PDF Ch10

Statistics for Business and Economics
Ninth Edition, Global Edition
Chapter 10
Hypothesis
Testing: Additional
Topics
Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 1

Chapter Goals
After completing this chapter, you should be able to:
• Test hypotheses for the difference between two population means
– Two means, matched pairs
– Independent populations, population variances known
– Independent populations, population variances unknown but equal
• Complete a hypothesis test for the difference between two proportions
(large samples)
• Use the F table to find critical F values
• Complete an F test for the equality of two variances

Two Sample Tests

Section 10.1 Dependent Samples
Dependent Samples
Tests of the Difference Between Two Normal Population
Means: Dependent Samples
Tests Means of 2 Related Populations
• Paired or matched samples
• Repeated measures (before/after)
• Use difference between paired values:
di = xi − yi
• Assumptions:
– Both Populations Are Normally Distributed
Test Statistic: Dependent Samples
Population Means, Dependent Samples
For tests of the

The test statistic for the mean
following form: difference is a t value, with n − 1
degrees of freedom:
𝐻0 : 𝜇𝑥 − 𝜇𝑦 ≥ 0 𝑑ሜ
𝐻0 : 𝜇𝑥 − 𝜇𝑦 ≤ 0 𝑡= 𝑠 σ 𝑑𝑖
𝑑 where 𝑑 = 𝑛
𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0 𝑛
𝑠𝑑 = sample standard dev . of differences

iation
n = the sample size (number of pairs)

Decision Rules: Matched Pairs
Matched or Paired Samples
Lower-tail test: Upper-tail test: Two-tail test:
𝐻0 : 𝜇𝑥 − 𝜇𝑦 ≥ 0 𝐻0 : 𝜇𝑥 − 𝜇𝑦 ≤ 0 𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0
𝐻1 : 𝜇𝑥 − 𝜇𝑦 < 0 𝐻1 : 𝜇𝑥 − 𝜇𝑦 > 0 𝐻1 : 𝜇𝑥 − 𝜇𝑦 ≠ 0
𝑑
Where 𝑡 = 𝑠 has 𝑛 − 1 d. f.
𝑑
𝑛
Matched Pairs Example
• Assume you send your salespeople to a “customer
service” training workshop. Has the training made a
difference in the number of complaints? You collect the
following data: σ𝑑 𝑖
𝑑=
𝑛
= −4.2
σ(𝑑𝑖 − 𝑑)2
𝑠𝑑 =
𝑛−1
= 5.67

Matched Pairs: Solution
• Has the training made a difference in the number of
complaints (at the 𝛼 = 0.05 level)?
H0 : μx − μy = 0
H1 : μx − μy ≠ 0
𝛼 = .05 𝑑 = −4.2
Critical Value = ±2.776
d. f. = n − 1 = 4 Decision: Do not reject H0

Test Statistic: (t stat is not in the reject region)
𝑑 −4.2 Conclusion: There is not a
𝑡= 𝑠 = = −1.66
𝑑 5.67
𝑛 significant change in the
5
number of complaints.
Section 10.2 Independent Samples
Population means, independent samples
Tests of the Difference Between Two Normal Population
Means: Dependent Samples
Goal: Form a confidence interval
for the difference between two
population means, 𝜇𝑥 − 𝜇𝑦
• Different populations
– Unrelated
– Independent
▪ Sample selected from one population has no effect on the
sample selected from the other population
– Normally distributed
Difference Between Two Means

Sigma Sub x Squared and Sigma Sub
y Squared Known (1 of 2)
Assumptions:
• Samples are randomly and
independently drawn
• both population distributions
are normal
• Population variances are
known

Sigma Sub x Squared and Sigma Sub
y Squared Known (2 of 2)
2 2
When 𝜎𝑥 and 𝜎𝑦 are known and
both populations are normal, the
variance of 𝑋ሜ − 𝑌ሜ is
𝜎𝑥 2 𝜎𝑦 2
𝜎 2 𝑋−
ሜ 𝑌ሜ = +
𝑛𝑥 𝑛𝑦
…and the random variable

