CHAPTER 10

10.1 df = n1 + n2 – 2 = 12 + 15 – 2 = 25

(n1 − 1) ⋅ S12 + (n2 − 1) ⋅ S 2 2 (7) ⋅ 42 + (14) ⋅ 52

10.2=
(a) S p2 = = 22
(n1 − 1) + (n2 − 1) 7 + 14

=tSTAT
(X 1 )
− X 2 − ( µ1 − µ2 )
=
(=
42 − 34 ) − 0
3.8959
 1 1 1 1 
S p2  +  22  + 
 n1 n2   8 15 
(b) d.f. = (n1 – 1) + (n2 – 1) = 7 + 14 = 21
(c) Decision rule: d.f. = 21. If tSTAT > 2.5177, reject H0.
(d) Decision: Since t = 3.8959 is greater than the critical bound of 2.5177, reject H0. There is
enough evidence to conclude that the first population mean is larger than the second
population mean.

10.3 Assume that you are sampling from two independent normal distributions having equal variances.

1 1 1 1 
10.4 (X 1 )
− X 2 ± t S p2  +  =( 42 − 34 ) ± 2.0796 22  + 
 8 15 
3.7296 ≤ µ1 − µ2 ≤ 12.2704
 n1 n2 

10.5 df = n1 + n2 − 2 = 5 + 4 – 2 = 7

10.6 PHStat output:

Data
Hypothesized Difference 0
Level of Significance 0.01
Population 1 Sample
Sample Size 5
Sample Mean 42
Sample Standard Deviation 4
Population 2 Sample
Sample Size 4
Sample Mean 34
Sample Standard Deviation 5

Intermediate Calculations
Population 1 Sample Degrees of Freedom 4
Population 2 Sample Degrees of Freedom 3
Total Degrees of Freedom 7
Pooled Variance 19.85714
Difference in Sample Means 8
t Test Statistic 2.676242

Copyright ©2019 Pearson Education Ltd.

373
374 Chapter 10: Two-Sample Tests

10.6
cont.

Upper-Tail Test
Upper Critical Value 2.997949
p-Value 0.015856
Do not reject the null hypothesis
H0: µ1 ≤ µ2 H1: µ1 > µ2

Test statistic: tSTAT =

(X 1 )
− X 2 − ( µ1 − µ2 )
= 2.6762
 1 1 
S p2  + 
 n1 n2 
Decision: Since tSTAT = 2.6762 is smaller than the upper critical bounds of 2.9979, do not reject
H0. There is not enough evidence of a difference in the means of the two populations.

10.7 (a) H0: µ1 ≥ µ2 The mean estimated amount of calories in the cheeseburger is not lower
for the people who thought about the cheesecake first than for the people who thought
about the organic fruit salad first.
H1: µ1 < µ2 The mean estimated amount of calories in the cheeseburger is lower for
the people who thought about the cheesecake first than for the people who thought about
the organic fruit salad first.
(b) Type I error is the error made in concluding that the mean estimated amount of calories in
the cheeseburger is lower for the people who thought about the cheesecake first than for
the people who thought about the organic fruit salad first when the mean estimated
amount of calories in the cheeseburger is in fact not lower for the people who thought
about the cheesecake first than for the people who thought about the organic fruit salad
first.
(c) Type II error is the error made in concluding that the mean estimated amount of calories
in the cheeseburger is not lower for the people who thought about the cheesecake first
than for the people who thought about the organic fruit salad first when the mean
estimated amount of calories in the cheeseburger is in fact lower for the people who
thought about the cheesecake first than for the people who thought about the organic fruit
salad first.
(d) PHStat output:
Separate-Variances t Test for the Difference Between
Two Means
(assumes unequal population variances)
Data
Hypothesized Difference 0
Level of Significance 0.01
Population 1 Sample
Sample Size 20
Sample Mean 780
Sample Standard Deviation 128
Population 2 Sample
Sample Size 20
Sample Mean 1041
Sample Standard Deviation 140

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 375

10.7 (d)
cont.
Intermediate Calculations
Numerator of Degrees of Freedom 3237120.6400
Denominator of Degrees of Freedom 85867.8232
Total Degrees of Freedom 37.6989
Degrees of Freedom 37
Separate Variance Denominator 42.4170
Difference in Sample Means –261
Separate-Variance t Test Statistic –6.1532

Lower-Tail Test
Lower Critical Value –2.4314
p-Value 0.0000
Reject the null hypothesis

tSTAT =
(X 1 − X 2 ) − ( µ1 − µ2 )
= –6.1532
S12 S 22
+
n1 n2
Decision: Since tSTAT = –6.1532 is smaller than the critical bound of –2.4314, reject H0.
There is evidence that the mean estimated amount of calories in the cheeseburger is lower
for the people who thought about the cheesecake first than for the people who thought
about the organic fruit salad first.
(e) The commercial would feature foods high in calories such as peanut butter and chocolate.
Based on the results from 10d, presentation of foods high in calories would decrease
estimates of the amount of calories associated with a cheeseburger.

