31  38  35 5

2  3  1  1  3 10   2.0 5 5

SSE = (19 - 22)2 + (24 - 27)2 + (32 - 31)2 + (38 - 35)2 = 4 + 9 + 1 + 1 + 9 = 24 Technique 2:

MAD 

19  22  24  24  28  26  32  28  38  30 5

3  0  2  4  8 15    3.0 5 5 SSE = (19 - 22)2 + (24 - 24)2 + (28 - 26)2 + (32 - 28)2 + (28 - 30)2 = 9 + 0 + 4 + 16 + 64 = 93

809


Technique 3:

MAD 

19  17  24  20  28  25  32  31  38  39 5

2  4  3  1  1 11    2.2 5 5 SSE = (19 - 17)2 + (24 - 20)2 + (28 - 25)2 + (32 - 31)2 + (38 - 39)2 = 4 + 16 + 9 + 1+ 1 = 31 By both measures, technique 1 is the most accurate.

20.34 Quarter

t

1 2 3 4

41 42 43 44

20.35

20.36

yˆ  150  3t 273 276 279 282

SI

Forecast

.7 1.2 1.5 .6

191.1 331.2 418.5 169.2

Day

t

yˆ  120  2.3t

SI

Forecast

Sunday Monday Tuesday Wednesday Thursday Friday Saturday

29 30 31 32 33 34 35

186.7 189.0 191.3 193.6 195.9 198.2 200.5

1.5 .4 .5 .6 .7 1.4 1.9

280.1 75.6 95.7 116.2 137.1 277.5 381.0

yˆ t  625 1.3 yt 1  625 1.3(65)  540.5

20.37 yˆ t  155  21yt 1  155  21(11)  386 20.38 F17  F18  F19  F20  S16  43.08

20.39 Day

t

1 2 3 4 5

21 22 23 24 25

SI yˆ  16 .8  .366 t 24.49 .675 24.85 .892 25.22 .864 25.58 1.212 25.95 1.357

810

Forecast 16.53 22.17 21.79 31.01 35.21


20.40 Quarter

t

1 2 3 4

21 22 23 24

20.41 Quarter 1 2 3 4 1 2 3 4

yˆ  47 .7  1.06t 25.44 24.38 23.32 22.26

yˆ  61 .75  1.18t

t 17 18 19 20 21 22 23 24

81.81 82.99 84.17 85.35 86.53 87.71 88.89 90.07

SI

Forecast

1.207 .959 .972 .863

30.71 23.38 22.67 19.21

SI 0.839 1.058 1.270 0.833 0.839 1.058 1.270 0.833

Forecast 68.64 87.80 106.90 71.10 72.60 92.80 112.89 75.03

20.42a ŷ 2007  1,775 .2  .9054 y 2006 = 1,775.2 + .9054(17,672) = 17,775 b F2007  S2006  17,146. 20.43 a ŷ 2007  4.2245  1.1203 2006 = −4.2245 + 1.1203(−817.3) = −919.8

F2007  S2006  −719.5

20.44 Quarter

t

1 2 3 4

25 26 27 28

20.45 Day 1 2 3 4 5 6 7

t 29 30 31 32 33 34 35

yˆ  143  7.42t 328.50 335.92 343.34 350.76

yˆ  145  1.66t 193.14 194.80 196.46 198.12 199.78 201.44 203.10

811

SI 1.404 0.517 0.515 0.621 0.675 1.145 2.123

SI

Forecast

1.063 .962 .927 1.048

349.20 323.16 318.28 367.60

Forecast 271.17 100.71 101.18 123.03 134.85 230.65 431.18


20.46 Day 1 2 3 4

yˆ  90.4  2.02t

t 17 18 19 20

124.74 126.76 128.78 130.80

SI 1.094 0.958 0.688 1.260

Forecast 136.47 121.44 88.60 164.81

20.47 60

Revenues

50 40 30 20 10 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 Quarter

20.48 There is a small upward trend and seasonality.

20.49 ŷ = 20.21 + .732t

20.50 A 1 Seasonal Indexes 2 3 Season 4 1 5 2 6 3 7 4

B

Index 0.646 1.045 1.405 0.904

20.51 Quarter 1 2 3 4

t 21 22 23 24

yˆ  20 .21  .732 t 35.58 36.31 37.05 37.78 812

SI 0.646 1.045 1.405 0.904

Forecast 23.00 37.95 52.05 34.14


20.52a

b. Period Month

ŷ = 16.34-.1008t Seasonal Index

Forecasts

Actual

61

January

10.19

.7192

7.33

2.2

62

February

10.09

.6729

6.79

3.6

63

March

9.99

.8936

8.93

5.3

64

April

9.89

1.0500

10.38

4.5

65

May

9.79

1.2148

11.89

5.7

66

June

9.69

1.3998

13.56

8.1

67

July

9.59

1.1908

11.42

6.1

68

August

9.49

1.1651

11.05

6.5

69

September

9.38

1.0246

9.62

5.8

70

October

9.28

1.0288

9.55

4.9

71

November

9.18

.8535

7.84

4.8

72

December

9.08

.7870

7.15

4.4

c. MAD = 4.47 SSE = 254.32

813


814


Chapter 21 21.1Chance variation is caused by a number of randomly occurring events that are part of the production process and that in general cannot be eliminated without changing the process. 21.3 Special variation is caused by specific events or factors that are frequently temporary and that can usually be identified and eliminated 21.4a Chance variation represents the variation in student achievement caused by differences in preparation, motivation, and ability. b Special variation represents variation due to specific event or factors that can be corrected. 21.5   P(| z | 2.5)  2(1  .9938 )  .0124

21.6 ARL =

1 = 81 .0124

21.7   P(| z | 2.0)  2(1  .9772 )  .0456

21.8 ARL =

1 = 22 .0456

21.9 ARL =

1 = 385 .0026

Number of units = Production  ARL = 100(385) = 38,500 21.10a From Beta-mean spreadsheet,  = .6603 b Probability = .6603 8 = .0361

21.11 P = 1 −  = 1 − .6603 = .3397; ARL =

1 1  = 2.94 P .3397

21.12 Number of units = Production  ARL = 50(385) = 19,250 21.13a From Beta-mean spreadsheet,  = .8133 b Probability = .8133 8 = .1914 815


21.14 P = 1 −  = 1 − .8133 = .1867; ARL =

1 1  = 5.36 P .1867

21.15 Sampling 3 units per hour means that on average we will produce 38,500 units before erroneously concluding that the process is out of control when it isn’t. Sampling 2 units per half hour reduces this figure by 50%. Sampling 4 units per hour means that when the process goes out of control, the probability of not detecting a shift of 1.5 standard deviations is .6603 and we will produce on average 2.94  100 = 294 units until the chart indicates a problem. Sampling 2 units per half hour increases the probability of not detecting the shift to .8133 and decreases the average number of units produced when the process is out of control to 50  5.36 = 268. 21.16 Number of units = Production  ARL = 2000(385) = 770,000 21.17a From Beta-mean spreadsheet,  = .7388 b Probability = .7388 4 = .2979

21.18 P = 1 −  = 1 − .7388 = .2612; ARL =

1 1  = 3.83 P .2612

21.19 Number of units = Production  ARL = 4000(385) = 1,540,000 21.20a From Beta-mean spreadsheet,  = .3659 b Probability = .3659 4 = .0179

21.21 P = 1 −  = 1 − .3659 = .6341; ARL =

1 1  = 1.58 P .6341

21.22 Sampling 10 units per half hour means that on average we will produce 770,000 units before erroneously concluding that the process is out of control when it isn’t. Sampling 20 units per hour doubles this figure. Sampling 10 units per half hour means that when the process goes out of control, the probability of not detecting a shift of .75 standard deviations is .7388 and we will produce on average 3.83  2000 = 7660 units until the chart indicates a problem. Sampling 20 units per hour decreases the probability of not detecting the shift to .3659 and decreases the average number of units produced when the process is out of control to 4000  1.58 = 6320.

