Chapter 7

Sampling and Sampling Distributions

Learning Objectives

1. Understand the importance of sampling and how results from samples can be used to

provide estimates of population characteristics such as the population mean, the

population standard deviation, or the population proportion.

2. Know what simple random sampling is and how simple random samples are selected.

3. Understand the concept of a sampling distribution.

4. Understand the central limit theorem and the important role it plays in sampling.

5. Specifically know the characteristics of the sampling distribution of the sample mean ( )

and the sampling distribution of the sample proportion ( ).

6. Learn about a variety of sampling methods, including stratified random sampling, cluster

sampling, systematic sampling, convenience sampling, and judgment sampling.

7. Know the definition of the following terms:

parameter target population

sampled population sampling distribution

sample statistic finite population correction factor

simple random sampling standard error

sampling without replacement central limit theorem

sampling with replacement unbiased

point estimator relative efficiency

point estimate consistency

© 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.
Solutions

1. a. AB, AC, AD, AE, BC, BD, BE, CD, CE, DE

b. With 10 samples, each has a 1/10 probability.

c. E and C because 8 and 0 do not apply; 5 identifies E; 7 does not apply; 5 is skipped

because E is already in the sample; 3 identifies C; 2 is not needed because the sample of

size 2 is complete.

2. Using the last three digits of each five-digit grouping provides the random numbers:

601, 022, 448, 147, 229, 553, 147, 289, 209

Numbers greater than 350 do not apply, and the 147 can only be used once. Thus, the simple

random sample of four includes 22, 147, 229, and 289.

3. 459, 147, 385, 113, 340, 401, 215, 2, 33, 348

4. a. 5, 0, 5, 8

Vale SA, Ford, General Electric

b.

5. 283, 610, 39, 254, 568, 353, 602, 421, 638, 164

6. 2782, 493, 825, 1807, 289

7. 108, 290, 201, 292, 322, 9, 244, 249, 226, 125, (continuing at the top of column 9) 147, and

113.

8. Random numbers used: 13, 8, 27, 23, 25, 18

The second occurrence of the random number 13 is ignored.

Companies selected: Goldman Sachs, Coca Cola, UnitedHealth, Pfizer,

Travelers Companies Inc, and JP Morgan Chase.

© 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.
9. 102, 115, 122, 290, 447, 588, 351, 157, 835, 498, 55, 801

10. a. Finite population. A frame could be constructed obtaining a list of licensed drivers from

the New York State driver’s license bureau.

b. Infinite population. Sampling from a process. The process is the production line

producing boxes of cereal.

c. Infinite population. Sampling from a process. The process is one of generating arrivals

to the Golden Gate Bridge.

d. Finite population. A frame could be constructed by obtaining a listing of students

enrolled in the course from the professor.

e. Infinite population. Sampling from a process. The process is one of generating orders

for the mail-order firm.

11. a.

b.

= (–4)2 + (–1)2 + 12 (–2)2 + 12 + 52 = 48

s=

12. a. = 75/150 = .50

b. = 55/150 = .3667

13. a.

© 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.

94 +1 1

100 +7 49

85 –8 64

94 +1 1

92 –1 1

Totals 465 0 116

14. a. Two of the 40 stocks in the sample received a five-star rating.

b. Seventeen of the 40 stocks in the sample are rated Above Average with respect to risk.

c. There are eight stocks in the sample that are rated one star or two stars.

15. a.

b.

© 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.
16. a. The sampled population is U. S. adults that are 50 years of age or older.

b. We would use the sample proportion for the estimate of the population proportion.

c. The sample proportion for this issue is .74, and the sample size is 426. The number of

respondents citing education as “very important” is (.74) 426 = 315.

d. We would use the sample proportion for the estimate of the population proportion.

e. The inferences in parts b and d are being made about the population of U.S. adults who

are age 50 or older. So the population of U.S. adults who are age 50 or older is the target

population. The target population is the same as the sampled population. If the sampled

population was restricted to members of AARP who were 50 years of age or older, the

sampled population would not be the same as the target population. The inferences

made in parts b and d would only be valid if the population of AARP members age 50

or older was representative of the U.S. population of adults age 50 and older.

