Chapter 8

Interval Estimation

Learning Objectives

1. Know how to construct and interpret an interval estimate of a population mean, a

population proportion, or both.

2. Understand and be able to compute the margin of error.

3. Learn about the t distribution and its use in constructing an interval estimate when  is

unknown for a population mean.

4. Be able to determine the size of a simple random sample necessary to estimate a

population mean and/or a population proportion with a specified margin of error.

5. Know the definition of the following terms:

confidence interval margin of error

confidence coefficient degrees of freedom

confidence level

Solutions

1. a.  x =  / n = 5 / 40 = .79

b. At 95%, z / n = 196
. (5 / 40 ) = 155
.

2. a. 32 + 1.645 (6 / 50 )

32 + 1.4 or 30.6 to 33.4

b. 32 + 1.96 (6 / 50 )

32 + 1.66 or 30.34 to 33.66

c. 32 + 2.576 (6 / 50 )

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 32 + 2.19 or 29.81 to 34.19

3. a. 80 + 1.96 (15 / 60 )

80 + 3.8 or 6.2 to 83.8

b. 80 + 1.96 (15 / 120 )

80 + 2.68 or 77.32 to 82.68

c. Larger sample provides a smaller margin of error.

160 + 152
x= = 156
Sample mean 2

Margin of error = 160 – 156 = 4

1.96( / n ) = 4

n = 1.96 / 4 = 1.96(15) / 4 = 7.35

n = (7.35)2 = 54

5. a. With 99% confidence z /2 = z.005 = 2.576

Margin of error = 2.576 / n = 2.576(6 / 64 ) = 1.93

b. Confidence Interval: 21.52 + 1.93 or 19.59 to 23.45

6. A 95% confidence interval is of the form x  z.025 ( / n ) .

Using Excel and the web file TravelTax, the sample mean is x = 40.31 and the sample size

is n = 400. The sample standard deviation  = 85 is known. The confidence interval is

40.31 + 1.96(8.5/ 200 )

40.31  1.18 40.31 + 1.18 or 39.13 to 41.49

7. a. Margin of error = z.025 ( / n )

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 z.025 = 1.96 ,  = $255 , n = 50

Margin of error = 1.96(255/ 50 ) = 70.68

b. From data set, sample mean x = 1,836.84.

95% confidence interval: 1836.84 + 70.68 (1766.16 to 1907.52)

8. a. Because n is small, an assumption that the population is at least approximately normal

is required so that the sampling distribution of x can be approximated by a normal

distribution.

b. Margin of error: z.025 ( / n ) = 1.96(5.5 / 10) = 3.41

c. Margin of error: z.005 ( / n ) = 2.576(5.5 / 10) = 4.48

9. x + z.025 ( / n)

19,100 + 1.96 (3027 / 55)

19,100 + 800 ($18,300 to $19,900)

The average cost to repair the smoke and fire damage that result from home fires of all

causes is $11,389. The 95% confidence interval shows the mean cost of fires resulting

from tobacco use is higher and more expensive.


x  z / 2
10. a. n

3,486 + 1.645 (650 / 120 )

3,486 + 98 or $3388 to $3584

b. 3,486 + 1.96 (650 / 120 )

3,486 + 116 or $3370 to $3602

c. 3,486 + 2.576 (650 / 120 )

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 3,486 + 153 or $3,333 to $3,639

The confidence interval gets wider as we increase our confidence level. We need a

wider interval to be more confident that the interval will contain the population mean.

11. a. .025

b. 1 – .10 = .90

c. .05

d. .01

e. 1 – 2(.025) = .95

f. 1 – 2(.05) = .90

12. a. 2.179

b. –1.676

c. 2.457

d. Use .05 column, –1.708 and 1.708

e. Use .025 column, –2.014 and 2.014

xi 80
x= = = 10
13. a. n 8

b.

xi ( xi − x ) ( xi − x ) 2

10 0 0

8 –2 4

12 2 4

15 5 25

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 13 3 9

11 1 1

6 –4 16

5 –5 25

84

( xi − x ) 2 84
s= = = 3.464
n −1 7

c. t.025 ( s / n ) = 2.365(3.464 / 8) = 2.9

d. x  t.025 ( s / n)

10 ± 2.9 or 7.1 to 12.9

14. x  t / 2 ( s / n) , df = 53

a. 22.5 ± 1.674 (4.4 / 54)

22.5 ± 1 or 21.5 to 23.5

b. 22.5 ± 2.006 (4.4 / 54)

22.5 ± 1.2 or 21.3 to 23.7

c. 22.5 ± 2.672 (4.4 / 54)

22.5 ± 1.6 or 20.9 to 24.1

d. As the confidence level increases, there is a larger margin of error and a wider

confidence interval.

x  t / 2 ( s / n)
15.