(𝑥ǉ − 𝑦)
ǉ − (𝜇𝑋 − 𝜇𝑌 )
𝑍=
𝜎2𝑥 𝜎 2𝑦
+
𝑛𝑋 𝑛𝑌
has a standard normal distribution

Test Statistic, Sigma Sub x Squared
and Sigma Sub y Squared Known
𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0
The test statistic for

𝜇𝑥 − 𝜇𝑦 is:
(𝑥ǉ − 𝑦)
ǉ
𝑧=
𝜎𝑥 2 𝜎𝑦 2
+
𝑛𝑥 𝑛𝑦

Hypothesis Tests for Two Population
Means
Two Population Means, Independent Samples

𝐻0 : 𝜇𝑥 ≥ 𝜇𝑦 𝐻0 : 𝜇𝑥 ≤ 𝜇𝑦 𝐻0 : 𝜇𝑥 = 𝜇𝑦
𝐻1 : 𝜇𝑥 < 𝜇𝑦 𝐻1 : 𝜇𝑥 > 𝜇𝑦 𝐻1 : 𝜇𝑥 ≠ 𝜇𝑦
i. e. , i. e. , i. e. ,
𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≥ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≤ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 = 0
𝐻1 : 𝜇𝑥 – 𝜇𝑦 < 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 > 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 ≠ 0

Decision Rules (1 of 2)
Two Population Means, Independent Samples, Variances
Known
𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≥ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≤ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 = 0
𝐻1 : 𝜇𝑥 – 𝜇𝑦 < 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 > 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 ≠ 0

Sigma Sub x Squared and Sigma Sub y
Squared Unknown, Assumed Equal (1 of 2)
Assumptions:
independently drawn
• Populations are normally
distributed
unknown but assumed equal

Squared Unknown, Assumed Equal (2 of 2)
• The population variances are
assumed equal, so use the two
sample standard deviations
and pool them to estimate 𝜎
• use a t value with (𝑛𝑥 + 𝑛𝑦 − 2)

degrees of freedom

Test Statistic, Sigma Sub x Squared and
Sigma Sub y Squared Unknown, Equal

𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0 is:
(𝑥ǉ − 𝑦)
ǉ
𝑡=
𝑠𝑝2 𝑠𝑝2
+
𝑛𝑥 𝑛𝑦
Where t has (𝑛1 + 𝑛2 − 2) d.f.,

and
(𝑛𝑥 − 1)𝑠𝑥2 + (𝑛𝑦 − 1)𝑠𝑦2

𝑠𝑝2 =
𝑛𝑥 + 𝑛𝑦 − 2

Decision Rules (2 of 2)
Two Population Means, Independent Samples, Variances
Unknown
𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≥ 0 𝐻0 : 𝜇𝑥 − 𝜇𝑦 ≤ 0 𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0
𝐻1 : 𝜇𝑥 – 𝜇𝑦 < 0 𝐻1 : 𝜇𝑥 − 𝜇𝑦 > 0 𝐻1 : 𝜇𝑥 − 𝜇𝑦 ≠ 0

Pooled Variance t Test: Example
You are a financial analyst for a brokerage firm. Is there a
difference in dividend yield between stocks listed on the NYSE
& NASDAQ? You collect the following data:
Blank NYSE NASDAQ

Number 21 25
Sample mean 3.27 2.53
Sample std dev 1.30 1.16
Assuming both populations are
approximately normal with equal
variances, is there a difference in
average yield 𝛼 = 0.05 ?
Calculating the Test Statistic
H0 : μ1 − μ2 = 0 i. e. (μ1 = μ2 )
H1 : μ1 − μ2 ≠ 0 i. e. (μ1 ≠ μ2 )
The test statistic is:

(𝑋ሜ 1 − 𝑋ሜ 2 ) (3.27 − 2.53)
𝑡= = = 2.040
1 1 1 1
𝑆𝑝2 + 1.5021 +
𝑛1 𝑛2 21 25
(𝑛1 − 1)𝑆1 2 + (𝑛2 − 1)𝑆2 2 (21 − 1)1.302 + (25 − 1)1.162

