Modern Business Statistics (6e)

Essentials of Modern
Business Statistics (7e)
Anderson, Sweeney, Williams, Camm, Cochran
© 2018 Cengage Learning

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or 1
otherwise on a password-protected website or school-approved learning management system for classroom use.
 Modern Business Statistics (6e)

Chapter 10, Part B

Inference About Means and Proportions with Two Populations

 Inferences About the Difference Between Two Population

Proportions

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
2
 Modern Business Statistics (6e)

Inferences About the Difference Between Two

Population Proportions
 Interval Estimation of p1 - p2
 Hypothesis Tests about p1 - p

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
3
 Modern Business Statistics (6e)

Sampling Distribution of 𝑝𝑝1̅ − 𝑝𝑝2̅ , Part 1

 Expected Value

 Standard Deviation (Standard Error)

where: n1 = size of sample taken from population1

n2 = size of sample taken from population 2
© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
4
 Modern Business Statistics (6e)

Sampling Distribution of 𝑝𝑝1̅ − 𝑝𝑝2̅ , Part 2

 If the sample sizes are large, the sampling distribution of
can be approximated by a normal probability distribution.

 The sample sizes are sufficiently large if all of these conditions are
met:

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
5
 Modern Business Statistics (6e)

Interval Estimation of p1 - p2 , Part 1

 Interval Estimate

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
6
 Modern Business Statistics (6e)

Interval Estimation of p1 - p2 , Part 2

Example: Tax preparation firm

A tax preparation firm is interested in comparing the quality of work at

two of its regional offices. By randomly selecting samples of tax returns
prepared at each office and verifying the sample return’s accuracy, the
firm will be able to estimate the proportion of erroneous returns
prepared at each office.

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
7
 Modern Business Statistics (6e)

Interval Estimation of p1 - p2 , Part 3

Example: Tax preparation firm

Independent random samples from the two offices provide the

following information:

Office 1 Office 2
n1 = 250 n1 = 250
Number of returns with Number of returns with
errors = 35 errors = 27

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
8
 Modern Business Statistics (6e)

Point Estimator of the Difference Between Two Population Proportions

Example: Tax preparation firm

p1 = proportion of erroneous returns for population 1 (office 1)

p2 = proportion of erroneous returns for population 2 (office 2)

𝑝𝑝1̅ = sample proportion for a random sample from population 1

𝑝𝑝̅2 = sample proportion for a random sample from population 1

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
9
 Modern Business Statistics (6e)

Interval Estimation of p1 - p2, Part 4

Example: Tax preparation firm

The 90% confidence interval for the difference in the

proportion of erroneous tax returns prepared by two
offices is .005 to .095.
© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
10
 Modern Business Statistics (6e)

Interval Estimation of p1 - p2, Part 5

Excel Formula
Office 1 Office 2
Office 1 Office 2
No No
No No
No No Interval Estimate of Difference in

and Value
No No
Interval Estimate of Difference in Population Proportions
No No
No No Population Proportions
No No
No No

worksheet
Empty cell Office 1 Office 2

Empty cell Office 1 Office 2 No No

No No Sample size 250 300

Sample size =COUNTA =COUNTA(B2 Yes No

Yes No (A2:A251) :B301) Response of Interest Yes Yes

No No Count for Response 35 27

No No Response of Interest Yes Yes

Count for Response =COUNTIF =COUNT(B2: No No Sample Proportion 0.14 0.09

No No (A2:A25,E B301,F6)
6) Empty cell Empty cell Empty cell
No No
No No
Sample Proportion =E7/E5 =F7/F5 Confidence Coefficient 0.9 Empty cell
No No
No No Empty cell Empty cell Empty cell Level of Significance 0.1 Empty cell
No No
No No Confidence Coefficient 0.9 Empty cell Z Value 1.645 Empty cell

Level of Significance =1-E10 Empty cell No No Empty cell Empty cell Empty cell
No No
Z Value =NORM.S.I Empty cell No No Standard of Error 0.0275 Empty cell
No No NV(1-
E11/2)
Margin of Error 0.0452 Empty cell
No No
No No Empty cell Empty cell Empty cell
Empty cell Empty cell Empty cell
No Yes
No Yes Standard of Error =SQRT(E8 Empty cell
Point Estimate of Difference 0.05 Empty cell
*(1-
E8)/E5+F8 Yes No
Yes No *(1- Lower Limit 0.0048 Empty cell
F8)/F5)
No No Upper Limit 0.0952 Empty cell
No No Margin of Error =E12*E14 Empty cell