90% confidence df = 64 t.05 = 1.669

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 19.5 ± 1.669 (5.2 / 65)

19.5 ± 1.08 or 18.42 to 20.58

95% confidence: df = 64 t.025 = 1.998

19.5 ± 1.998 (5.2 / 65)

19.5 ± 1.29 or 18.21 to 20.79

16. a. Using JMP or Excel, x = 9.7063 and s = 7.9805

The sample mean years to maturity is 9.7063 years with a standard deviation of

7.9805.

b. x  t.025 ( s / n) , df = 39, t.025 = 2.023

9.7063 ± 2.023 (7.9805 / 40)

9.7063 ± 2.5527 or 7.1536 to 12.2590

The 95% confidence interval for the population mean years to maturity is 7.1536 to

12.2590 years.

c. Using JMP or Excel, x = 3.8854 and s = 1.6194

The sample mean yield on corporate bonds is 3.8854% with a standard deviation of

1.6194.

d. x  t.025 ( s / n) df = 39 t.025 = 2.023

3.8854 ± 2.023 (1.6194 / 40)

3.8854 ± .5180 or 3.3674 to 4.4034

The 95% confidence interval for the population mean yield is 3.3674 to 4.4034%.

17. Using JMP or Excel, x = 6.34 and s = 2.163

x  t.025 ( s / n) df = 49 t.025 = 2.010

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 6.34 ± 2.010 (2.163 / 50)

6.34 ± .61 or 5.73 to 6.95

18. For the JobSearch data set, x = 22 and s = 11.8862

a. x = 22 weeks

b. margin of error = t.025 s / n = 2.023(11.8862) / 40 = 3.8020

c. The 95% confidence interval is x  margin of error

22  3.8020 or 18.20 to 25.80

d. Skewness = 1.0062, data are skewed to the right. Use a larger sample next time.

19. a. t.025 ( s / n) , df = 41

t.025 = 2.02, s = 6.827

2.02 (6.827 / 42) = 2.13

x  t.025 ( s / n)
b.

32.66 ± 2.13 or $30.53 to $34.79

c. The mean cost for a mid-range meal for two in Tokyo ($40) is not in the 95%

confidence interval for comparable meals in Hong Kong. The mean cost of

comparable meals in Hong Kong is less than in Tokyo.

x = xi / n = 51, 020 / 20 = 2551

20. a.

The point estimate of the mean annual auto insurance premium in Michigan is

$2,551.

( xi − x ) 2 ( xi − 2551) 2
s= = = 301.3077
b. n −1 20 − 1

95% confidence interval: x + t.025 ( s / n) , df = 19

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 2,551 + 2.093 (301.3077 / 20 )

$2,551 + 141.01 or $2,409.99 to $2,692.01

c. The 95% confidence interval for Michigan does not include the national average of

$1,503 for the United States. We would be 95% confident that auto insurance

premiums in Michigan are above the national average.

xi 1420
x= = = 71
21. n 20

( xi − x ) 2 9492
s= = = 22.3513
n −1 20 − 1

t.025 = 2.093 df = 19

95% confidence interval: x + t.025 ( s / n)

71 + 2.093 (22.3513 / 20)

71 + 10.46 or $60.54 to $81.46 per visit

Sxi 693,000
x= = = $23,100
22. a. n 30

The point estimate of the population mean ticket sales revenue per theater is $23,100.

S(xi - x)2
s= = 3981.89
n -1

95% confidence interval: x + t.025 ( s / n) with df = 29 t.025 = 2.045

23,100 + 3981.89
2.045
30

23,100 + 1,487

The 95% confidence interval for the population mean is $21,613 to $24,587.

b. Mean number of customers per theater = 23,100/8.11 = 2,848 customers per theater

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 c. Total number of customers = 4,080(2,848) = 11,619,840

Total box office ticket sales for the weekend = 4,080(23,100) = $94,248,000
2
z.025 2 (1.96) 2 (40) 2
n= = = 61.47 Use n = 62
23. E2 102

24. a. Planning value of  = Range/4 = 36/4 = 9

z.2025 2 (196
. ) 2 ( 9) 2
n= = = 34.57 Use n = 35
b. E2 32

. ) 2 ( 9) 2
(196
n= = 77.79 Use n = 78
c. 22

(1.96) 2 (6.84) 2
n= = 79.88 Use n = 80
(1.5) 2
25.