17. a. = 2,977/4,135 = .7200

b. = 2,770/4,135 = .6699

c. = 2,233/4,135 = .5400

18. a.

b.

c. Normal with E ( ) = 200 and =5

d. It shows the probability distribution of all possible sample means that can be observed

© 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.
 with random samples of size 100. This distribution can be used to compute the

probability that is within a specified + from μ.

19. a. The sampling distribution is normal with

E = μ = 200

For +5,

Using standard normal probability table:

At = 205, = .8413

At = 195, = .1587

= .8413 – .1587 = .6826

b. For + 10,

Using standard normal probability table:

At = 210, = .9772

At = 190, = .0228

= .9772 – .0228 = .9544

20.

© 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 standard error of the mean decreases as the sample size increases.

21. a.

b. n / N = 50 / 50,000 = .001

Use

c. n / N = 50 / 5000 = .01

Use

d. n / N = 50 / 500 = .10

Note: Only case d where n /N = .10 requires the use of the finite population correction

factor.

22. a. E( ) = 71,800 and

The normal distribution for is based on the central limit theorem.

b. For n = 120, E ( ) remains $71,800, and the sampling distribution of can still be

© 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.
 approximated by a normal distribution. However, is reduced to = 365.15.

c. As the sample size is increased, the standard error of the mean, , is reduced. This

appears logical from the point of view that larger samples should tend to provide sample

means that are closer to the population mean. Thus, the variability in the sample mean,

measured in terms of , should decrease as the sample size is increased.

23. a. With a sample of size 60

At = 72,300,

P( ≤ 72,300) = P(z ≤ .97) = .8340

At = 71,300,

P( < 71,300) = P(z < –.97) = .1660

P(71,300 ≤ ≤ 72,300) = .8340 – .1660 = .6680

b.

At = 72,300,

P( ≤ 72,300) = P(z ≤ 1.37) = .9147

At = 71,300,

P( < 71,300) = P(z < –1.37) = .0853

P(71,300 ≤ ≤ 72,300) = .9147 – .0853 = .8294

© 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.
24. a. Normal distribution,

b. Within one week means 16.5   18.5.

At = 18.5, P(z ≤ 1.75) = .9599

At = 16.5, z = –1.75. P(z < –1.75) = .0401

So P(16.5 ≤ ≤ 18.5) = .9599 – .0401 = .9198

c. Within 1/2 week means 17.0 ≤ ≤ 18.0.

At = 18.0, P(z ≤ .88) = .8106

At = 17.0, z = –.88 P(z < –.88) = .1894

P(17.0 ≤ ≤ 18.0) = .8106 – .1894 = .6212

25. This value for the standard error can be used for parts a and b.

a. P(z ≤ .95) = .8289

P(z < –.95) = .1711

Probability = .8289 – .1711 =.6578

b. P(z ≤ .95) = .8289

P(z < –.95) = .1711

Probability = .8289 – .1711 =.6578

c. The probability of being within 10 of the mean on the evidence-based reading 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.
 writing portion of the test is exactly the same as the probability of being within 10 on

the Mathematics portion of the SAT. This is because the standard error is the same in

both cases. The fact that the means differ does not affect the probability calculation.

26. a.

Within 200 means – 16,642 must be between –200 and +200.

The z value for – 16,642 = –200 is the negative of the z value for – 16,642

= 200. So we just show the computation of z for – 16,642 = 200.

n = 30 P(–.46 ≤ z ≤ .46) = .6772 – .3228 = .3544

n = 50 P(–.59 ≤ z ≤ .59) = .7224 – .2776 = .4448

n = 100 P(–.83 ≤ z ≤ .83) = .7967 – .2033 = .5934

n = 400 P(–.1.67 ≤ z ≤ .1.67) = .9525 – .0475 = .9050

b. A larger sample increases the probability that the sample mean will be within a

specified distance of the population mean. In this instance, the probability of being

within +200 of μ ranges from .3544 for a sample of size 30 to .9050 for a sample of size

400.

27. a.

At = 38.39, P(z ≤ 1.54) = .9382

At = 36.39, z = –1.54 P(z < –1.54) = .0618, thus

© 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(36.39 ≤ ≤ 38.39) = .9382 – .0618 = .876

b.