21.23 Centerline = x = 453.6

816


 12 .5   = 434.85 = 453.6 − 3  n  4 

Lower control limit = x 

3S

Upper control limit = x 

3S

 12 .5   = 472.35 = 453.6 + 3  n  4 

21.24 Centerline = x = 181.1

 11 .0   = 170.1 = 181.1 − 3   n  9 

Lower control limit = x 

3S

Upper control limit = x 

3S

 11 .0   = 192.1 = 181.1 + 3   n  9 

Zone boundaries: 170.10, 173.77, 177.44, 181.10, 184.77, 188.43, 192.10

21.25 a Centerline = x = 13.3

 3.8   = 7.6 = 13.3 − 3  n  4

Lower control limit = x 

3S

Upper control limit = x 

3S

 3.8   = 19.0 = 13.3 + 3  n  4

c The process is out of control at points 8, 9, 21, 22, and 25. 21.26a S Chart

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

D

Data 10.0885 4.452 0

817


12

10

8

6

4

2

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

x Chart

1 2 3 4 5 6 7 8

A B C Statistical Process Control

D

Data Upper control limit 19.9668 Centerline 12.7386 Lower control limit 5.5103 Pattern Test #2 Failed at Points: 29 Pattern Test #6 Failed at Points: 29, 30

818


25

20

15

10

5

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

c The process is out of control at samples 29 and 30. d A level shift occurred. 21.27a S Chart

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

2by4 lumber 0.1851 0.0721 0

819


0.2 0.18 0.16 0.14 0.12 0.1

0.08 0.06 0.04 0.02 0 1

1 2 3 4 5 6 7 8 9

3

5

7

9

11

13

15

17

A B C Statistical Process Control

D

19

21

23

E

25

27

29

31

F

2by4 lumber Upper control limit 96.0635 Centerline 95.9231 Lower control limit 95.7827 Pattern Test #1 Failed at Points: 1, 4, 5, 28, 35, 36, 37, 39, 40 Pattern Test #5 Failed at Points: 3, 4, 5, 6, 9, 10, 35, 36, 37, 38, 39, 40 Pattern Test #6 Failed at Points: 5, 6, 7, 8, 9, 35, 36, 37, 38, 39, 40

820

33

G

35

37

39


96.2 96.1 96 95.9 95.8 95.7 95.6 95.5 1

3

5

7

9

11

13

15

17

b The process is out of control at the first sample. c Clean out the sawdust more frequently. 21.28

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

AEU 0.0031 0.0015 0

821

19

21

23

25

27

29

31

33

35

37

39


0.0035 0.003 0.0025 0.002 0.0015 0.001 0.0005 0 1

1 2 3 4 5 6

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A B C Statistical Process Control AEU 0.4408 0.4387 0.4366

Upper control limit Centerline Lower control limit

0.442 0.441 0.44 0.439 0.438

0.437 0.436 0.435 0.434 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is under control. 822


S

21.29

S 5

n

Upper control limit  Centerline .4408  .4387   .0007 3 3

 .0007 ; S  .0016

CPL 

x  LSL .4387  .4335   1.08 3S 3(.0016 )

CPU 

USL  x 4435  .4387   1.00 3S 3(.0016 )

C pk  Min (CPL, CPU) = 1.00

21.30

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

Volume 1.8995 0.9093 0

2 1.8 1.6 1.4 1.2 1

0.8 0.6 0.4 0.2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

Volume 1.3871 0.092 -1.2031 823


2 1.5 1 0.5 0 -0.5 -1 -1.5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The process is under control.

S

21.31

S 5

n

Upper control limit  Centerline 1001 .3871  1000 .092   .4317 3 3

 .4317 ; S  .9653

CPL 

x  LSL 1000 .092  995   1.76 3S 3(.9653 )

CPU 

USL  x 1005  1000 .092   1.69 3S 3(.9653 )

Cpk  Min (CPL, CPU) = 1.69

21.32

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

Headrest 2.3313 0.9078 0

824


2.5

2

1.5

1

0.5

0 1

1 2 3 4 5 6 7 8

2

3

4

5

6

7

8

A B C Statistical Process Control

9

10

11

12

13

14

15

16

17

18

19

20

9

10

11

12

13

14

15

16

17

18

19

20

D

Headrest Upper control limit 241.3248 Centerline 239.5617 Lower control limit 237.7986 Pattern Test #1 Failed at Points: 19, 20 Pattern Test #5 Failed at Points: 20

242 241 240 239 238 237 236 235 1

2

3

4

5

6

7

8

825


a The process is out of control. b The process is out of control at sample 19. c The width became too small. 21.33

1 2 3 4 5 6 7

A B C Statistical Process Control

D

Nuts Upper control limit 4.6143 Centerline 2.0363 Lower control limit 0 Pattern Test #1 Failed at Points: 21

6

5

4

3

2

1

0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is out of control at sample 21. It is not necessary to draw the x chart.

21.34

1 2 3 4 5 6 7

A B C Statistical Process Control

D

Seats Upper control limit 11.1809 Centerline 5.3523 Lower control limit 0 Pattern Test #1 Failed at Points: 23, 24

826


14 12 10 8 6 4 2 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is out of control at sample 23. It is not necessary to draw the x chart.

21.35

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

Headers 0.0045 0.002 0

827


0.005

0.004

0.003

0.002

0.001

0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A B C 1 Statistical Process Control 2 3 Headers 4 Upper control limit 4.9872 5 Centerline 4.9841 6 Lower control limit 4.9809 4.988

4.986

4.984

4.982

4.98

4.978

4.976 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is under control.