(1.645) 2 (6.84) 2
n= = 31.65 Use n = 32
22

26. a.
2
z.025 2 (1.96) 2 (.25) 2 Use 25.
n= = = 24.01
E2 (.10) 2

If the normality assumption for the population appears questionable, this should be

adjusted upward to at least 30.

b. (1.96) 2 (.25) 2
n= = 49
(.07) 2

Use 49 to guarantee a margin of error no greater than .07. However, the U.S. EIA

may choose to increase the sample size to a round number of 50.

c. (1.96) 2 (.25) 2 Use 97

n= = 96.04
(.05) 2

For reporting purposes, the U.S. EIA might decide to round up to a sample size of

100.

60, 000 − 45, 000

27. Planning value  = = 3750
4

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 2
z.025 2 (1.96) 2 (3750) 2
n= = = 216.09 Use n = 217
E2 (500) 2
a.

(1.96) 2 (3750) 2
n= = 1350.56 Use n = 1351
(200) 2
b.

(1.96) 2 (3750) 2
n= = 5402.25 Use n = 5403
(100) 2
c.

d. Sampling 5,403 college graduates to obtain the $100 margin of error would be viewed

as too expensive and too much effort by most researchers.

z2 /2 2 (1.645) 2 (25) 2

n= 2
= = 187.92 Use n = 188
E (3) 2
28. a.

(1.96) 2 (25) 2
n= = 266.78 Use n = 267
(3) 2
b.

(2.576) 2 (25) 2
n= = 460.82 Use n = 461
(3) 2
c.

d. The sample size gets larger as the confidence level is increased. We would not

recommend 99% confidence. The sample size must be increased by 79 respondents

(267 – 188) to go from 90% to 95%. This may be reasonable. However, increasing

the sample size by 194 respondents (461 – 267) to go from 95% to 99% would

probably be viewed as too expensive and time consuming for the 4% gain in

confidence level.

29. a. 75 seconds is 1.25 minutes, so

(1.96) 2 (4) 2 Use n = 40

n= = 39.3380
(1.25) 2

(1.96) 2 (4) 2
n= = 61.4656
b. 12 Use n = 62

30. Planning value from previous study:  = 2000

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 2
z.025 2 (1.96) 2 (2000) 2
n= = = 1536.64
E2 (100) 2

Use n = 1537 to guarantee the margin of error will not exceed 100.

31. a. p = 100/400 = .25

p (1 − p ) .25(.75)
= = .0217
b. n 400

p (1 − p )
c. p  z.025
n

.25 + 1.96 (.0217)

.25 + .0424 or .2076 to .2924

32. a. .70 + 1.645 .70(.30)

800

.70 + .0267 or .6733 to .7267

.70(.30)
b. .70 + 1.96 800

.70 + .0318 or .6682 to .7318

2
z.025 p  (1 − p  ) (1.96) 2 (.35)(.65)
n= = = 349.59 Use n = 350
E2 (.05) 2
33.

34. Use planning value p* = .50

(1.96) 2 (.50)(.50)
n= = 1067.11 Use n = 1068
(.03) 2

35. a. p = 574/1007 = .57

b. Margin of Error

p (1 − p ) 0.57(0.43)
z.05 = 1.645 = 0.0257
n 1007

c. Confidence interval:

.57 ± .0257 or .5443 to .5927

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 0.5433～0.5957

d. Margin of Error

p (1 − p ) 0.57(0.43)
z.025 = 1.96 = 0.0306
n 1007

95% confidence interval

.57 + .0306 or .5394 to .6006

36. a. p = 46/200 = .23

p (1 − p ) .23(1 − .23)
= = .0298
b. n 200

p (1 − p )
p  z.025
n

.23 + 1.96(.0298)

.23 + .0584 or .1716 to .2884

p = 384 /1007 = .3813

37. a.

b. Margin of error:

p (1 − p ) .3813(1 − .3813)
z.025 = 1.96 = .0300
n 1007

p  .0300
c.

confidence interval: .3813 + .0300 or .3513 to .4113

d. p = 553 /1007 = .5492

p (1 − p ) .5492(1 − .5492)
z.025 = 1.96 = .0307
n 1007

confidence interval: .5492  .0307 or .5185 to .5799

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 e. Part c margin of error = .0300 Part d margin of error = .0307

The confidence interval in part c has the smaller margin of error. This is because p is

farther away from 0.5 in part c (the point at which the standard error is the largest)

than it is in part d.