At = 28.83, P(z ≤ 1.72) = .9573

At = 26.83, z = –1.72, P(z < –1.72) = .0427, thus

P(26.83 ≤ ≤ 28.83) = .9573 – .0427 = .915

c. In part b we have a higher probability of obtaining a sample mean within $1.00 of the

population mean because the standard error for female graduates (.5798) is smaller than

the standard error for male graduates (.6505).

d. With n = 120,

At = 27.23,

P( < 27.23) = P(z < –1.60) = .055

28. a. This is a graph of a normal distribution with = 22 and .

b. Within 1 inch means 21   23

P(21   23) = P(–1.37 ≤ z ≤ 1.37) = .9147 – .0853 = .8294

The probability the sample mean will be within 1 inch of the population mean of 22

is .8294.

© 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.

Within 1 inch means 41   43

P(41   43) = P(–1.68 ≤ z ≤ 1.68) = .9535 – .0465 = .9070

The probability the sample mean will be within 1 inch of the population mean of 42

is .9070.

The probability of being within 1 inch is greater for New York in part c because

the sample size is larger.

29.  = 273  = 100

a. n = 30 Within 16 means

P(257   289) = P(–.88  z  .88) = .8106 – .1894 = .6212

b. n = 50 Within 16 means

P(257   289) = P(–1.13  z  1.13) = .8708 – .1292 = .7416

c. n = 100 Within 16 means

P(257   289) = P(–1.60  z  1.60) = .9452 – .0548 = .8904

d. None of the sample sizes in parts a, b, and c is sufficiently large. The sample size will

© 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.
 need exceed n = 100, which was used in part c.

30. a. n / N = 40 / 4000 = .01 < .05; therefore, the finite population correction factor is not

necessary.

b. With the finite population correction factor

Without the finite population correction factor

Including the finite population correction factor provides only a slightly different value

for than when the correction factor is not used.

c. P(z ≤ 1.54) = .9382

P(z < –1.54) = .0618

Probability = .9382 – .0618 = .8764

31. a. E( ) = p = .40

b.

c. Normal distribution with E( ) = .40 and = .0490.

d. It shows the probability distribution for the sample proportion .

32. a. E( ) = .40

Within ± .03 means .37 ≤ ≤ .43

© 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(z ≤ .87) = .8078

P(z < –.87) = .1922

P(.37 ≤ ≤ .43) = .8078 – .1922 = .6156

b. P(z ≤ 1.44) = .9251

P(z < –1.44) = .0749

P(.35 ≤ ≤ .45) = .9251 – .0749 = .8502

33.

The standard error of the proportion, decreases as n increases.

34. a.

Within ± .04 means .26 ≤ ≤ .34.

P(z ≤ .87) = .8078

P(z < –.87) = .1922

© 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(.26 ≤ ≤ .34) = .8078 – .1922 = .6156

b.

P(z ≤ 1.23) = .8907

P(z < –1.23) = .1093

P(.26 ≤ ≤ .34) = .8907 – .1093 = .7814

c.

P(z ≤ 1.95) = .9744

P(z < –1.95) = .0256

P(.26 ≤ ≤ .34) = .9744 – .0256 = .9488

d. P(z ≤ 2.76) = .9971

P(z < –2.76) = .0029

P(.26 ≤ ≤ .34) = .9971 – .0029 = .9942

e. With a larger sample, there is a higher probability will be within + .04 of the

population proportion p.

35. a.

© 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(1  p) .30(.70)
p   .0458
n 100

p
.30

The normal distribution is appropriate because np = 100(.30) = 30 and n(1 – p) =

100(.70) = 70 are both greater than 5.

b. P (.20 ≤ ≤ .40) = ?

P(z ≤ 2.18) = .9854

P(z < –2.18) = .0146

P(.20 ≤ ≤ .40) = .9854 – .0146 = .9708

c. P (.25 ≤ ≤ .35) = ?

P(z ≤ 1.09) = .8621

P(z < –1.09) = .1379

P(.25 ≤ ≤ .35) = .8621 – .1379 = .7242

36. a. This is a graph of a normal distribution with a mean of = .55 and

b. Within ± .05 means .50 ≤ ≤ .60

© 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(.50   .60) = P(–1.42 ≤ z ≤ 1.42) = .9222 – .0778 = .8444

c. This is a graph of a normal distribution with a mean of = .45 and

d.

Within ± .05 means .40 ≤ ≤ .50.