828


S

21.36

S 4

n

Upper control limit  Centerline 4.9873  4.9841   .00107 3 3

 .00107 ; S  .00214

CPL 

x  LSL 4.9841  4.978   .95 3S 3(.00214 )

CPU 

USL  x 4.990  4.9841   .92 3S 3(.00214 )

Cpk  Min (CPL, CPU) = .92

21.37

1 2 3 4 5 6

A B C Statistical Process Control Bolts 1.2895 0.5021 0

Upper control limit Centerline Lower control limit

1.4 1.2 1 0.8 0.6 0.4 0.2 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

829


A B C Statistical Process Control

1 2 3 4 5 6 7 8

D

E

Bolts Upper control limit 12.7864 Centerline 11.81 Lower control limit 10.8336 Pattern Test #5 Failed at Points: 25 Pattern Test #6 Failed at Points: 23, 24, 25 13

12.5 12 11.5 11 10.5 10 9.5 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process went out of control at sample 23. 21.38

A B C 1 Statistical Process Control 2 3 Bottles 4 Upper control limit 69.6821 5 Centerline 33.3567 6 Lower control limit 0

830


80 70 60 50 40

30 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

1 2 3 4 5 6 7 8 9

A B C Statistical Process Control

D

Bottles Upper control limit 803.9376 Centerline 756.4267 Lower control limit 708.9157 Pattern Test #1 Failed at Points: 30 Pattern Test #5 Failed at Points: 29, 30 Pattern Test #6 Failed at Points: 30

850 830 810 790 770 750

730 710 690 670 650 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

831


The process went out of control at sample 29. 21.39a

A B C Statistical Process Control

1 2 3 4 5 6

Upper control limit Centerline Lower control limit

Calls 40.592 17.9131 0

50 45 40 35 30 25

20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

1 2 3 4 5 6 7 8 9

A B C Statistical Process Control

D

Calls Upper control limit 114.1052 Centerline 85.0217 Lower control limit 55.9381 Pattern Test #1 Failed at Points: 25 Pattern Test #5 Failed at Points: 30 Pattern Test #6 Failed at Points: 29, 30

832


140 120 100 80 60 40 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

b The process went out of control at sample 25 21.40

A B C 1 Statistical Process Control 2 Pipes 3 0.0956 4 Upper control limit 0.0372 5 Centerline 0 6 Lower control limit

833


0.12

0.1

0.08

0.06

0.04

0.02

0 1

1 2 3 4 5 6

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

8

9

10

11

12

13

14

15

16

17

18

19

20

A B C Statistical Process Control Pipes 3.0615 2.9892 2.9168

Upper control limit Centerline Lower control limit 3.1

3.05

3

2.95

2.9

2.85

2.8 1

2

3

4

5

6

7

The process is under control.

834


S

21.41

S 3

n

Upper control limit  Centerline 3.0615  2.9892   .0241 3 3

 .0241; S  .0417

CPL 

x  LSL 2.9892  2.9   .71 3S 3(.0417 )

CPU 

USL  x 3.1  2.9892   .89 3S 3(.0417 )

Cpk  Min (CPL, CPU) = .71

21.42

S 5

S n

Upper control limit  Centerline 1504 .572  1496 .952   2.54 3 3

 2.54; S  5.68

CPL 

x  LSL 1496 .952  1486   .64 3S 3(5.68 )

CPU 

USL  x 1506  1496 .952   .53 3S 3(5.68 )

Cpk  Min (CPL, CPU) = .53 The value of the index is low because the statistics used to calculate the control limits and centerline were taken when the process was out of control. 21.43 Centerline = p = .035 Lower control limit = p  3

(.035 )(1  .035 ) p (1  p ) = .035 − 3 = .0174 1000 n

Upper control limit = p + 3

(.035 )(1  .035 ) p (1  p ) = .035 + 3 = .0524 1000 n

21.44 Centerline = p = .0324 Lower control limit = p  3

p (1  p ) (.0324 )(1  .0324 ) = .0324 − 3 = −.00516 (= 0) 200 n

Upper control limit = p + 3

p (1  p ) (.0324 )(1  .0324 ) = .0324 + 3 = .06996 200 n

835


1 2 3 4 5 6 7

A B C Statistical Process Control

D

Copiers Upper control limit 0.07 Centerline 0.0324 Lower control limit 0 Pattern Test #1 Failed at Points: 25

0.014 0.012 0.01 0.008 0.006 0.004 0.002 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is out of control at sample 25. 21.45 Centerline = p = .00352 Lower control limit = p  3

p (1  p ) (.00352 )(1  .00352 ) = .00352 − 3 = −.00443 (=0) 500 n

Upper control limit = p + 3

p (1  p ) (.00352 )(1  .00352 ) = .00352 + 3 = .01147 500 n

1 2 3 4 5 6

A B C Statistical Process Control

Upper control limit Centerline Lower control limit

PCBs 0.0115 0.0035 0

836


0.012

0.01

0.008

0.006

0.004

0.002

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The process is under control. 21.46 Centerline = p = . 0383 Lower control limit = p  3

(.0383 )(1  .0383 ) p (1  p ) = .0383 − 3 = −.0193 (= 0) 100 n

Upper control limit = p + 3

(.0383 )(1  .0383 ) p (1  p ) = .0383 + 3 = .0959 100 n

1 2 3 4 5 6 7 8

A B C Statistical Process Control

D

Telephones Upper control limit 0.0959 Centerline 0.0383 Lower control limit 0 Pattern Test #1 Failed at Points: 25, 30

837


0.12

0.1

0.08

0.06

0.04

0.02

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The process is out of control at samples 25 and 30. 21.47 Centerline = p = .0169 Lower control limit = p  3

p (1  p ) (.0169 )(1  .0169 ) = .0169 − 3 = .0047 1000 n

Upper control limit = p + 3

p (1  p ) (.0169 )(1  .0169 ) = .0169 + 3 = .0291 1000 n

1 2 3 4 5 6 7

A B C Statistical Process Control

D

Pages Upper control limit 0.0291 Centerline 0.0169 Lower control limit 0.0047 Pattern Test #2 Failed at Points: 37

838


0.035 0.03 0.025 0.02 0.015 0.01 0.005 0 1

3

5

7

9

11 13 15 17 19 21 23 25 27 29 31 33 35 37 39

The process is out of control at samples 37-40. 21.48

A B C D 1 Statistical Process Control 2 Batteries 3 0.047 4 Upper control limit 0.0257 5 Centerline 0.0045 6 Lower control limit 7 Pattern Test #1 Failed at Points: 28, 29, 30

839

E


0.06

0.05

0.04

0.03

0.02

0.01

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The process is out of control at sample 28. 21.49

A B C D 1 Statistical Process Control 2 3 Courier 4 Upper control limit 0.0088 5 Centerline 0.0044 6 Lower control limit 0 7 Pattern Test #1 Failed at Points: 23, 26 8 Pattern Test #2 Failed at Points: 9, 29

840


0.01

0.008

0.006

0.004

0.002

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

The process went out of control at sample 9. 21.50

A B C 1 Statistical Process Control 2 Scanners 3 0.0275 4 Upper control limit 0.0126 5 Centerline 0 6 Lower control limit 7 Pattern Test #1 Failed at Points: 24

D

841


0.035 0.03 0.025 0.02 0.015 0.01 0.005 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

The process is out of control at sample 24.

842


Chapter 22 22.1

a1

a2

s1

0

29

s2

0

5

s3

14

0

s4

36

0

22.2

22.3 EMV(a 1 ) = .4(55) + .1(43) + .3(29) + .2(15) = 38.0 EMV(a 2 ) = .4(26) + .1(38) + .3(43) + .2(51) = 37.3

843


22.4

22.5

a1

a2

a3

s1

0

15

21

s2

0

3

4

s3

20

5

0

22.6 EOL(a 1 ) = .2(0) + .6(0) + .2(20) = 4.0 EOL(a 2 ) = .2(15) + .6(3) + .2(5) = 5.8 EOL(a 3 ) = .2(21) + .6(4) + .2(0) = 6.6 The EOL decision is a 1 .