38. a. p = 81/142 = .5704

b. Margin of error = p (1 − p ) (.5704)(.4296)

1.96 = 1.96 = .0814
n 142

Confidence interval: .5704 + .0814 or .4890 to .6518

1.962 (.5704)(.4296)
n= = 1046
(.03) 2
c.
2
z.025 p (1 − p  ) (1.96) 2 (.16)(1 − .16)
39. a. n= = = 573.68 Use 574
E2 (.03) 2

2
z.005 p (1 − p ) (2.576) 2 (.16)(1 − .16)
n= = = 990.94
E2 (.03) 2
b. Use 991

p (1 − p ) .52(1 − .52)
40. Margin of error: z.025 = 1.96 = .0346
n 800

95% confidence interval: p ± .0346

.52 + .0346 or .4854 to .5546

41. a. Margin of error = p (1 − p ) .639(1 − .639)

z.025 = 1.96 = .0228
n 1200

a.0.027

Interval estimate: .639  .0228 or .6162 to .6618

p (1 − p ) .417(1 − .417)
b. Margin of error = z.025 = 1.96 = .0234
n 1200

Interval estimate: .417 ± .0234 or .3936 to .4404

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
b.0.028

c. The margin of error is larger in part b. This is because the sample proportion is closer

to .50 in part b than in part a. This also leads to a larger interval estimate in part b.

p * (1 − p*) .50(1 − .50)

= = .0226
42. a. n 491

p * (1 − p*) = 1.96(.0226) = .0442

z.025
n

2
z.025 p  (1 − p  )
n=
b. E2

1.962 (.50)(1 − .50)

September n= = 600.25 Use 601
.042

1.962 (.50)(1 − .50)

October n= = 1067.11 Use 1068
.032

1.962 (.50)(1 − .50)

November n= = 2401
.022

1.962 (.50)(1 − .50)

Preelection n= = 9604
.012

43. a. Margin of Error = z / 2 p (1 − p ) (.90)(.10)

= 1.96 = .0201
n 857

95% confidence interval: .90 + .0201 or .8799 to .9201

b. Margin of error = 1.96 (.67)(.33) = .0315

857

95% confidence interval: .67 + .0315 or .6385 to .7015

(.56)(.44)
c. Margin of error = 1.96 = .0332
857

95% confidence interval: .56 + .0332 or .5268 to .5932

d. The margin of error increases as p gets closer to .50.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
44. a. 𝑥̅ = 326.6674

b. degrees of freedom = 10,000 and 𝑠𝑥̅ = 12,318.01, so 𝑡𝛼⁄2 𝑠𝑥̅ =

1.9600(12,318.01/√10,001) = 241.4165.

c. 𝑥̅ ± 𝑡𝛼⁄2 𝑠𝑥̅ = 326.6674 ± 241.4165 = (85.25, 568.08)

45. a. 𝑥̅ = 60.3964

b. 𝑧𝛼⁄2 𝜎𝑥̅ = 2.576(33.4460/√3500) = 1.4999

c. 𝑥̅ ± 𝑧𝛼⁄2 𝜎𝑥̅ = 60.40 ± 1.5023 = (58.8965, 61.8963)

d. The 99% confidence interval for the mean sick hours taken last year does not include

the value for the mean sick hours taken two years ago. This suggests that the mean

sick hours taken changed from two years ago to last year.

46. a. 𝑝̅ = 65120/102519 = .6352

b. 𝑧𝛼⁄2 𝜎𝑝̅ = 1.96(.0015) = .0029

c. 𝑥̅ ± 𝑧𝛼⁄2 𝜎𝑝̅ = .6352 ± .0029 = (.6323, .6381)

47. a. 𝑝̅ = 9252/20000 = .4626

b. 𝑧𝛼⁄2 𝜎𝑝̅ = 2.576(.0035) = .0090

c. 𝑥̅ ± 𝑧𝛼⁄2 𝜎𝑝̅ = .4626 ± .0090 = (.4536, .4716)

d. The 99% confidence interval for the proportion of vehicles on U.S. roads does not

include the value for the proportion reported by ABC News. This suggests that the

proportion of vehicles on U.S. roads that are speeding differs from the proportion

reported by ABC News. However, the sample was taken at 6 P.M. EST on a recent

Tuesday afternoon. We must question whether the drivers on U.S. roads at this time

are representative of all U.S. drivers. If it doesn’t, then our sample suffers from

nonsampling error.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
  15
48. a. Margin of error: z.025 = 1.96 = 4.00
n 54

b. Confidence interval: x + margin of error

33.77 + 4.00 or $29.77 to $37.77

49. a. x + t.025 ( s / n) df = 63 t.025 = 1.998

252.45 + 1.998 (74.50 / 64 )

252.45 + 18.61 or $233.84 to $271.06

b. Yes. The lower limit for the population mean at Niagara Falls is $233.84, which is

greater than $215.60.