P(.40   .50) = P(–1.42 ≤ z ≤ 1.42) = .9222 – .0778 = .8444

e. No, the probabilities are exactly the same. This is because , the standard error, and the

width of the interval are the same in both cases. Notice the formula for computing the

standard error. It involves . So whenever p = 1 – p, the standard error will be the

same. In part b, p = .45 and 1 – p = .55. In part d, p = .55 and 1 – p = .45.

f. For n = 400, .

Within ± .05 means .50 ≤ ≤ .60.

P(.50   .60) = P(–2.01 ≤ z ≤ 2.01) = .9778 – .0222 = .9556

The probability is larger than in part b. This is because the larger sample size has

© 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.
 reduced the standard error from .0352 to .0249.

37. a. np = 525(.10) = 52.5 and n(1 – p) = 525(1 – .10) = 472.5, so is normal distributed with

and

b. P(z ≤ 2.29) = .9890

P(z < –2.29) = .0110

P(.07 ≤ ≤ .13) = .9890 – .0110 = .9780

c. P(z ≤ 1.15) = .8749

P(z < –1.15) = .1251

P(.085 ≤ ≤ .115) = .8749 – .1251 = .7498

38. a. It is a normal distribution with E( ) = .42.

b. P(z ≤ 1.05) = .8531

P(z < –1.05) = .1469

P(.39 ≤ ≤ .44) = .8531 – .1469 = .7062

c. P(z ≤ 1.75) = .9599

P(z < –1.75) = .0401

© 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(.39 ≤ ≤ .44) = .9599 – .0401 = .9198

d. The probabilities would increase. This is because the increase in the sample size makes

the standard error, , smaller.

39. a. np = 200(.80) = 160 and n(1 – p) = 200(1 – .80) = 40, so is normal distributed normal

distribution with and

b.

P(z ≤ 1.41) = .9207

P(z < –1.41) = .0797

P(.76   .84) = P(–1.41  z  1.41) = .9207 – .0797 = .8410

c. np = 450(.80) = 360 and n(1 – p) = 450(1 – .80) = 90, so is normal distributed normal

distribution with and

d.

P(z ≤ 2.12) = .9830

P(z < –2.12) = .0170

P(.76   .84) = P(–2.12  z  2.12) = .9830 – .0170 = .9660

© 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. The probability of the sample proportion being within .04 of the population mean is

increased .8410 to .9660. The likelihood of the sample proportion being within 0.04 of

the population proportion increases by .125 by increasing the sample size from 200 to

450. If the extra cost of using the larger sample size is not too great, we should probably

do so.

40. a. E ( ) = .76

Normal distribution because np = 400(.76) = 304 and n(1 – p) = 400(.24) = 96.

b. P(z ≤1.40) = .9192

P(z < –1.40) = .0808

P(.73   .79) = P(–1.40  z  1.40) = .9192 – .0808 = .8384

c.

P(z ≤ 1.92) = .9726

P(z < –1.92) = .0274

P(.73   .79) = P(–1.92  z  1.92) = .9726 – .0274 = .9452

41. a. E( ) = .17

© 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.
 Distribution is approximately normal because np = 800(.17) = 136 > 5 and n(1 – p) =

800(.83) = 664 > 5

b. P(z ≤ 1.51) = .9345

P(z < –1.51) = .0655

P(.15   .19) = P(–1.51  z  1.51) = .9345 – .0655 = .8690

c.

P(z ≤ 2.13) = .9834

P(z < –2.13) = .0166

P(.15   .19) = P(–2.13  z  2.13) = .9834

– .0166 = .9668

42. a. E ( x )=μ=400

b. σ x =σ / √ n=100 / √ 100,000=.3162

c. Normal with E( ) = 400 and = .3162.

d. It shows the probability distribution of all possible sample means that can be observed

with random samples of size 100,000. This distribution can be used to compute the

probability that is within a specified + from μ.

43.

© 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 standard error of the mean decreases and becomes extremely small as the sample

size becomes huge.

44. a. E( p) = p = .75

b.

c. Normal distribution with E( p) = .75 and = .0014

d. It shows the probability distribution of all possible sample proportions that can be

observed with random samples of size 100,000. This distribution can be used to

compute the probability that p is within a specified + from p.

45.

© 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 standard error of the sample proportion, decreases and becomes extremely small as

the sample size becomes huge.