22.7a

Produce

Demand

a0

a1

a2

a3

s0

0

-3.00

-6.00

-9.00

s1

0

5.00

2.00

-1.00

s2

0

5.00

10.00

7.00

s3

0

5.00

10.00

15.00

844


b

Produce

Demand

a0

a1

a2

a3

s0

0

3.00

6.00

9.00

s1

5.00

0

3.00

6.00

s2

10.00

5.00

0

3.00

s3

15.00

10.00

5.00

0

c

22.8a EMV(a 0 ) = 0 EMV(a 1 ) = .25(-3.00) + .25(5.00) + .25(5.00) + .25(5.00) = 3.00 EMV(a 2 ) = . 25(-6.00) + .25(2.00) + .25(10.00) + .25(10.00) = 4.00 EMV(a 3 ) = . 25(-9.00) + .25(-1.00) + .25(7.00) + .25(15.00) = 3.00

845


EMV decision is a 2 (bake 2 cakes) b EOL(a 0 ) = .25(0) + .25(5.00) + .25(10.00) + .25(15.00) = 7.50 EOL(a 1 ) = .25(3.00) + .25(0) + .25(5.00) + .25(10.00) = 4.50 EOL(a 2 ) = . 25(6.00) + .25(3.00) + .25(0) + .25(5.00) = 3.50 EOL(a 2 ) = . 25(9.00) + .25(6.00) + .25(3.00) + .25(0) = 4.50 EOL decision is a 2 (bake 2 cakes)

22.9

a 1 (flat fee)

a 2 Pay per snowfall

s0

-40,000

0

s1

-40,000

-18,000

s2

-40,000

-36,000

s3

-40,000

-54,000

s4

-40,000

-72,000

22.10 EMV(a 1 ) = -40,000 EMV(a 2 ) = . 05(0) + .15(-18,000) + .30(-36,000) + .40(-54,000) +.10(-72,000) = -42,300 EMV decision is a 1

22.11a Payoff Table

s 100

s 150

s 200

s 250

a 100

a 200

a 300

12(100)-10(100)

12(100)-9(200)+6(100)

12(100)-8.50(300)+6(200)

= 200

=0

= -150

12(100)- 10(100)

12(150)-9(200)+6(50)

12(150)-8.50(300)+6(150)

= 200

= 300

= 150

12(100)-10(100)

12(200)-9(200)

12(200)-8.50(300)+6(100)

= 200

= 600

= 450

12(100)-10(100)

12(200)–9(200)

12(250)-8.50(300)+6(50)

= 200

= 600

= 750

846


b

Opportunity Loss Table a 100

a 200

a 300

s 100

0

200

350

s 150

100

0

150

s 200

400

0

150

s 250

550

150

0

c

22.12 EMV(a 100 ) = 200 EMV(a 200 ) = . 20(0) + .25(300) + .40(600) + .15(600) = 405 EMV(a 300 ) = . 20(-150) + .25(150) + .40(450) + .15(750) = 300 EMV decision is order 200 shirts.

847


22.13 P(s 0 ) = .607, P(s 1 ) = .303, P(s 2 ) = .076, P(s 3 ) = .012, P(s 4 ) = .002 Payoff Table a0

a1

a2

a3

s0

0

-6,000

-12,000

-18,000

s1

0

7,000

1,000

-5,000

s2

0

7,000

14,000

8,000

s3

0

7,000

14,000

21,000

Opportunity Loss Table a0

a1

a2

a3

s0

0

6,000

12,000

18,000

s1

7,000

0

6,000

12,000

s2

14,000

7,000

0

6,000

s3

21,000

14,000

7,000

0

22.14a EMV(Small) = .15(-220) + .55(-330) + .30(-440) = -346.5 EMV(Medium) = .15(-300) + .55(-320) + .30(-390) = -338.0 EMV(Large) = .15(-350) + .55(-350) + .30(-350) =-350.0 EMV decision: build a medium size plant; EMV*= -338.0 b

Opportunity Loss Table Small

Medium

Large

Low

0

80

130

Moderate

10

0

30

High

90

40

0

c EOL(Small) = .15(0) + .55(10) + .30(90) = 32.5 EOL(Medium) = .15(80) + .55(0) + .30(40) = 24.0 EOL(Large) = .15(130) + .55(30) + .30(0) = 36.0 EOL decision: build a medium size plant

848


22.15a P(s 10 ) = 9/90 = .10, P(s 11 ) = 18/90 = .20, P(s 12 ) = 36/90 = .40, P(s 13 ) = 27/90 = .30 Payoff Table

s 10

s 11

s 12

s 13

a 10

a 11

a 12

a 13

30

10(5)- 11(2)+2

10(5)-12(2)+3.50

10(5)-13(2)+4.50

= 30

= 29.50

= 28.50

11(5)-11(2)

11(5)-12(2)+2

11(5)-13(2)+3.50

= 33

= 33

= 32.50

11(5)-11(2)

12(5)-12(2)

12(5)- 13(2)+2

= 33

= 36

= 36

11(5)-11(2)

12(5)-12(2)

13(5)-13(2)

= 33

= 36

= 39

30

30

30

b EMV(a 10 ) = 30 EMV(a 11 ) = .10(30) + .20(33) + .40(33) + .30(33) = 32.70 EMV(a 12 ) = .10(29.50) + .20(33) + .40(36) + .30(36) = 34.75 EMV(a 13 ) = .10(28.50) + .20(32.50) + .40(36) + .30(39) = 35.45 EMV decision: buy 13 bushels 22.16

Payoff Table Decision Produce

Don’t produce

5%

-28 million

0

10%

2 million

0

15%

8 million

0

Market share

EMV(produce) = .15(-28 million) + .45(2 million) + .40 (8 million) = -.1 million EMV (don’t produce) = 0 EMV decision: don’t produce 22.17 EPPI = .10(110) + .25(150) + .50(220) + .15(250) = 196 EMV(a 1 ) = .10(60) + .25(40) + .50(220) + .15(250) = 163.5 EMV(a 2 ) = .10(110) + .25(110) + .50(120) + .15(120) = 116.5 EMV(a 3 ) = .10(75) + .25(150) + .50(85) + .15(130) = 107 EVPI = EPPI – EMV* = 196 – 163.5 = 32.5 849


22.18

Opportunity Loss Table a1

a2

a3

s1

50

0

35

s2

110

40

0

s3

0

100

135

s4

0

130

120

EOL(a 1 ) = .10(50) + .25(110) + .50(0) + .15(0) = 32.5 EOL(a 2 ) = .10(0) + .25(40) + .50(100) + .15(130) = 79.5 EOL(a 3 ) = .10(35) + .25(0) + .50(135) + .15(120) = 89 EOL* = 32.5 22.19 EPPI = .5(65) + .5(110) = 87.5 EMV(a 1 ) = .5(65) + .5(70) = 67.5 EMV(a 2 ) = .5(20) + .5(110) = 65.0 EMV(a 3 ) = .5(45) + .5(80) = 62.5 EMV(a 4 ) = .5(30) + .5(95) = 62.5 EVPI = EPPI – EMV* = 87.5 – 67.5 = 20 22.20 a EPPI = .75(65) + .25(110) = 76.25 EMV(a 1 ) = .75(65) + .25(70) = 66.25 EMV(a 2 ) = .75(20) + .25(110) = 42.5 EMV(a 3 ) = .75(45) + .25(80) = 53.75 EMV(a 4 ) = .75(30) + .25(95) = 46.25 EVPI = EPPI – EMV* = 76.25 – 66.25 = 10 b EPPI = .95(65) + .05(110) = 67.25 EMV(a 1 ) = .95(65) + .05(70) = 65.25 EMV(a 2 ) = .95(20) + .05(110) = 24.5 EMV(a 3 ) = .95(45) + .05(80) = 46.75 EMV(a 4 ) = .95(30) + .05(95) = 33.25 EVPI = EPPI – EMV* = 67.25 – 65.25 = 2 22.21 As the difference between the two prior probabilities increases EVPI decreases.