𝑆𝑝2 = = = 1.5021
(𝑛1 − 1) + (𝑛2 − 1) (21 − 1) + (25 − 1)

Solution
H0 : μ1 − μ2 = 0 i. e. μ1 = μ2
H1 : μ1 − μ2 ≠ 0 i. e. μ1 ≠ μ2
α = 0.05
df = 21 + 25 − 2 = 44
Critical Values: t = ±2.0154
Test Statistic: Decision:

3.27 − 2.53 Reject 𝐻0 at 𝛼 = 0.05
𝑡= = 2.040
1 1 Conclusion:
1.5021 +
21 25
There is evidence of a
difference in means.
Squared Unknown, Assumed Unequal (1 of 2)
Assumptions:
independently drawn
• Populations are normally
distributed
unknown and assumed
unequal

Squared Unknown, Assumed Unequal (2 of 2)
Forming interval estimates:
• The population variances are
assumed unequal, so a pooled
variance is not appropriate
• use a t value with v degrees of
freedom, where
2
𝑠𝑥2 𝑠𝑦2
+
𝑛𝑥 𝑛𝑦
𝑣= 2 2 2
𝑠𝑥2 𝑠𝑦
𝑛𝑥 𝑛𝑦
+
𝑛𝑥 − 1 𝑛𝑦 − 1

Test Statistic, Sigma Sub x Squared and
Sigma Sub y Squared Unknown, Unequal
𝐻0 : 𝜇𝑥 − 𝜇𝑦 = 0 is:
𝑥ǉ − 𝑦ǉ
𝑡=
2
𝑠𝑥2 𝑠𝑦
+
𝑛𝑋 𝑛𝑌
2
𝑠𝑥2 𝑠𝑦2
+
𝑛𝑥 𝑛𝑦
𝑣= 2 2
Where t has  degrees of freedom: 𝑠𝑥2 𝑠𝑦2
𝑛𝑥 𝑛𝑦
+
𝑛𝑥 − 1 𝑛𝑦 − 1

Section 10.3 Two Population
Proportions
Population proportions
Tests of the Difference Between Two Population Proportions
(Large Samples)
Goal: Test hypotheses for the

difference between two population
proportions, 𝑃𝑥 − 𝑃𝑦
Assumptions:
Both sample sizes are large,
𝑛𝑃 1 − 𝑃 > 5

Two Population Proportions
• The random variable
𝑝Ƹ𝑥 − 𝑝Ƹ𝑦 − 𝑃𝑥 − 𝑃𝑦
𝑍=
𝑃𝑥 1 − 𝑃𝑥 𝑃𝑦 1 − 𝑃𝑦
+
𝑛𝑥 𝑛𝑦
has a standard normal distribution

Test Statistic for Two Population
Proportions
𝐻0 : 𝑃𝑥 − 𝑃𝑦 = 0
is a z value:
𝑝Ƹ𝑥 − 𝑝Ƹ𝑦
𝑧=
𝑝Ƹ0 1 − 𝑝Ƹ0 𝑝Ƹ 1 − 𝑝Ƹ0
+ 0
𝑛𝑥 𝑛𝑦
𝑛𝑥 𝑝Ƹ𝑥 + 𝑛𝑦 𝑝Ƹ𝑦
𝑝Ƹ0 =
Where 𝑛𝑥 + 𝑛𝑦

Decision Rules: Proportions
𝐻0 : 𝑃𝑥 − 𝑃𝑦 ≥ 0 𝐻0 : 𝑃𝑥 − 𝑃𝑦 ≤ 0 𝐻0 : 𝑃𝑥 − 𝑃𝑦 = 0
𝐻1 : 𝑃𝑥 − 𝑃𝑦 < 0 𝐻1 : 𝑃𝑥 − 𝑃𝑦 > 0 𝐻1 : 𝑃𝑥 − 𝑃𝑦 ≠ 0