46. a. Normal distribution,

b. Within ±1 hour means 99   101

At = 101, P(z ≤ 2.55) = .9946

At = 99, P(z < –2.55) = .0054

So P(99 ≤ x ≤ 101) = .9946 – .0054 = .9892.

c. In part a we found that P(99 ≤ x ≤ 101) = .9892, so the probability the mean annual

number of hours of vacation time earned for a sample of 15,000 blue collar and service

employees who work for small private establishments and have at least 10 years of

service differs from the population mean μ by more than 1 hour is 1 – . 9892 – .0108, or

approximately 1%.

Because these sample results are unlikely if the population mean is 100, the

sample results suggest either (a) the mean annual number of hours of vacation time

earned by blue-collar and service employees who work for small private establishments

and have at least 10 years of service is not 100 or (b) the sample is not representative of

the population of blue-collar and service employees who work for small private

establishments and have at least 10 years of service. The sampling procedure should be

carefully reviewed.

47. . This value for the standard error can be used for parts a

and b.

© 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.
 a. P (25.15 ≤ x ≤ .25.25) = ?

P(z ≤ 2.20) = .9863

P(z < –2.20) = .0137

P (25.15 ≤ x ≤ .25.25) = .9863 – .0137 = .973

b. P (25.19 ≤ x ≤ .25.21) = ?

P(z ≤ .44) = .6700

P(z < –.44) = .3300

P (25.19 ≤ x ≤ .25.21) = .6700 – .3300 =.3400

For samples of the same size, the probability of being within 05 of the mean miles per

gallon is much higher than the probability of being within .01 miles per gallon.

c. P (25.19 ≤ x ≤ .25.21) = ?

The standard error is now smaller because the sample size is larger.

P(z ≤ 0.50) = .6915

P(z < –0.50) = .3085

P (25.19 ≤ x ≤ .25.21) = .6915 – .3085 =.3830

The probability is larger here than it is in part b because the larger sample size has made

the standard error smaller.

© 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.
 d. P (24.2 ≤ x ≤ .26.2) = ?

P(z ≤ 24.20) ≈ 1.0000

P(z < –24.20) ≈ .0000

P (24.2 ≤ x ≤ .26.2) ≈ 1.0000 – .0000 ≈ 1.000

Because the probability the mean miles per gallon for a sample of 70,000 new cars and

light trucks sold in the United States in 2017 differs from the population mean μ by less

than 1 gallon is approximately 1.0000, the probability the mean miles per gallon for a

sample of 70,000 new cars and light trucks sold in the United States in 2017 differs

from the population mean μ by more than 1 gallon is approximately 1 – 1.0000 = .0000.

These sample results are unlikely if the population mean is 25.2, so these sample

results would suggest either (1) the population mean miles per gallon for new cars and

light trucks sold in the United States in 2017 is not 25.2 or (2) the sample is not

representative of the population of new cars and light trucks sold in the United States in

2017. The sampling procedure should be carefully reviewed.

48. a.

p (1  p ) .42(.58)
p   .0015
n 108, 700

p
.30
.42

The normal distribution is appropriate because np = 108,700(.42) = 45,654 and n(1 – p)

=108,700(.58) = 63,046 are both greater than 5.

© 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. P (.419 ≤ p ≤ .421) = ?

P(z ≤ 0.67) = .7486

P(z < –0.67) = .2514

P(.419 ≤ p≤ .421) = .7486 – .2514 = .4972

c. P (.4175 ≥ p ≥ .4225) = ?

P(z ≤ 1.67) = .9525

P(z < –1.67) = .0475

P(.25 ≤ p ≤ .35) = .9525 – .0475 = .9050

For samples of the same size, the probability of being within 1% of the population

proportion of repeat purchasers is much smaller than the probability of being

within .25% of the population proportion of repeat purchasers.

d. P (.41 ≤ p ≤ .43) = ?

P(z ≤ 6.68) ≈ 1.0000

P(z < –6.68) ≈ .0000

P(.41 ≤ p ≤ .43) ≈ 1.0000 – .0000 = .0000

Because the probability the proportion of orders placed by repeat customers for a

sample of 108,700 orders from the past six months differs from the population

proportion p by less than 1% is approximately 1.0000, the probability the proportion of

© 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.
 orders placed by repeat customers for a sample of 108,700 orders from the past six

months differs from the population proportion p by more than 1% is approximately 1 –

1.0000 = .0000.