850


22.22 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .25 .40 (.25)(.40) = .10 .12/.20 = .500 s2

.40

.25

(.40)(.25) = .10

.10/.20 = .500

s3

.35

0

(.35)(0) = .0

0/.20 = 0

P(I 1 ) = .20 Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .25 .30 (.25)(.30) = .075 .075/.28 = .268 s2

.40

.25

(.40)(.25) = .10

.10/.28 = .357

s3

.35

30

(.35)(.30) = .105

.105/.28 = .375

P(I 2 ) = .28 Posterior Probabilities for I 3 sj

P(s j )

P(I 3 |s j )

P(s j and I 3 )

P(s j | I 3 )

__________________________________________________________________________ s1 .25 .20 (.25)(.20) = .05 .05/.29 = .172 s2

.40

.25

(.40)(.25) = .10

.10/.29 = .345

s3

.35

.40

(.35)(.40) = .14

.14/.29 = .483

P(I 3 ) = .29 Posterior Probabilities for I 4 sj

P(s j )

P(I 4 |s j )

P(s j and I 4 )

P(s j | I 4 )

__________________________________________________________________________ s1 .25 .10 (.25)(.10) = .025 .025/.23 = .109 s2

.40

.25

(.40)(.25) = .10

.10/.23 = .435

s3

.35

.30

(.35)(.30) = .105

.105/.23 = .456

P(I 4 ) = .23 22.23 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .5 .98 (.5)(.98) = .49 .49/.515 = .951 s2

.5

.05

(.5)(.05) = .025 P(I 1 ) = .515

851

.025/.515 = .049


Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .5 .02 (.5)(.02) = .01 .01/.485 = .021 s2

.5

.95

(.5)(.95) = .475

.475/.485 = .979

P(I 2 ) = .485 22.24a Prior probabilities: EMV(a 1 ) = .5(10) + .5(22) = 16 EMV(a 2 ) = .5(18) + .5(19) = 18.5 EMV(a 3 ) = .5(23) + .5(15) = 19 EMV* = 19

I 1 : EMV(a 1 ) = .951(10) + .049(22) = 10.588 EMV(a 2 ) = .951(18) + .049(19) = 18.049 EMV(a 3 ) = .951(23) + .049(15) = 22.608 Optimal act: a 3

I 2 : EMV(a 1 ) = .021(10) + .979(22) = 21.748 EMV(a 2 ) = .021(18) + .979(19) = 18.979 EMV(a 3 ) = .021(23) + .979(15) = 15.168 Optimal act: a 1 b EMV` = .515(22.608) + .485(21.748) = 22.191 EVSI = EMV` - EMV* = 22.191 – 19 = 3.191 22.25 Prior probabilities: EMV(a 1 ) = .333(60) + .333(90) + .333(150) = 100 EMV(a 2 ) = 90 EMV* = 100 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .333 .7 (.333)(.7) = .233 .233/.467 = .499 s2

.333

.5

(.333)(.5) = .167

.167/.467 = .358

s3

.333

.2

(.333)(.2) = .067

.067/.467 = .143

P(I 1 ) = .467

852


Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .333 .3 (.333)(.3) = .100 .100/.534 = .187 s2

.333

.5

(.333)(.5) = .167

.167/.534 = .313

s3

.333

.8

(.333)(.8) = .267

.267/.534 = .500

P(I 2 ) = .534 I 1 : EMV(a 1 ) = .499(60) + .358(90) + .143(150) = 83.61 EMV(a 2 ) = 90 I 2 : EMV(a 1 ) = .187(60) + .313(90) + .500(150) = 114.39 EMV(a 2 ) = 90 EMV` = .467(90) + .534(114.39) = 103.11 EVSI = EMV` - EMV* = 103.11 – 100 = 3.11 22.26 Prior probabilities: EMV(a 1 ) = .5(60) + .4(90) + .1(150) = 81 EMV(a 2 ) = 90 EMV* = 90

Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .5 .7 (.5)(.7) = .35 .35/.57 = .614 s2

.4

.5

(.4)(.5) = .20

.20/.57 = .351

s3

.1

.2

(.1)(.2) = .02

.02/.57 = .035

P(I 1 ) = .57 Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ .5 .3 (.5)(.3) = .15 .15/.43 = .349 s1 s2

.4

.5

(.4)(.5) = .20

.20/.43 = .465

s3

.1

.8

(.1)(.8) = .08

.08/.43 = .186

P(I 1 ) = .43 I 1 : EMV(a 1 ) = .614(60) + .351(90) + .035(150) = 73.68 EMV(a 2 ) = 90

I2: EMV(a 1 ) = .349(60) + .465(90) + .186(150) = 90.69 EMV(a 2 ) = 90 853


EMV` = .57(90) + .43(90.69) = 90.30 EVSI = EMV` - EMV* = 90.30 – 90 = .30 22.27 Prior probabilities: EMV(a 1 ) = .90(60) + .05(90) + .05(150) = 66 EMV(a 2 ) = 90 EMV* = 90 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ .90 .7 (.90)(.7) = .63 .63/.665 = .947 s1 s2

.05

.5

(.05)(.5) = .025

.025/.665 = .038

s3

.05

.2

(.05)(.2) = .01

.01/.665 = .015

P(I 1 ) = .665 Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ .90 .3 (.90)(.3) = .27 .27/.335 = .806 s1 s2

.05

.5

(.05)(.5) = .025

.025/.335 = .075

s3

.05

.8

(.05)(.8) = .04

.04/.335 = .119

P(I 1 ) = .335 I 1 : EMV(a 1 ) = .947(60) + .038(90) + .015(150) = 62.49 EMV(a 2 ) = 90 I 2 : EMV(a 1 ) = .806(60) + .075(90) + .119(150) = 72.96 EMV(a 2 ) = 90 EMV` = .665(90) + .335(90) = 90 EVSI = EMV` - EMV* = 90 – 90 = 0 22.28 As the prior probabilities become more diverse EVSI decreases. 22.29 Payoff Table Demand