Example 1: Two Population
Proportions (1 of 3)
Is there a significant difference between the
proportion of men and the proportion of
women who will vote Yes on Proposition A?
• In a random sample, 36 of 72 men and 31 of

50 women indicated they would vote Yes
• Test at the .05 level of significance

• The hypothesis test is:
𝐻0 : 𝑃𝑀 − 𝑃𝑊 = 0 (the two proportions are equal)
𝐻1 : 𝑃𝑀 − 𝑃𝑊 ≠ 0 (there is a significant difference between proportions)
• The sample proportions are:
36
𝑝Ƹ𝑀 = = .50
– Men: 72
31
𝑝Ƹ𝑊 = = .62
– Women: 50
• The estimate for the common overall proportion is:

36 31
𝑛𝑀 𝑝Ƹ𝑀 + 𝑛𝑊 𝑝Ƹ𝑊 72 + 50
𝑝Ƹ0 = = 72 50 = 67 = .549
𝑛𝑀 + 𝑛𝑊 72 + 50 122

The test statistic for 𝑃𝑀 − 𝑃𝑊 = 0 is:
𝑝Ƹ𝑀 − 𝑝Ƹ𝑊
𝑧=
𝑝Ƹ0 1 − 𝑝Ƹ0 𝑝Ƹ 1 − 𝑝Ƹ0
+ 0
𝑛1 𝑛2
.50 − .62
=
.549 1 − .549 .549 1 − .549
+
72 50 Decision: Do not reject H0
Conclusion: There is not
= −1.31
significant evidence of a
difference between men
Critical Values = ±1.96
and women in proportions
For α = .05 who will vote yes.
Section 10.4 Tests of Equality of Two
Variances
• Goal: Test hypotheses about two
population variances
𝐻0 : 𝜎𝑥2 ≥ 𝜎𝑦2
𝐻1 : 𝜎𝑥2 < 𝜎𝑦2 Lower-tail test
𝐻0 : 𝜎𝑥2 ≤ 𝜎𝑦2
𝐻1 : 𝜎𝑥2 > 𝜎𝑦2 Upper-tail test
𝐻0 : 𝜎𝑥2 = 𝜎𝑦2
𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2 Two-tail test
The two populations are assumed to be

independent and normally distributed
Hypothesis Tests for Two Variances
The random variable
𝑠𝑥2
𝜎𝑥2
𝐹= 2
𝑠𝑦
𝜎𝑦2
Has an F distribution with 𝑛𝑥 − 1

numerator degrees of freedom and
𝑛𝑦 − 1
denominator degrees of
freedom
Denote an F value with 𝑣1 numerator and 𝑣2
denominator degrees of freedom by 𝐹𝑣1,𝑣2
Test Statistic
The critical value for a hypothesis test
about two population variances is
𝑠𝑥2
𝐹= 2
𝑠𝑦
where F has 𝑛𝑥 − 1 numerator

degrees of freedom and 𝑛𝑦 − 1
denominator degrees of freedom

Decision Rules: Two Variances
2
Use 𝑠𝑥 to denote the larger variance. 𝐻0 : 𝜎𝑥2 = 𝜎𝑦2
𝐻0 : 𝜎𝑥2 ≤ 𝜎𝑦2 𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2
𝐻1 : 𝜎𝑥2 > 𝜎𝑦2
• rejection region for a two-tail

test is:
Reject 𝐻0 if 𝐹 > 𝐹𝑛𝑥 −1,𝑛𝑦−1,𝛼 Reject H 0 if F  Fn −1,n 

x y −1,
2 2
where 𝑠𝑥 is the larger of
the two sample variances
Example 2: F Test
You are a financial analyst for a brokerage firm. You
want to compare dividend yields between stocks listed
on the NYSE & NASDAQ. You collect the following data:
Blank NYSE NASDAQ