These sample results are unlikely if the population proportion is .42, so these

sample results would suggest either (1) the population proportion of orders placed by

repeat customers from the past six months is not 42% or (2) the sample is not

representative of the population of orders placed by repeat customers from the past six

months. The sampling procedure should be carefully reviewed.

49. a. For a sample of 207,000 U.S. homes, the normal distribution is appropriate because np

= 207,000(.492) = 101,844 and n(1 – p) =207,000(.508) = 105156 are both greater than

5. The mean is = .492 and the standard error is

b. Within ± .002 means .490 ≤ p ≤ .494.

P(.490  p .494) = P(–1.82 ≤ z ≤ 1.82) = .9656 – .0344 = .9322

c. For a sample of 86,800 U.S. homes, the normal distribution is appropriate because np =

86,800(.492) = 42,705.6 and n(1 – p) =86,800(.508) = 44,094.4 are both greater than 5.

The mean is E ( p ) = .492 and the standard error is

d. Within ± .002 means .490 ≤ p ≤ .494

© 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(.490  p  .494) = P(–1.18 ≤ z ≤ 1.18) = .8810 – .1190 = .7620

e. Yes, the probability in d is smaller than the probability in b because the sample size in d

is smaller than the sample size in b.

50. a. Sorting the list of companies and random numbers to identify the five companies

associated with the five smallest random numbers provides the following sample.

Company Random Number

LMI Aerospace .008012

Alpha & Omega Semiconductor .055369

Olympic Steel .059279

Kimball International .144127

International Shipholding .227759

b. Step 1: Generate a new set of random numbers in column B. Step 2: Sort the random

numbers and corresponding company names into ascending order and then select the

companies associated with the five smallest random numbers. It is extremely unlikely

that you will get the same companies as in part a.

51. a.With n = 100, we can approximate the sampling distribution with a normal distribution

having

E( ) = 10,348

© 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. P(z ≤ .80) = .7881

P(z < –.80) = .2119

P(10,148   10,548) = P(–.80  z  .80) = .7881 – .2119 = .5762

The probability that the sample mean will be within $200 of the population mean

is .5762.

c. At 12,000,

P( ≥ 11,000) = 1 – P( ≤ 11,000) = 1 – P(z ≤ 6.61) 0

Yes, the research firm should be questioned. A sample mean this large is extremely

unlikely (almost 0 probability) if a simple random sample is taken from a population

with a mean of $10,348.

52. a. Normal distribution with

E = 406

b. P(z ≤ 1.50) = .9332

P(z < –1.50) = .0668

P(391   421) = P(–1.50  z  1.50) = .9332 – .0668 = .8664

c. At = 380,

© 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( ≤ 380) = P(z ≤ –2.60) = .0047

Yes, this is an unusually low performing group of 64 stores. The probability of a sample

mean annual sales per square foot of $380 or less is only .0047.

53. With n = 60, the central limit theorem allows us to conclude the sampling distribution for

part a and part b is approximately normal with

E = 320.51

b. Within ± 20 of the population mean μ means 300.51   340.51.

At = 340.51, P(z ≤ 1.94) = .9738

At = 300.51, P(z < –1.94) = .0262

P(300.51   340.51) = P(–1.94  z  1.94) = .9738 – .0262 = .9476

c. Within ± 10 of the population mean μ means 310.51   330.51.

At = 330.51, P(z ≤ .97) = .8340

At = 310.51, P(z < –.97) = .1660

P(310.51   330.51) = P(–.97  z  .97) = .8340 – .1660 = .6680

54.  = 27,175  = 7400

a.

b.

© 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( > 27,175) = P(z > 0) = .50

Note: This could have been answered easily without any calculations; 27,175 is the

expected value of the sampling distribution of .

c. P(z ≤ 1.05) = .8531

P(z < –1.05) = .1469

P(26,175   28,175) = P(–1.05  z  1.05) = .8531 – .1469 = .7062

d.

P(z ≤ 1.35) = .9115

P(z < –1.35) = .0885

P(26,175   28,175) = P(–1.35  z  1.35) = .9115 – .0885 = .8230

55. a.

