Chapter 20: Additional Topics in Sampling 511

Chapter 20:

Additional Topics in Sampling

20.1 Answers should refer to each of the steps outlined in Figure 20.1

20.2 Answers should refer to each of the steps outlined in Figure 20.1

20.3 Answers should refer to each of the steps outlined in Figure 20.1

20.4 Answers should refer to each of the steps outlined in Figure 20.1

20.5 Answers should refer to each of the steps outlined in Figure 20.1

20.6 Answers should deal with issues such as (a) the identification of the correct population, (b) selection (nonresponse) bias, (c) response bias

20.7 Answers should deal with issues such as (a) the identification of the correct

population, (b) selection (nonresponse) bias, (c) response bias

20.8  Answers should deal with issues such as (a) the identification of the correct population, (b) selection (nonresponse) bias, (c) response bias

20.9  While the recall method will lower the number of nonresponses, a bias may be built in since this technique will reduce the participation of individuals who are absent on Thursday evenings (e.g., night students, second shift workers, shopping mall employees, etc.)

20.10 Within Minitab, go to Calc è Make Patterned Data… in order to generate a simple set of numbers of size ‘n’ or ‘N’. Enter first value as 1, last value as the total number of stocks traded on the New York Stock Exchange. Go to Calc è Random Data è Sample from Columns… in order to generate a simple random sample of size ‘n’ from the number of all stocks. “Sample ____ rows from column(s):” For Exercise 20.10 enter 20 as the number of rows to sample from. The results will be the observation numbers in the list to include in the sample.


20.11 Within Minitab, go to Calc è Make Patterned Data… in order to generate a simple set of numbers of size ‘n’ or ‘N’. Enter first value as 1, last value as the total number of houses advertised for sale in your city. Go to Calc è Random Data è Sample from Columns… in order to generate a simple random sample of size ‘n’ from the number of all houses advertised for sale in your city. “Sample ____ rows from column(s):” For Exercise 20.11 enter 15 as the number of rows to sample from. The results will be the observation numbers in the list to include in the sample.

20.12 Within Minitab, go to Calc è Make Patterned Data… in order to generate a simple set of numbers of size ‘n’or ‘N’. Enter first value as 1, last value as 12,723. Go to Calc è Random Data è Sample from Columns… in order to generate a simple random sample of size ‘n’. “Sample ____ rows from column(s):” Enter 100 as the number of rows to sample from. The results will be the observation numbers in the list to include in the sample.

20.13 Within Minitab, go to Calc è Make Patterned Data… in order to generate a simple set of numbers of size ‘n’ or ‘N’. Enter first value as 1, last value as 984 which is the total number of pages in the text. Go to Calc è Random Data è Sample from Columns… in order to generate a simple random sample of size ‘n’. “Sample ____ rows from column(s):” Enter 50 as the number of rows to sample from. The results will be the observation numbers in the list to include in the sample.

20.14 , = .7519

9.7 ± 1.96 (.7519) (8.2262, 11.1738)

20.15 a.

b. = 28.9216

c. 127.43 ± 1.645 () (118.5834, 136.2766)

d.  [137.43 – 117.43]/2 = 10 = , solving for z: 1.86

tabled value of .9686 yields a confidence level of 93.72% or an of .0628

20.16 = .6936

7.28 ± 2.58 (.6936) (5.4904, 9.0696)


20.17 a. false: as n increases, the confidence interval becomes narrower for a given N and s2

b.  true

c.  true: the finite population correction factor is larger to account for the fact that a smaller proportion of the population is represented as N increases relative to n.

d.  true

20.18

20.19 99% confidence interval:

where,

= 142.1167

1833.30

1466.6390 < < 2199.9610

20.20 95% confidence interval: using

, where

= 4,409.8619

104,492.6

95,849.2706 < < 113,135.9294

20.21  90% confidence interval:

, where

= 86.7054

910

767.3696 < < 1,052.6304


20.22 = 143/35 = 4.0857

90% confidence interval:

, where

= 52.9210

490.2857

403.2307 < < 577.3407

20.23 = x/n = 39/400 = .0975

= .0125

95% confidence interval: .0975 ± 1.96(.0125): .073 up to .1220

20.24 = 56/100 = .56

= .0435

90% confidence interval: .56 ± 1.645(.0435): .4884 up to .6316

20.25 = 37/120 = .3083

= .0309

95% confidence interval: .3083 ± 1.96(.0309): .2477 up to .3689

20.26 = 31/80 = .3875

= .0493

90% confidence interval: .3875 ± 1.645(.0493): .3064 up to .4686

128.688 < Np < 196.812 or between 129 and 197 students intend to take the final.

20.27 a. = = 342.089


b.

,

c. 95% confidence interval: 342.089 ± 1.96

335.609 up to 348.569

20.28 a. ,

90% confidence interval: 43.3 ± 1.645 : 40.806 up to 45.794

b. = = 37.3306

c. ,

90% confidence interval: 37.3306 ± 1.645 : 36.0313 up to 38.6299

95% confidence interval: 37.3306 ± 1.96 : 35.7825 up to 38.8787

20.29 a.