50. a. Margin of error = t.025 ( s / n)

df = 79 t.025 = 1.990 s = 550

1.990 (550 / 80) = 122

b. x ± margin of error

1873 + 122 or $1,751 to $1,995

c. Because 92 million Americans were age 50 and older, the estimate of total

expenditures = 92(1873) = 172,316.

In dollars, we estimate that $172,316 million dollars are spent annually by Americans

of age 50 and older on restaurants and carryout food.

d. We would expect the median to be less than the mean of $1,873. The few individuals

who spend much more than the average cause the mean to be larger than the median.

This is typical for data of this type.

51. a. From the sample of 7,044 people who have never been married, the proportion who

have a healthy sleep duration is p = 4437 / 7044 = .63 . This is our point estimate of the

proportion for the population.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 The margin of error is

p (1 − p ) .63(1 − .63)
Margin of error = z.025 = 1.96 = .0113
n 6044

95% confidence interval: .63 + .0113 or .6186 to .6413

b. From the sample of 7044 people who have never been married, the sample mean

number of hours of sleep is x = 7.2257 . Thus, our point estimate of the population

mean number of hours of sleep for people who have never been married is 7.226.

Using Excel, we find that the sample standard deviation is s = .8211 .

Therefore,

s .8211
Margin of error = t.025 = 1.96 = .0192
n 7044

95% confidence interval: 7.2257 + .0192 or 7.2065 to 7.2449

c. From the sample, we see that 4,437 people had a healthy sleep duration. The mean

number of hours of sleep for those who had a healthy sleep duration is given by

x = 34, 295.1/ 4437 = 7.7293 .

52. a. Using Excel shows that the sample mean and sample standard deviation are x = 773

and s = 738.3835 .

Margin of error = t.05  s   738.3835 

 = 1.645   = 60.73
 n  400 

90% Confidence Interval: 773  60.73 or $712.27 to $833.73

b. Using Excel shows that the sample mean and sample standard deviation are x = 187

and s = 178.6207 .

Margin of error = t.05  s   178.6207 

 = 1.645   = 14.69
 n  400 

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 90% confidence interval: 187 + 14.69 or $172.31 to $201.69

c. There were 136 employees who had no prescription medication cost for the year:

p = 136 / 400 = .34 .

d. The margin of error in part a is 60.73; the margin of error in part c is 14.69. The

margin of error in part a is larger because the sample standard deviation in part a is

larger. The sample size and confidence level are the same in both parts.

53. a. Using Excel or JMP: x = 27.63

The estimated mean BMI for adults in the U.S. is 27.63. This is less than the standard

established by the National Institute of Diabetes and Digestive and Kidney Diseases.

b. Using Excel or JMP: s = 10.8484 minutes

c. x  t.025 s / n df = 1009 t.025 = 1.984

27.63 + 1.984 (10.8484/ 101 )

27.63 + 2.14 or 25.49 to 29.77

(2.33) 2 (2.6) 2
n= = 36.7 Use n = 37
54. 12

. ) 2 (8) 2
(196
55. n= = 6147
. Use n = 62
22

(2.576) 2 (8) 2
n= = 10617
. Use n = 107
22

. ) 2 (675) 2
(196
n= = 175.03 Use n = 176
56. 100 2

p (1 − p )
p  1.96
57. a. n

.47 + 1.96 (.47)(.53)

450

.47 + .0461 or .4239 to .5161

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 b. .47 + 2.576 (.47)(.53)
450

.47 + .0606 or .4094 to .5306

c. The margin of error becomes larger.

58. a. p = 200/369 = .5420

p (1 − p ) (.5420)(.4580)
1.96 = 1.96 = .0508
b. n 369

c. .5420 + .0508 or .4912 to .5928

59. a. p = .78

Margin of error = z.025 p (1 − p ) (.78)(.22)

= 1.96 = .0363
n 500

95% confidence interval: .78 + .0363 or .7437 to .8163

b. p = .45

Margin of error = z.005 p (1 − p ) (.45)(.55)

= 2.576 = .0573
n 500

95% confidence interval: .45 + .0573 or .3927 to .5073

c. The margin of error is larger in part b for two reasons: (1) with p = .45, the estimate

of the standard error is larger than with p = .78, (2) And z.005 = 2.576 for 99%

confidence is larger than z.025 = 1.96 for 95% confidence.