N = 2,000

N = 5,000

N = 10,000

Note: With n / N ≤ .05 for all three cases, common statistical practice would be to ignore

© 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 finite population correction factor and use for each case.

b. N = 2,000

P(z ≤ 1.24) = .8925

P(z < –1.24) = .1075

Probability = P(–1.24  z  1.24) = .8925 – .1075 = .7850

N = 5,000

P(z ≤ 1.23) = .8907

P(z < –1.23) = .1093

Probability = P(–1.23  z  1.23) = .8907 – .1093 = .7814

N = 10,000

P(z ≤ 1.23) = .8907

P(z < –1.23) = .1093

Probability = P(–1.23  z  1.23) = .8907 – .1093 = .7814

All probabilities are approximately .78, which indicates that a sample of size 50 will

work well for all three firms.

56. a.

= 500/20 = 25 and n = (25)2 = 625

b. For + 25,

P(z ≤ 1.25) = .8944

© 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(z < –1.25) = .1056

Probability = P(–1.25  z  1.25) = .8944 – .1056 = .7888

57. Sampling distribution of

 
x  
n 30

0.05 0.05

 x
1.9 2.1

The area below = 2.1 must be 1 – .05 = .95. An area of .95 in the standard normal table

shows z = 1.645.

Thus,

Solve for .

58. p = .15

a. This is the graph of a normal distribution with E( ) = p = .15 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.
 b. Within ± .04 means .11 ≤ ≤ .19

P(.11   .19) = P(–1.74 ≤ z ≤ 1.74) = .9591 – .0409 = .9182

c. Within ±.02 means .13 ≤ ≤ .17.

P(.13   .17) = P(–.87 ≤ z ≤ .87) = .8078 – .1922 = .6156

59.

P ( ≥ .375) = ?

P(z < –1.02) = .1539

P ( ≥ .375) = 1 – P ( ≤ .375) = 1 – P (z ≤ –1.02) = 1 – .1539 = .8461

60. a.

Within ± .04 means .36 ≤ ≤ .44.

P(.36   .44) = P(–1.59 ≤ z ≤ 1.59) = .9441 – .0559 = .8882

b. We want P( .45).

© 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( .45) = 1 – P( ≤ .45) = 1 – P(z ≤ 1.99) = 1 – .9767 = .0233

61. a. Normal distribution with E( ) = .15 and

b. P (.12   .18) = ?

P(z ≤ 1.03) = .8485

P(z < –1.03) = .1515

P(.12   .18) = P(–1.03  z  1.03) = .8485 – .1515 =.6970

62. a.

Solve for n

b. Normal distribution with E( ) = .25 and = .0625.

[Note: (48)(.25) = 12 > 5, and (48)(.75) = 36 > 5.]

c. P ( ≥ .30) = ?

P(z ≤ .80) = .7881

P ( ≥ .30) = 1 – P ( ≤ .30) = 1 – P(z ≤ .80) = 1 – .7881 1 – .7881

= .2119

63. a. Normal distribution, .

© 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. Within ±4 hours means 7,996   8,006.

At = 8,004, P(z ≤ 1.56) = .9406

At = 7,996, P(z < –1.56) = .0594

So P(7,996 ≤ x ≤ 8,004) = .9406 – .0594 = .8812.

c. Within ±1 hour means 7,999 ≤ x ≤ 8,001.

x
At = 8,001, P(z ≤ 0.39) = .6517

x
At = 7,999, P(z < –0.39) = .3483

P(7,999 ≤ x ≤ 8,001) = .6517 – .6483 = .3034

d. In part b we found that P(7,996 ≤ x ≤ 8,004) = .8812, so the probability the mean life of

a sample of 35,000 14-watt CFLs will differ from the population mean life μ by more

than 4 hours is 1 – .8812 = .1188. If the population mean is 8,000, sample results as

extreme as this would occur more than one time out of 10 samples. The results therefore

do not provide strong evidence the population mean is different from 8,000.

64. a. Normal distribution,

b. Within ±3 minutes of the mean is the same as within 3/60 – .05 hours of the mean, i.e.,

17.55   17.65.