Purchase lot

Don’t purchase lot

10,000

10,000(5)-125,000 = -75,000

0

30,000

30,000(5) – 125,000 = 25,000

0

50,000

50,000(5)-125,000 = 125,000

0

EMV(purchase) = .2(-75,000) + .5(25,000) + .3(125,000) = 35,000 854


EMV(don’t purchase) = 0 EPPI = .2(0) + .5(25,000) + .3(125,000) = 50,000 EVPI = EPPI – EMV* = 50,000 – 35,000 = 15,000 22.30 EMV* = 0 EPPI = .15(0) + .45(2 million) + .40(8 million) = 4.1 million EVPI = EPPI – EMV* = 4.1 million – 0 = 4.1 million 22.31 Likelihood probabilities (binomial probabilities) P(I | s 1 ) = P(x = 3, n= 25 | p = .05) = .0930 P(I | s 2 ) = P(x = 3, n= 25 | p = .10) = .2265 P(I | s 3 ) = P(x = 3, n= 25 | p = .15) = .2174 Posterior Probabilities P(s j ) sj

P(I | s j )

P(s j and I)

P(s j | I)

__________________________________________________________________________ s1 .15 .0930 (.15)(.0930) = .0140 .0140/.2029 = .0690 s2 .45 .2265 (.45)(.2265) = .1019 .1019/.2029 = .5022 s3 .40 .2174 (.40)(.2174) = .0870 .0870/.2029 = .4288 P(I) = .2029 EMV(produce) = .0690(-28 million) + .5022(2 million) + .4288(8 million) = 2.503 million EMV (don’t produce) = 0 EMV decision: produce 22.32a

Payoff Table

Market share

Switch

Don’t switch

5%

5(100,000) – 700,000 = -200,000

285,000

10%

10(100,000) – 700,000 = 300,000

285,000

20%

20(100,000)-700,000 = 1,300,000

285,000

b EMV(switch) = .4(-200,000) + .4(300,000) + .2(1,300,000) = 300,000 EMV(don’t switch) = 285,000 Optimal act: switch (EMV* = 300,000) c EPPI = .4(285,000) + .4(300,000) + .2(1,300,000) = 494,000 EVPI = EPPI – EMV* = 494,000 – 300,000= 194,000

855


22.33

Payoff Table

Participating Households

Proceed

Don’t proceed

50,000

50(500) – 55,000 = -30,000

0

100,000

100(500) – 55,000 = -5,000

0

200,000

200(500) – 55,000 = 45,000

0

300,000

300(500) – 55,000 = 95,000

0

Likelihood probabilities (binomial probabilities) P(I | s 1 ) = P(x = 3, n= 25 | p = .05) = .0930 P(I | s 2 ) = P(x = 3, n= 25 | p = .10) = .2265 P(I | s 3 ) = P(x = 3, n= 25 | p = .20) = .1358 P(I | s 4 ) = P(x = 3, n= 25 | p = .30) = .0243

Posterior Probabilities sj P(s j )

P(I | s j )

P(s j and I)

P(s j | I)

__________________________________________________________________________ s1 .5 .0930 (.5)(.0930) = .0465 .0465/.1305 = .3563 s2

.3

.2265

(.3)(.2265) = .0680

.0680/.1305 = .5211

s3

.1

.1358

(.1)(.1358) = .0136

.0136/.1305 = .1042

s4

.1

.0243

(.1)(.0243) = .0024 .0024/.1305 = .0184 P(I) = .1305 EMV(proceed) = .3563(-30,000) + .5211(-5,000) + .1042(45,000) + .0184(95,000) = -6,858 EMV (don’t proceed = 0 EMV decision: don’t proceed 22.34 Likelihood probabilities (binomial probabilities) P(I | s 1 ) = P(x = 12, n= 100 | p = .05) = .0028 P(I | s 2 ) = P(x = 12, n= 100 | p = .10) = .0988 P(I | s 3 ) = P(x = 12, n= 100 | p = .20) = .0128 P(I | s 4 ) = P(x = 12, n= 100 | p = .30) = .000013

Posterior Probabilities sj P(s j )

P(I | s j )

P(s j and I)

P(s j | I)

__________________________________________________________________________ s1 .5 .0028 (.5)(.0028) = .0014 .0014/.0323 = .0433 s2

.3

.0988

(.3)(.0988) = .0296

.0296/.0323 = .9164

s3

.1

.0128

(.1)(.0128) = .0013

.0013/.0323 = .0402

s4

.1

.000013

(.1)(.000013) = .000001 .000001/.0323 = .000031 P(I) = .0323 856


EMV(proceed) = .0433(-30,000) + .9164(-5,000) + .0402(45,000) + .000031(95,000) = -4,069 EMV (don’t proceed = 0 EMV decision: don’t proceed 22.35 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ .15 .5 (.15)(.5) = .075 .075/.30 = .25 s1 s2

.55

.3

(.55)(.3) = .165

.165/.30 = .55

s3

.30

.2

(.30)(.2) = .06

.06/.30 = .20

P(I 1 ) = .30 Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .15 .3 (.15)(.3) = .045 .045/.435 = .103 s2

.55

.6

(.55)(.6) = .33

.33/.435 = .759

s3

.30

.2

(.30)(.2) = .06

.06/.435 = .138

P(I 2 ) = .435 Posterior Probabilities for I 3 sj

P(s j )

P(I 3 |s j )

P(s j and I 3 )

P(s j | I 3 )

__________________________________________________________________________ .15 .2 (.15)(.2) = .03 .03/.265 = .113 s1 s2

.55

.1

(.55)(.1) = .055

.055/.265 = .208

s3

.30

.6

(.30)(.6) = .18

.18/.265 = .679

P(I 3 ) = .265 I 1 : EMV(a 1 ) = .25(-220) + .55(-330) + .20(-440) = -324.5 EMV(a 2 ) = .25(-300) + .55(-320) + .20(-390) = -329.0 EMV(a 3 ) = .251(-350) + .55(-350) + .20(-350) = -350 Optimal act: a 1 I 2 : EMV(a 1 ) = .103(-220) + .759(-330) + .138(-440) = -333.85 EMV(a 2 ) = .103(-300) + .759(-320) + .138(-390) = -327.59 EMV(a 3 ) = .103(-350) + .759(-350) + .138(-350) = -350 Optimal act: a 2

857


I 3 : EMV(a 1 ) = .113(-220) + .208(-330) + .679(-440) = -392.26 EMV(a 2 ) = .113(-300) + .208(-320) + .679(-390) = -365.28 EMV(a 3 ) = .113(-350) + .208(-350) + .679(-350) = -350 Optimal act: a 3 EMV` = .30(-324.5) + .435(-327.59) + .265(-350) = -332.60 EVSI = EMV` - EMV* = -332.60 – (-338) = 5.40 22.36 I 0 = neither person supports format change I 1 = one person supports format change I 2 = both people support format change Likelihood probabilities P( I i | s j ) I1 .0950 .18 .32

I0 .9025 .81 .64

5% 10% 20%

I2 .0025 .01 .04

Posterior Probabilities for I 0 sj

P(s j )

P(I 0 |s j )

P(s j and I 0 )

P(s j | I 0 )

__________________________________________________________________________ .4 .9025 (.4)(.9025) = .361 .361/.813 = .444 s1 s2

.4

.81

(.4)(.81) = .324

.324/.813 = .399

s3

.2

.64

(.2)(.64) = .128

.128/.813 = .157

P(I 0 ) = .813 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .4 .0950 (.4)(.0950) = .038 .038/.174 = .218 s2