60. a. With 165 out of 750 respondents rating the economy as good or excellent,

p = 165/750 = .22

p (1 − p ) .22(1 − .22)
b. Margin of error = 1.96 = 1.96 = .0296
n 750

95% Confidence interval: .22 + .0296 or .1904 to .2496

c. With 315 out of 750 respondents rating the economy as poor,

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 p = 315/750 = .42

Margin of error = 1.96 p (1 − p ) .42(1 − .42)

= 1.96 = .0353
n 750

95% Confidence interval: .42 + .0353 or .3847 to .4553

d. The confidence interval in part c is wider. This is because the sample proportion is

closer to .5 in part c.

(1.96) 2 (.3)(.7)
n= = 2016.84 Use n = 2017
(.02) 2
61. a.

b. p = 520/2017 = .2578

p (1 − p )
p  1.96
c. n

.2578 + 1.96 (.2578)(.7422)

2017

.2578 + .0191 or .2387 to .2769

(2.33) 2 (.70)(.30)
n= = 1266.74 Use n = 1267
(.03) 2
62. a.

(2.33) 2 (.50)(.50)
n= = 1508.03 Use n = 1509
(.03) 2
b.

63. a. For the 18–24 age group:

.67(1 − .67)
95% margin of error: 1.96 = .0432
455

95% confidence interval: .67  .0432 or .6268 to .7132

For the 25–34 age group:

.83(1 − .83)
95% margin of error: 1.96 = .0345
455

95% Confidence interval: .83  .0345 or .7955 to .8645

For the 35–49 age group

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 .76(1 − .76)
95% margin of error: 1.96 = .0392
455

95% confidence interval: .76  .0392 or .7208 to .7992

For the 50+ age group

.78(1 − .78)
95% margin of error: 1.96 = .0380
455

95% confidence interval: .78  .0381 or .7420 to .8180

b. The 18–24 age group has the largest margin of error, so the sample size must be large

enough to reduce the margin of error for the 18–24 age group to .05 or less. Using the

proportions found in the study reported by CreditCards.com as a planning value

p* = .67.

2
(1.96) (0.67)(0.33)
n= 2
= 943.75 Use n = 944
(0.03)

This is an increase of 489 people surveyed in each age group. The added cost of the

larger sample size would have to be taken into account before deciding whether the

smaller margin of error is worth the added cost of obtaining the data.

64. a. p = 618/1993 = .3101

p (1 − p )
b. p  1.96
1993

.3101 + 1.96 (.3101)(.6899)

1993

.3101 + .0203 or .2898 to .3304

z 2 p  (1 − p  )
n=
c. E2

(1.96) 2 (.3101)(.6899)
z= = 8218.64 Use n = 8219
(.01) 2

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 No, the sample appears unnecessarily large. The .02 margin of error reported in part b

should provide adequate precision.

65. a. 𝑥̅ = 42.5926

b. degrees of freedom = 15,716 and s = 7.7835,

so 𝑡𝛼⁄2 𝑠𝑥̅ = 1.9600(7.7835/√15,717) = .1217

c. 𝑥̅ ± 𝑡𝛼⁄2 𝑠𝑥̅ = 42.5926 ± .1217 = (42.4710, 42.7143)

66. a. 𝑥̅ = 34.9872

b. degrees of freedom = 28,584 and s = 2.9246,

so 𝑡𝛼⁄2 𝑠𝑥̅ = 2.58(2.9246/√28,584) = .0446

c. 𝑥̅ ± 𝑡𝛼⁄2 𝑠𝑥̅ = 34.9872 ± .0446 = (34.9426, 35.0318)

d. The 99% confidence interval for the mean hours worked hours during the past week

does not include the value for the mean hours worked during the same week one year

ago. This suggests that the mean hours worked taken changed from last year to this

year.

67. a. 𝑝̅ = 18,233/42,296 = .4311

b. 𝑧𝛼⁄2 𝜎𝑝̅ = 1.96(.0024) = .0047

c. 𝑥̅ ± 𝑧𝛼⁄2 𝜎𝑝̅ = .4311 ± .0047 = (.4264, .4358)

The 95% confidence interval for the proportion of reports filed from Florida that dealt

with instances of fraud does not include the value for the proportion filed by the

entire country. This suggests that the proportion of instances of fraud in Florida

differs from the proportion for the United States.

68. a. 𝑝̅ = 490/8749 = .0560

b. 𝑧𝛼⁄2 𝜎𝑝̅ = 1.64(.0025) = .0040

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 c. 𝑥̅ ± 𝑧𝛼⁄2 𝜎𝑝̅ = .0560 ± .0040 = (.0520, .0600)

d. The 90% confidence interval for the proportion of California bridges that are deficient

does not include the value for the proportion of deficient bridges in the entire country.