Yes No
Yes No Empty cell Empty cell Empty cell

Point Estimate of Difference =E8-F8 Empty cell No No

No No
Lower Limit =E17-E15 Empty cell Empty Cell No
Empty Cell No
Upper Limit =E17+E15 Empty cell
Empty Cell No
Empty Cell No

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
11
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 1

We focus on tests involving no difference between the two population
proportions (i.e. p1 = p2)

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
12
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 2

Pooled Estimator of p when p1 = p2 = p

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
13
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 3

Test Statistic

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
14
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2 , Part 4

Example: Tax preparation firm

Let us assume that the firm wants to use a hypothesis test to

determine whether the error proportions differ between the two
offices.
Can we conclude, using a .10 level of significance that the error
proportions differ between the two offices?

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
15
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 5

Example: Tax preparation firm
p -Value approach
1. Develop the hypotheses.

p1 = proportion of erroneous returns for population 1 (office 1)

p2 = proportion of erroneous returns for population 2 (office 2)

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
16
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 6

Example: Tax preparation firm
p -Value approach
2. Specify the level of significance. α = .10
3. Compute the value of the test statistic.

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
17
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 7

Example: Tax preparation firm
p –Value Approach
4. Compute the p –value.
For z = 1.85, the area in the upper tail is 1.0000 - .9678 = .0322
Doubling this for a two-tailed test, p–value = 2(0.0322) the p–value
= .0644
5. Determine whether to reject H0.
Because p–value < α = .10, we reject H0.
We can conclude that the error rates differ between two offices.

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
18
 Modern Business Statistics (6e)

Hypothesis Tests about p1 - p2, Part 8

Excel Office 1

No
Office 2

No
Office 1

No
Office 2

No

Formula and
Hypothesis Test Concerning Difference Hypothesis Test Concerning Difference
No No Between Population Proportions No No Between Population Proportions
No No Empty cell Office 1 Office 2
No No

Value
Empty cell Office 1 Office 2

No No No No
Sample size =COUNTA(A2:A251) =COUNTA(B2:B301) Sample size 250 300

No No No No
Response of Interest Yes Yes Response of Interest Yes Yes

worksheet
Yes No Yes No
Count for Response =COUNTIF(A2:A251,E6) =COUNTIF(B2:B301,F6) Count for Response 35 27
No No No No

Sample Proportion 0.14 0.09

No No Sample Proportion =E8-F8 =F7/F5 No No

No No Empty cell Empty cell Empty cell No No Empty cell Empty cell Empty cell

No No No No Hypothesized Value 0 Empty cell

Hypothesized Value 0 Empty cell

No No No No
Point Estimate of Difference =E8-F8 Empty cell Point Estimate of Difference 0.05 Empty cell

No No No No
Empty cell Empty cell Empty cell Empty cell Empty cell Empty cell
No No No No
Pooled Estimate of p =(E5*E8+F5*F8)/(E5+F5) Empty Cell Pooled Estimate of p 0.1127 Empty cell
No No No No

No Yes Standard Error =SQRT(E13*(1- Empty cell No Yes Standard Error 0.0271 Empty cell
E13)*(1/E5+1/F5)

Yes No Test Statistic =(E11-E10)/E14 Empty cell Yes No Test Statistic 1.8462 Empty cell

No No No No
Empty cell Empty cell Empty cell Empty cell Empty cell Empty cell

Yes No Yes No
p-value (lower Tail) =NORM.S.DIST(E15,TRUE) Empty cell p-value (lower Tail) 0.9676 Empty cell

No No No No
p-value (Upper Tail) =1- Empty cell p-value (Upper Tail) 0.0324 Empty cell
Empty No NORM.S.DIST(E15,TRUE) Empty No
Cell Cell p-value (Two Tail 0.0649 Empty cell
p-value (Two Tail =2*MIN(E17,E18) Empty cell

Empty No Empty No
Cell Cell

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or
otherwise on a password-protected website or school-approved learning management system for classroom use.
19
 Modern Business Statistics (6e)

End of Chapter 10, Part B

© 2018 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or 20
otherwise on a password-protected website or school-approved learning management system for classroom use.