90% confidence interval: 2.5± 1.645 : 2.3102 up to 2.6897

b. = = 3.2387

c. ,

,

90% confidence interval: 3.2387 ± 1.645 : 3.1177 up to 3.3596

95% confidence interval: 3.2387 ± 1.96 : 3.0947 up to 3.3827

20.30  a. ; 3.12 ± 1.96 :

2.8435 up to 3.3965

b. ; 3.37 ± 1.96 : 3.1431 up to 3.5969

c. ; = 3.2339

3.2339 ± 1.96 : 3.0513 up to 3.4166

20.31 a. 90% confidence interval:

, where

= 315.3409

9006.4

8487.6642 < < 9525.1358

b. from Exercise 20-28: 37.3306,

(487)(37.3306) = 18,180.0022

90% confidence interval: 18,180.0022 ± 1.645(384.6620):

17,547.2332 < N < 18,812.7716

20.32  a. = 237(120) + 198(150) + 131(180) = 81,720

b. , ,

95% confidence interval: 181.6(450) ± 1.96(450) : 77,542.3153 < N < 85,897.6847

20.33  a.


b.

95% confidence interval: .1638 ± 1.96 : .0931 up to .2345

20.34  a.

b.

,

90% confidence interval: .3467 ± 1.645 : .2550 up to .4383

95% confidence interval: .3467 ± 1.96 : .2375 up to .4559

20.35  a.

b.

,

90% confidence interval: .7214 ± 1.645 : .6649 up to .7779 95% confidence interval: .7214 ± 1.96 : .6541 up to .7887

20.36  a. = 56 observations

b. = 68 observations

20.37  a. = 37 observations

b. = 32 observations

20.38  a. = 55 observations

b. = 60 observations

20.39  a. observations

b. = 52 observations

20.40  a. = 74 observations

b. = 88 observations

20.41  , = 262 observations

20.42  How large n?

= 57.988, take 58 observations

20.43

= 217 observations

20.44 , = 211 observations

20.45  Proportional allocation:

= 225 observations

Optimal allocation:

= 196 observations

20.46  Proportional allocation:

= 498 observations

Optimal allocation:

= 471 observations

20.47  a.

b.

90% confidence interval: 24.979 ± 1.645 : 22.8627 up to 27.0953

20.48  a.

b. = 66.409

99% confidence interval: 91.6761 ± 2.58

70.6920 up to 112.6602

20.49  a.

b.

90% confidence interval: .4636 ± 1.645 : .4069 up to .5203

20.50  a.

b.

95% confidence interval: .4507 ± 1.96: .38 up to .5214

20.51 

90% confidence interval: .231 ± 1.645 : .2146 up to .2489

20.52 , = 127 observations.

Additional sample observations needed are 127-20 = 107

20.53

= 196

observations. Additional sample observations needed: 196-60 = 136

20.54

= 160

observations. Additional sample observations needed: 160-30 = 130

20.55  - 20.57 Discussion questions – various answers

20.58  a. , s = 11.44,

90% confidence interval: 74.7 ± 1.645 : 69.089 up to 80.311

b. The interval would be wider; the z-score would increase to 1.96

20.59 a.

99% confidence interval: 492.36 ± 2.58 :

442.9139 up to 541.7501

b. 95% confidence interval: 492.36 ± 1.96 :

454.8175 up to 529.9025

c. The 90% interval is narrower; the z-score would decline to 1.645

20.60  a. ,

90% confidence interval: .623 ± 1.645: .559 up to .687

b.  If the sample information is not randomly selected, the resulting conclusions may be biased

20.61  ,

95% confidence interval: .6 ± 1.96: .504 up to .696

20.62  a.

, ,

99% confidence interval for managers in subdivision 1:

9.2 ± 2.575 : 6.997 up to 11.403

b.  99% confidence interval for all managers:

11.5845 ± 2.575: 9.8247 up to 13.3444

20.63  a.

,

95% confidence interval for mean number of errors per page in the book: 1.084 ± 1.96 : .8747 up to 1.2933

b.

where,

, 325.2

99% confidence interval for the total number of errors in the book:

242.4279 up to 407.9721 or from 243 total errors up to 408 total errors.

20.64

90% confidence interval is:

or 0.5147 up to 0.7453

20.65  a. = 30 observations

b. = 23 observations

20.66  a. observations

b. = 22 observations

20.67  Refer to section 20.4 – Stratified Random Sampling

20.68 , = 76 observations

20.69 , = 178 observations

20.70 Various answers. Answers should include a discussion of the potential for stratification of the population. Because different counties utilize different ballots and ballot techniques, a stratification by county may prove reasonable. The method used by the county could also be utilized in the stratification – e.g., butterfly ballots and hand-punched vs. electronic ballots.