Number 21 25
Mean 3.27 2.53
Std dev 1.30 1.16
Is there a difference in the

variances between the NYSE
NASDAQ at the α = 0.10 level?
F Test: Example Solution (1 of 2)
• Form the hypothesis test:
𝐻0 : 𝜎𝑥2 = 𝜎𝑦2 (there is no difference between variances)
𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2 (there is a difference between variances)
10
• Find the F critical values for α =. :
2
Degrees of Freedom:
• Numerator Fnx −1, ny −1, 
(NYSE has the larger 2
standard deviation):
=𝐹 10 = 2.03
– 𝑛𝑥 − 1 = 21 − 1 = 20 d.f. 20,24,0.
2
• Denominator:
– 𝑛𝑦 − 1 = 25 − 1 = 24 d.f.
F Test: Example Solution (2 of 2)
• The test statistic is: 𝐻0 : 𝜎𝑥2 = 𝜎𝑦2
𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2
• F = 1.256 is not in the rejection

region, so we do not reject 𝐻0
• Conclusion: There is not sufficient evidence

of a difference in variances at α = .10
Some Comments on Hypothesis
Testing
• A test with low power can result from:
– Small sample size
– Large variances in the underlying populations
– Poor measurement procedures
• If sample sizes are large it is possible to find
significant differences that are not practically
important
• Researchers should select the appropriate level of
significance before computing p-values

Two-Sample Tests in Excel
For paired samples (t test):
– Data data analysis 𝑡 − test: paired two sample for means
For independent samples:

• Independent sample z test with variances known:
– Data data analysis 𝑧 − test: two sample for means
For variances…
• F test for two variances:
– Data data analysis 𝐹 − test: two sample for variances

Chapter Summary (1 of 2)
• Compared two dependent samples (paired
samples)
– Performed paired sample t test for the mean difference
• Compared two independent samples
– Performed z test for the differences in two means
– Performed pooled variance t test for the differences in
two means
• Compared two population proportions
– Performed z-test for two population proportions

Chapter Summary (2 of 2)
• Performed F tests for the difference between two
population variances
• Used the F table to find F critical values

PDF Ch10

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PDF Ch10

Uploaded by

Copyright:

Available Formats

Statistics for Business and Economics

Ninth Edition, Global Edition

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 1

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 2

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 3

For tests of the

𝑠𝑑 = sample standard dev . of differences

n = the sample size (number of pairs)

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 5

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 7

Critical Value = ±2.776

d. f. = n − 1 = 4 Decision: Do not reject H0

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 10

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 11

…and the random variable

has a standard normal distribution

The test statistic for

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 13

Lower-tail test: Upper-tail test: Two-tail test:

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 14

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 15

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 18

• use a t value with (𝑛𝑥 + 𝑛𝑦 − 2)

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 19

The test statistic for

Where t has (𝑛1 + 𝑛2 − 2) d.f.,

(𝑛𝑥 − 1)𝑠𝑥2 + (𝑛𝑦 − 1)𝑠𝑦2

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 20

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 21

Blank NYSE NASDAQ

The test statistic is:

(𝑛1 − 1)𝑆1 2 + (𝑛2 − 1)𝑆2 2 (21 − 1)1.302 + (25 − 1)1.162

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 23

Test Statistic: Decision:

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 27

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 28

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 29

Goal: Test hypotheses for the

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 30

has a standard normal distribution

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 31

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 32

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 33

• In a random sample, 36 of 72 men and 31 of

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 34

• The estimate for the common overall proportion is:

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 35

The two populations are assumed to be

Has an F distribution with 𝑛𝑥 − 1

where F has 𝑛𝑥 − 1 numerator

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 40

• rejection region for a two-tail

Reject 𝐻0 if 𝐹 > 𝐹𝑛𝑥 −1,𝑛𝑦−1,𝛼 Reject H 0 if F  Fn −1,n 

Blank NYSE NASDAQ

Is there a difference in the

• F = 1.256 is not in the rejection

• Conclusion: There is not sufficient evidence

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 48

For independent samples:

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 50

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 51

Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 52

You might also like