10.8 (a) Because tSTAT = 2.8990 > 1.6620 or p-value = 0.0024 < 0.05, reject H0. There is evidence
that the mean amount of Walker Crisps eaten by children who watched a commercial
featuring a long-standing sports celebrity endorser is higher than for those who watched a
commercial for an alternative food snack.
(b) 3.4616 ≤ µ1 − µ2 ≤ 18.5384
(c) The results cannot be compared because (a) is a one-tail test and (b) is a confidence
interval that is comparable only to the results of a two-tail test.
(d) You would choose the commercial featuring a long-standing celebrity endorser.

10.9 (a) H0: μold = μnew H1: μold ≠ μnew

new old
Sample mean 4.48 2.62
Sample variance 1.29 1.90
Sample size 13 15
Pooled variance 1.62
Hypothesized difference 0
Degrees of Freedom 26
t-Test statistic 3.85
Lower critical value -2.06
Upper critical value 2.06
p-value 0.00069
Reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

376 Chapter 10: Two-Sample Tests

10.9 (b) Since the p-value is low, we reject the null hypothesis. There is enough evidence to
cont. conclude that the mean growth rates of the two groups are different. Given that the null
hypothesis is true, the probability of obtaining two samples that will yield a t-test statistic
as extreme or even more extreme than 3.85 is 0.00069.
(c) We have to assume that the distributions are normally distributed.
(d) Using a 95% confidence interval, the lower and upper limits of the confidence interval
estimate are 0.866 and 2.848. The interval does not contain 0 so we can conclude that
there is a difference in the means.
(e) As there is a difference in the means, we can say that the mean growth rate of new
member states is different from that of the old states in 2017.

10.10 (a) H0: μHUN = μPOL H1: μHUN ≠ μPOL

Hungary Poland
Sample mean 5.018 4.060
Sample variance 1.027 1.037
Sample size 50 50
Pooled variance 1.032
Hypothesized difference 0
Degrees of Freedom 98
t-Test statistic 4.715
Lower critical value -1.984
Upper critical value 1.984
p-value 8.002E-06
Reject the null hypothesis

(b) Since the p-value is low, we reject the null hypothesis. There is enough evidence to
conclude that the mean growth rates of the two groups are different. The probability of
obtaining two samples that will yield a t-test statistic as extreme or even more extreme
than 4.715 is only 0.000008002, given that the null hypothesis is true.
(c) We have to assume that the distributions are normally distributed and that the population
variances are equal.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 377

10.11 (a) H0: μc = μnc H1: μc ≠ μnc

captain non-captain
Sample mean 32.406 27.689
Sample variance 12.636 14.593
Sample size 32 704
Pooled variance 14.510
Hypothesized difference 0
Degrees of Freedom 734
t-Test statistic 6.851
Lower critical value -1.963
Upper critical value 1.963
p-value 1.55E-11
Reject the null hypothesis

Since the p-value is very low, we reject the null hypothesis. There is enough evidence to
conclude that the mean ages of captains and non-captains are different..
(b) The p-value is 1.55–11, which is the probability of obtaining two samples that will yield a
t-test statistic as extreme or even more extreme than 6.851, given that the null hypothesis
is true.
(c) The lower and upper limits of the confidence interval estimate are 3.366 and 6.069. The
estimated population mean difference of ages is between 3.37 and 6.07 years.

10.12 (a) H 0 : µO = µY
H1 : µO ≠ µY
BBS_Older BBS_Younger
Sample mean 37.65 47.5
Sample variance 69.71316 35
Sample size 20 20
Pooled variance 52.35658
Hypothesized difference 0
Degrees of Freedom 38
t-Test statistic -4.30478
Lower critical value -2.02439
Upper critical value 2.024394
p-value 0.000113
Since the p-value of 0.000113 is very low, we reject the null hypothesis. There is enough
evidence to conclude that the older and younger patients’ mean BBS values are different.
(b) The probability of obtaining two samples that will yield a t-test statistic as extreme or
even more extreme than -4.30478 is 0.000113, given that the null hypothesis is true.
(c) We need to assume that the two populations are normally distributed.
 1 1   1 1 
(d) (X O )
− X Y + t S p2  + = ( 37.65 − 47.5 ) ± 2.0244 52.35066  + 
 20 20 
 nO nY 
−14.4821 ≤ µO − µY ≤ −5.2179
You are 95% confident that the difference in mean BBS value between the older and
younger patients is between –14.4821 and –5.2179.

Copyright ©2019 Pearson Education Ltd.