At = 17.65, P(z ≤ 2.86) = .9979

© 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.
 At = 17.55, z = –2.86. P(z < –2.86) = .0021

So P(17.55 ≤ x ≤ 17.65) = .9979 – .0021 = .9958

c. In part b we found that if we assume the U.S. population mean and standard deviation

are appropriate for Florida, then P(17.55 ≤ x ≤ 17.65) = .9958, so the probability the

mean for a sample of 85,020 Floridians will differ from the U.S. population mean by

more than 3 minutes is 1 – .9958 = .0042. This result is would be highly unlikely and

would suggest that the home Internet usage of Floridians differs from home Internet

usage for the rest of the United States.

65. a.

p (1  p ) .043(.957)
p   .0006
n 114, 250

p
.30
.043

The normal distribution is appropriate because np = 114,250(.043) = 4,912.75 and n(1 –

p) =114,250(.957) = 109,337.25 are both greater than 5.

b. P (.042 ≤ p ≤ .044) = ?

P(z ≤ 1.67) = .9525

P(z < –1.67) = .0475

P(.042 ≤ p≤ .044) = .9525 – .0475 = .9050

c. P (.0425 ≤ p ≤ .0435) = ?

© 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(z ≤ 0.83) = .7967

P(z < –0.83) = .2033

P(.0425 ≤ p ≤ .0435) = .7967 – .2033 = .5934

For samples of the same size, the probability of being within .1% of the population

proportion of undeliverable mail pieces is much larger than the probability of being

within .05% of the population proportion of undeliverable mail pieces.

66. a.

p (1  p ) .58(.42)
p   .0035
n 20, 000

p
.30
.58

The normal distribution is appropriate because np = 20,000(.58) = 11,600 and n(1 – p)

=20,000(.42) = 8,400 are both greater than 5.

b. P (.57 ≤ p ≤ .59) = ?

P(z ≤ 2.87) = .9979

P(z < –2.87) = .0021

P(.57 ≤ p ≤ .59) = .9979 – .0021 = .9958

c. In part b we found that P(.57 ≤ x ≤ .59) = .9958, so the probability the proportion of a

sample of 20,000 drivers that is speeding will differ from the U.S. population proportion

© 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.
 of drivers that is speeding by more than 1% is 1 – .9958 = .0042.

These sample results are unlikely if the population proportion is .58, so these

sample results would suggest either (1) the population proportion of drivers who are

speeding is not 58% or (2) the sample is not representative of the population of U.S.

drivers. The sampling procedure should be carefully reviewed.

© 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: Marion Dairies

1. We know that

• E( x ) = μ regardless of the sample size, so the mean of all possible values of x is equal

to 80.

σ 25
• The sample size is 50, so the standard error of the mean is σ x = = =3.5355 .
√ n √50
• The simple random sample of 50 customers, so the central limit theorem enables us to

conclude that the sampling distribution of the mean can be approximated by a normal

distribution.

At x = 75, we have

75−80
z= =−1.41
3.5355

The area under the curve to the left of z = –1.41 is .0793. Therefore, P( x ≤ 75) = .0793

when n = 50. The probability the mean score of Blugurt given by a simple random

sample of 50 Marion Dairies customers will be 75 or less is .0793.

2. We know that:

• E( x ) = μ regardless of the sample size, so the mean of all possible values of x is again

equal to 80.

• The sample size is now 150, so the standard error of the mean is

σ 25
σ x= = =2.0412.
√ n √150
• The simple random sample of 150 customers and the central limit theorem enable us

to conclude that the sampling distribution of the mean can be approximated by a

© 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.
 normal distribution.

At x = 75, we have

75−80
z= =−2.45
2.0412

The area under the curve to the left of z = –2.45 is .0071. Therefore, P( x ≤ 75) = .0071

when n = 150. The probability the mean score of Blugurt given by a simple random

sample of 150 Marion Dairies customers will be 75 or less is .0071.

3. Whenever the sample size is increased, the standard error of the sample mean decreases.

With n = 50, the standard error of the sample mean for Blugurt scores is 3.5355. With the

increase in the sample size to n = 150, the standard error of the sample mean is decreased

to 2.0412. Because the sampling distribution of the sample mean with n = 150 has a

smaller standard error, the potential values of x that the sample can yield have less

variation and tend to be closer to the population mean than the potential values of x when

n = 50. With a smaller standard error, it is less likely that a sample will have a mean

Blugurt score of 75 or lower.

© 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.