This suggests that the proportion of California bridges that is deficient differs from

the proportion for the U.S.

Even though the IRC report indicates that California has a large proportion of the

nation’s deficient bridges, California has a large total number of bridges, so the

proportion of bridges in California that are deficient is smaller than the proportion of

deficient bridges nationwide.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Case Solutions

Case Problem 1: Young Professional Magazine

1. Descriptive statistics for the quantitative variables follow:

Variable N Mean SE Mean StDev Minimum Maximum Skewness

Age 410 30.112 0.199 4.024 19.000 42.000 -0.03

Investments 410 28538 781 15811 0.000 133400 1.71

Transactions 410 5.973 0.153 3.101 0.000 21.000 1.21

Household 410 74460 1720 34818 16200 322500 2.01

Income

Descriptive statistics for the qualitative variables follow:

Gender Male: 229 Proportion male: .5585

Plan R.E. purchase Yes: 181 Proportion yes: .4415

Broadband access Yes: 256 Proportion yes: .6244

Have Children Yes: 219 Proportion yes: .5341

2. 95% confidence intervals

Variable N Mean StDev SE Mean 95% CI

Age 410 30.1122 4.0240 0.1987 (29.7215, 30.5029)

Variable N Mean StDev SE Mean 95% CI

Household Income 410 74459.5 34818.2 1719.5 (71079.3, 77839.8)

We can conclude with 95% confidence that the mean age of subscribers to Young

Professional is between 29.72 and 30.50 years of age. And we can conclude with 95%

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
confidence that the mean household income of subscribers is between $71,079 and $77,840.

3. 95% confidence intervals

We can conclude with 95% confidence that the proportion of subscribers with broadband

access is between 57.75% and 67.13% and that the proportion of subscribers with

children is between 48.59% and 58.24%.

4. Young Professional should be a good advertising outlet for online brokers. We see that

most of the subscribers have financial investments exclusive of their homes (the mean

amount is $28,538), and some have substantial amounts of investments (several have

more than $100,000 in investments). Another factor to consider is the number of stock,

bond, and mutual fund transactions. The mean number is approximately six per year, and

several subscribers make significantly more transactions than that. Finally, a large

proportion of subscribers have broadband access (the sample proportion is .6244), and

this makes them more likely to do business with an online broker.

5. The survey results allow us to estimate that the mean age of subscribers is 30.12 years

and that 53.41% of subscribers have children. Given the age of subscribers, it is

reasonable to assume that their children are young. Thus, we conclude that subscribers to

Young Professional would be a good target market for companies selling educational

software and computer games for young children.

6. A variety of answers are possible here, but the survey results clearly show that articles

about investing would appeal to many readers. Articles about real estate and architecture

would probably appeal to those subscribers planning to make real estate purchases.

Technology-related articles would probably appeal to readers, as would occasional

articles on parenting and child care.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Case Problem 2: Gulf Real Estate Properties

The variables are as follows:

GV List The list price of a Gulf View condominium

GV Sale The sale price of the Gulf View condominium

GV Days The number of days to sell the Gulf View condominium

NGV List The list price of a No Gulf View condominium

NGV Sale The sale price of the No Gulf View condominium

NGV Days The number of days to sell the No Gulf View condominium

1/2. Descriptive statistics follow.

Descriptive Statistics: GV List, GV Sale, GV Days, NGV List, NGV Sale, NGV Days

Variable N Mean Median TrMean StDev SE Mean

GV List 40 474.0 437.0 462.0 197.3 31.2

GV Sale 40 454.2 417.5 441.2 192.5 30.4

GV Days 40 106.00 96.00 102.64 52.22 8.26

NGV List 18 212.8 212.5 210.0 48.9 11.5

NGV Sale 18 203.2 203.5 201.8 43.9 10.3

NGV Days 18 135.0 126.0 127.8 76.3 18.0

Variable Minimum Maximum Q1 Q3

GV List 169.9 975.0 332.7 551.9

GV Sale 165.0 975.0 314.3 530.6

GV Days 28.00 282.00 71.25 138.75

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 NGV List 148.0 322.0 174.9 241.0

NGV Sale 135.5 292.5 172.4 230.0

NGV Days 48.0 338.0 62.5 154.8

3. Gulf View condominiums are the more expensive units as usually anticipated. A Gulf

View condominium lists for a mean price of $474,000 and a median price of $437,000. A

No Gulf View condominium lists for a mean price of $212,800 and a median price of

$212,500. The Gulf View condominiums are more than twice as expensive as the No

Gulf View condominiums.