.4

.18

(.4)(.18) = .072

.072/.174 = .414

s3

.2

.32

(.2)(.32) = .064

.064/.174 = .368

P(I 1 ) = .174 Posterior Probabilities for I 3 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .4 .0025 (.4)(.0025) = .001 .001/.013 = .077 s2

.4

.01

(.4)(.01) = .004

.004/.013 = .308

s3

.2

.04

(.2)(.04) = .008

.008/.013 = .615

P(I 2 ) = .013

858


I 1 : EMV(switch) = .444(-200,000) + .399(300,000) + .157(1,300,000) = 235,000 EMV(don’t switch) = 285,000 Optimal act: don’t switch I 2 : EMV(switch) = .218(-200,000) + .414(300,000) + .368(1,300,000) = 559,000 EMV(don’t switch) = 285,000 Optimal act: switch I 3 : EMV(switch) = .077(-200,000) + .308(300,000) + .615(1,300,000) = 876,500 EMV(don’t switch) = 285,000 Optimal act: switch EMV` = .813(285,000) + .174(546,000) + .013(876,500) = 338,104 EVSI = EMV` - EMV* = 338,104 – 300,000 = 38,104 22.37 Likelihood probabilities (binomial probabilities) P(I | s 1 ) = P(x = 2, n= 25 | p = .05) = .2305 P(I | s 2 ) = P(x = 2, n= 25 | p = .10) = .2659 P(I | s 3 ) = P(x = 2, n= 25 | p = .20) = . 0708

Posterior Probabilities for I P(s j ) P(I|s j ) sj

P(s j and I)

P(s j | I)

__________________________________________________________________________ .4 .2305 (.4)(.2305) = .0922 .0922/.2127 = .4334 s1 s2

.4

.2659

(.4)(.2659) = .1064

.1064/.2127 = .5000

s3

.2

.0708

(.2)(.0708) = .0142

.0142/.2127 = .0667

P(I) = .2127 EMV(switch) = .4334(-200,000) + .5000(300,000) + .0667(1,300,000) = 149,873 EMV(don’t switch) = 285,000 Optimal act: don’t switch

859


22.38a Payoff Table Demand

Battery 1

Battery 2

Battery 3

50,000

20(50,000)-900,000

23(50,000)-1,150,000

25(50,000)-1,400,000

= 100,000

0

-150,000

20(100,000)-900,000

23(100,000)-1,150,000

25(100,000)-1,400,000

=1,100,000

1,150,000

1,100,000

20(150,000)-900,000

23(150,000)-1,150,000

25(150,000)-1,400,000

=2,100,000

2,300,000

2,350,000

100,000 150,000

b

Opportunity Loss table

Demand

Battery 1

Battery 2

Batter3

50,000

0

100,000

250,000

100,000

50,000

0

50,000

150,000

250,000

50,000

9

c EMV(Battery 1) = .3(100,000) + .3(1,100,000) + .4(2,100,000) = 1,200,000 EMV(Battery 2) = .3(0) + .3(1,150,000) + .4(2,300,000) = 1,265,000 EMV(Battery 3) = .3(-150,000) + .3(1,100,000) + .4(2,350,000) = 1,225,000 EMV decision: Battery 2 d EOL(Battery 2) = .3(100,000) + .3(0) + .4(50,000) = 50,000 EVPI = EOL* = 50,000 22.39

Payoff Table

Percentage change

Change ad

-2

-258,000

0

-1

-158,000

0

0

-58,000

0

1

42,000

0

2

142,000

0

Don’t change

EMV(Change ad) = -1(-258,000) + .1(-158,000) + .2(-58,000) + .3(42,000) + .3(142,000) = 2,000 EMV (don’t change) = 0. Optimal decision: change ad

860


22.40 I 0 = person does not believe the ad I 1 = person believes the ad Likelihood probabilities P( I i | s j ) I0 .70 .69 .68 .67 .66

30% 31% 32% 33% 34%

I1 .30 .31 .32 .33 .34

Posterior Probabilities for I 0 sj

P(s j )

P(I 0 |s j )

P(s j and I 0 )

P(s j | I 0 )

__________________________________________________________________________ s1 .1 .70 (.1)(.70) = .070 .070/.674 = .104 s2

.1

.69

(.1)(.69) = .069

.069/.674 = .102

s3

.2

.68

(.2)(.68) = .136

.136/.674 = .202

s4

.3

.67

(.3)(.67) = .201

.201/.674 = .298

s5

.3

.66

(.3)(.66) = .198

.198/.674 = .294

P(I 0 ) = .674 Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ s1 .1 .30 (.1)(.30) = .030 .030/.326 = .092 s2

.1

.31

(.1)(.31) = .031

.031/.326 = .095

s3

.2

.32

(.2)(.32) = .064

.064/.326 = .196

s4

.3

.33

(.3)(.33) = .099

.099/.326 = .304

s5

.3

.34

(.3)(.34) = .102

.102/.326 = .313

P(I 1 ) = .326 I 0 : EMV(Change ad) = .104(-258,000) + .102(-158,000) + .202(-58,000) + .298(42,000) + .294(142,000) = -400 EMV (don’t change) = 0. Optimal decision: don’t change ad

I 1 : EMV(Change ad) = .092(-258,000) + .095(-158,000) + .196(-58,000) + .304(42,000) + .313(142,000) = 7,100

861


EMV (don’t change) = 0. Optimal decision: change ad EMV` = .674(0) + .326(7,100) = 2,315 EVSI = EMV` - EMV* = 2,315 – 2,000 = 315 22.41 Likelihood probabilities (binomial probabilities) P(I | s 1 ) = P(x = 1, n = 5 | p = .30) = .3602 P(I | s 2 ) = P(x = 1, n = 5 | p = .31) = .3513 P(I | s 3 ) = P(x = 1, n = 2 | p = .32) = . 3421 P(I | s 4 ) = P(x = 1, n = 2 | p = .33) = . 3325 P(I | s 5 ) = P(x = 1, n = 2 | p = .34) = . 3226 Posterior Probabilities for I sj P(s j ) P(I|s j )

P(s j and I)

P(s j | I)

__________________________________________________________________________ .1 .3602 (.1)(.3602) = .0360 .0360/.3361 = .1072 s1 s2

.1

.3513

(.1)(.3513) = .0351

.0351/.3361 = .1045

s3

.2

.3421

(.2)(.3421) = .0684

.0684/.3361 = .2036

s4

.3

.3325

(.3)(.3325) = .0997

.0997/.3361 = .2968

s5

.3

.3226

(.3)(.3226) = .0968 P(I) = .3361

.0968/.3361 = .2879

EMV(Change ad) = .1072(-258,000) + .1045(-158,000) + .2036(-58,000) + .2968(42,000) + .2879(142,000) = -2,620 EMV (don’t change) = 0. Optimal decision: don’t change ad 22.42 EMV(25 telephones) = 50,000 EMV(50 telephones) = .50(30,000) + .25(60,000) + .25(60,000) = 45,000 EMV(100 telephones) = .50(20,000) + .25(40,000) + .25(80,000) = 40,000 Optimal decision: 25 telephones (EMV* = 50,000) I 1 = small number of calls I 2 = medium number of calls I 3 = large number of calls

862


Likelihood probabilities (Poisson distribution)

=5

I1

I2

I3

P(X < 8 |  = 5)

P(8  X < 17 |  = 5)

P(X  17 |  = 5)

= .8667

= .1334

=0

P(8  X < 17 |  = 10)

P(X  17 |  = 10)

= .7527

= .0270

P(8  X < 17 |  = 15)

P(X  17 |  = 15)

= .6461

= .3359

 = 10 P(X < 8 |  = 10) = .2202

 = 15 P(X < 8 |  = 15) = .0180

Posterior Probabilities for I 1 sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ .50 .8667 (.50)(.8667) = .4333 .4333/.4929 = .8792 s1 s2

.25

.2202

(.25)(.2202) = .0551

.0551/.4929 = .1117

s3

.25

.0180

(.25)(.0180) = .0045

.0045/.4929 = .0091

P(I 1 ) = .4929 Posterior Probabilities for I 2 sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ .50 .1334 (.50)(.1334) = .0667 .0667/.4164 = .1601 s1 s2

.25

.7527

(.25)(.7527) = .1882

.1882/.4164 = .4519

s3

.25

.6461

(.25)(.6461) = .1615

.1615/.4164 = .3879

P(I 2 ) = .4164 Posterior Probabilities for I 3 sj

P(s j )

P(I 3 |s j )

P(s j and I 3 )

P(s j | I 3 )

__________________________________________________________________________ s1 .50 .0 (.50)(0) = 0 0/.0907 = 0 s2

.25

.0270

(.25)(.0270) = .0068

.0068/.0907 = .0745

s3

.25

.3359

(.25)(.3359) = .0840

.0840/.0907 = .9254

P(I 3 ) = .0907 I 1 : EMV(25 telephones) = 50,000 EMV(50 telephones) = .8792(30,000) + .1117(60,000) + .0091(60,000) = 33,624 EMV(100 telephones) = .8792(20,000) + .1117(40,000) + .0091(80,000) = 22,780 Optimal act: 25 telephones

863


I 2 : EMV(25 telephones) = 50,000 EMV(50 telephones) = .1601(30,000) + .4519(60,000) + .3879(60,000) = 55,191 EMV(100 telephones) = .1601(20,000) + .4519(40,000) + .38791(80,000) = 52,310 Optimal act: 50 telephones

I 3 : EMV(25 telephones) = 50,000 EMV(50 telephones) = 0(30,000) + .0745(60,000) + .9254(60,000) = 60,000 EMV(100 telephones) = 0(20,000) + .0745(40,000) + .9254(80,000) = 77,012 Optimal act: 100 telephones EMV` = .4929(50,000) + .4164(55,191) + .0907(77,012) = 54,612 EVSI = EMV` - EMV* = 54,612 – 50,000 = 4,612 Because the value is greater than the cost ($4,000) Max should not sample. If he sees a small number of calls install 25 telephones. If there is a medium number install 50 telephones. If there is a large number of calls, install 100 telephones. 22.43a EMV(Model 101) = .2(20 million) + .4(100 million) + .4(210 million) = 128 million EMV (Model 202) = .1(70 million) + .4(100 million) + .5(150 million) = 122 million Optimal decision: Model 101 b Likelihood probabilities (binomial distribution) for Model 101 P(X =1, n = 10| p = .05) = .3151 P(X =1, n = 10| p = .10) = .3874 P(X =1, n = 10| p = .15) = .3474 Posterior Probabilities for Model 101 P(s j ) P(I|s j ) sj

P(s j and I)

P(s j | I)

__________________________________________________________________________ s1 .2 .3151 (.2)(.3151) = .0630 .0630/.3570 = .1766 s2

.4

.3874

(.4)(.3874) = .1550

.1550/.3570 = .4341

s3

.4

.3474

(.4)(.3474) = .1390

.1390/.3570 = .3893

P(I) = .3570 EMV(Model 101) = .1766(20 million) + .4341(100 million) + .3893(210 million) = 128.7 million Likelihood probabilities (binomial distribution) for Model 202 P(X =9, n = 20| p = .30) = .0654 P(X =9, n = 20| p = .40) = .1597 P(X =9, n = 20| p = .50) = .1602 864


Posterior Probabilities for Model 202 sj P(s j ) P(I|s j )

P(s j and I)

P(s j | I)

__________________________________________________________________________ s1 .1 .0654 (.1)(.0654) = .0065 .0065/.1505 = .0434 s2

.4

.1597

(.4)(.1597) = .0639

.0639/.1505 = .4245

s3

.5

.1602

(.5)(.1602) = .0801

.0801/.1505 = .5321

P(I) = .1505 EMV(Model 101) = .0434(70 million) + .4245(100 million) + .5321(150 million) = 125.3 million Optimal decision: Model 101 22.44 EMV( Release in North America) = .5(33 million) + .3(12 million) + .2(-15 million) = 17.1 million EMV(European distributor) = 12 million Optimal decision: Release in North America

Posterior Probabilities for I 1 (Rave review) sj

P(s j )

P(I 1 |s j )

P(s j and I 1 )

P(s j | I 1 )

__________________________________________________________________________ .5 .8 (.5)(.8) = .40 .40/.63 = .635 s1 s2

.3

.5

(.3)(.5) = .15

.15/.63 = .238

s3

.2

.4

(.2)(.4) = .08

.08/.63 = .127

P(I 1 ) = .63 EMV( Release in North America) = .635(33 million) + .238(12 million) + .127(-15 million) =21.9 million EMV(European distributor) = 12 million Optimal decision: Release in North America

Posterior Probabilities for I 2 (lukewarm response) sj

P(s j )

P(I 2 |s j )

P(s j and I 2 )

P(s j | I 2 )

__________________________________________________________________________ s1 .5 .1 (.5)(.1) = .05 .05/.20 = .25 s2

.3

.3

(.3)(.3) = .09

.09/.20 = .45

s3

.2

.3

(.2)(.3) = .06

.06/.20 = .30

P(I 2 ) = .20

865


EMV( Release in North America) = .25(33 million) + .45(12 million) + .30(-15 million) = 9.2 million EMV(European distributor) = 12 million Optimal decision: Sell to European distributor

Posterior Probabilities for I 3 (poor response) sj

P(s j )

P(I 3 |s j )

P(s j and I 3 )

P(s j | I 3 )

__________________________________________________________________________ .5 .1 (.5)(.1) = .05 .05/.17 = .294 s1 s2

.3

.2

(.3)(.2) = .06

.06/.17 = .353

s3

.2

.3

(.2)(.3) = .06

.06/.17 = .353

(I 3 ) = .17 EMV( Release in North America) = .294(33 million) + .353(12 million) + .353(-15 million) = 8.6 million EMV(European distributor) = 12 million Optimal decision: Sell to European distributor. EMV` = .63(21.9 million) + .20(12 million) + .17(12 million) = 18.2 million EVSI = EMV` - EMV* = 18.2 million – 17.1 million = 1.1 million Because EVSI is greater than the sampling cost (100,000) the studio executives should show the movie to a random sample of North Americans. If the response is a rave review release the movie in North America. If not sell it to Europe.

866


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.