The standard deviation of the list price for Gulf View condominiums is $197,300;

the standard deviation of the list price for No Gulf View condominiums is $48,900. Thus,

the Gulf View condominiums have a greater variation in list price.

The most expensive list price is for a Gulf View condominium at $975,000, and

the least expensive list price is a No Gulf View condominium at $148,000.

A box plot shows that the No Gulf View condominiums do not have any outliers.

A similar box plot shows the highest four Gulf View list prices are outliers: one each at

$885,000 and $895,000 and two at $975,000. These show 4/40 or 10% of the Gulf View

condominiums have an unusually high list price.

The mean number of days to sell a condominium is slightly better for Gulf View

condominiums than for No Gulf View condominiums (106 days vs. 135 days). On

average, selling a Gulf View condominium takes slightly more than three months and

selling a No Gulf View condominium takes a little more than four months. The quickest

sale was a Gulf View condominium that sold in less than a month (28 days). The slowest

sale was for a No Gulf View condominium that took almost a year to sell (338 days).

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 Gulf View condominiums with a mean list price of $474,000 and a mean sale

price of $454,200 sell on average $474,000 – $454,200 = $19,800 (4.2%) below list

price. No Gulf View condominiums with a mean list price of $212,800 and a mean sale

price of $203,200 appears to sell on average $212,800 – $203,200 = $9,600 (4.5%) below

list price. There is only a small percentage difference in list and sale price for the two

types of condominiums.

In summary, Gulf View condos are substantially more expensive and tend to sell

slightly faster. The discount off the list price is roughly 4.0% to 4.5% for both Gulf View

and No Gulf View condos.

4. 95% confidence intervals:

Gulf View mean sales price $392,700 to $515,800

Gulf View mean days to sell 89 to 123

5. 95% confidence intervals:

No Gulf View mean sales price $181,400 to $225,000

No Gulf View mean days to sell 97 to 173

6. Gulf View condominiums

With n = 40, the margin of error is $59,600. To reduce the margin of error to

$40,000, the recommended sample size:

2
z.025 2 1.962 (192.518)2
n= = = 89
E2 402

No Gulf View condominiums

With n = 18, the margin of error is $21,800. To reduce the margin of error to

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 $15,000, the recommended sample size:

2
z.025 2 1.962 (43.9) 2
n= = = 33
E2 152

7. From part 3, a Gulf View condominium sells on average 4.2% below its list price.

The estimated selling price is $589,000 × .958 = $564,262. Using the mean days,

it is estimated to sell in about 106 days.

From part 3, a No Gulf View condominium sells on average 4.5% below its list

price.

The estimated selling price is $285,000 × .955 = $272,175. Using the mean days,

it is estimated to sell in about 135 days.

Case Problem 3: Metropolitan Research, Inc.

Descriptive statistics follow.

Midpoint Count

25000 1 *

35000 6 * * * * * *

45000 1 *

55000 5 * * * * *

65000 11 * * * * * * * * * * *

75000 8 * * * * * * * *

85000 9 * * * * * * * * *

95000 3 * * *

105000 1 *

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 115000 3 * * *

125000 1 *

135000 1 *

N MEAN MEDIAN TRMEAN STDEV SEMEAN

MILES 50 73340 72705 72705 24899 3521

MIN MAX Q1 Q3

MILES 25066 138114 59881 87309

The 95% confidence interval for the population mean is 66,264 to 80,416, which

indicates there is a 95% confidence that this interval contains the population mean.

The mean is 73,340, and the median is 72,705. The first quartile of 59,881 shows

25% of the repairs occurred with fewer than 60,000 miles on the vehicle. Also, the

histogram shows seven (14%) repairs occurred with fewer than 40,000 miles. These data

tend to support the conclusion of early transmission failures for this automobile

manufacturer.

To bring the precision for the population mean to within  5,000 miles, a sample

size of 96 automobile transmission repair records would be needed as follows:

z.2025 2 (196
. ) 2 (24,899) 2
n= = = 95.3 or 96
E2 (5,000) 2

Additional information that would be helpful to more fully evaluate the

transmission problem include:

1. transmission failure data for other automobile manufacturers,

2. the proportion of all automobiles that experience the transmission failures, and

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 3. industry standards for transmission failures.

With this information, we could make comparative statements about how the

manufacturer in question compares to other manufacturers as well as industry standards.

However, with the data available, the manufacturer in question appears to have problems

with early transmission failures.

© 2019 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.