378 Chapter 10: Two-Sample Tests

10.13 H 0 : µ1 = µ2 Mean waiting times of Bank 1 and Bank 2 are the same.
H1 : µ1 ≠ µ2 Mean waiting times of Bank 1 and Bank 2 are different.
PHStat output:
Separate-Variances t Test for the Difference Between
Two Means
(assumes unequal population variances)
Data
Hypothesized Difference 0
Level of Significance 0.05
Population 1 Sample
Sample Size 15
Sample Mean 4.286666667
Sample Standard Deviation 1.637985115
Population 2 Sample
Sample Size 15
Sample Mean 7.114666667
Sample Standard Deviation 2.082189324

Intermediate Calculations
Numerator of Degrees of Freedom 0.2189
Denominator of Degrees of Freedom 0.0083
Total Degrees of Freedom 26.5293
Degrees of Freedom 26
Separate Variance Denominator 0.6840
Difference in Sample Means –2.828
Separate-Variance t Test Statistic –4.1343

Two-Tail Test
Lower Critical Value –2.0555
Upper Critical Value 2.0555
p-Value 0.0003
Reject the null hypothesis
Since the p-value of 0.00031 is less than the 5% level of significance, reject the null hypothesis.
There is enough evidence to conclude that the mean waiting times are different in the two banks.
Both t tests yield the same conclusion.

10.14 (a) Because tSTAT = 2.7349 > 2.0484, reject H0. There is evidence of a difference in the mean
time to start a business between developed and emerging countries.
(b) p-value = 0.0107. The probability that two samples have a mean difference of 14.62 or
more is 0.0107 if there is no difference in the mean time to start a business between
developed and emerging countries.
(c) You need to assume that the population distribution of the time to start a business of both
developed and emerging countries is normally distributed.
(d) 3.6700 ≤ µ1 − µ2 ≤ 25.5700

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 379

10.15

The results between the two analyses were approximately equal. The t test analysis without
assuming equal variances also revealed a significant difference in mean time required to start a
business between developed and emerging countries.

10.16 (a) Because tSTAT = –2.1554 < –2.0017 or p-value = 0.03535 < 0.05, reject H0. There is
evidence of a difference in the mean time per day accessing the Internet via a mobile
device between males and females.
(b) You must assume that each of the two independent populations is normally distributed.

Copyright ©2019 Pearson Education Ltd.

380 Chapter 10: Two-Sample Tests

10.17 (a) H 0 : µT = µ FI
H1 : µT ≠ µ FI

Technology Financial Institutions

Sample mean 79614.35 31776.33333
Sample variance 9111735600 892507533.9
Sample size 20 21
Pooled variance 4896746848
Hypothesized difference 0
Total Degrees of Freedom 39
t-Test statistic 2.188023874
Lower critical value -2.02269092
Upper critical value 2.02269092
p-value 0.034724355
Reject the null hypothesis

Decision rule: If the p-value < 0.05, reject H0.

(
Test statistic: tSTAT =
)
X SA − X UAB ( µ SA − µUAB )
= 2.188 p-value = 0.0347
 1 1 
S p2  + 
 nSA nUAB 
Decision: Since the p-value < 0.05, reject H0. There is enough evidence of a difference
between the technology sector and the financial institutions sector with respect to mean
brand value.
(b) H 0 : µT = µ FI
H1 : µT ≠ µ FI

Technology Financial Institutions

Sample mean 79614.35 31776.33333
Sample variance 9111735600 892507533.9
Sample size 20 21
Hypothesized difference 0
Total Degrees of Freedom 23
t-Test statistic 2.143485262
Lower critical value -2.06865761
Upper critical value 2.06865761
p-value 0.042879027
Reject the null hypothesis

Decision rule: If the p-value < 0.05, reject H0.

Decision: Since the p-value = 0.0429 < 0.05, reject H0. There is enough evidence of a
difference between the technology sector and the financial institutions sector with respect
to mean brand value.
(c) The conclusions in (a) and (b) are not different. The sample variance of the technology
sector is 10 times as big as that of the financial sector, the test in (b) is the appropriate
way, assuming that both the samples are drawn from normally distributed populations.

10.18 The degrees of freedom is 20 – 1, or 19.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 381

10.19 Since this is an illustration of the related population, paired t-test can be used.

If the Professor assumes that the difference scores are randomly and independently selected from
a population that is normally distributed, you can use the paired t test for the mean difference in
related populations to determine whether there is a significant population mean difference.

10.20 (a) tSTAT =

( –1.5566 ) =−3.2772. Because tSTAT = –3.2772 < –2.306 or
 1.424 
 
 9 
p-value = 0.0112 < 0.05, reject H0. There is enough evidence of a difference in the mean
summated ratings between the two brands.
(b) You must assume that the distribution of the differences between the two ratings is
approximately normal.
(c) p-value = 0.0112. The probability of obtaining a mean difference in ratings that results in
a test statistic that deviates from 0 by 3.2772 or more in either direction is 0.0112 if there
is no difference in the mean summated ratings between the two brands.
(d) −2.6501 ≤ µ D ≤ −0.4610. You are 95% confident that the mean difference in summated
ratings between brand A and brand B is somewhere between –2.6501 and –0.4610.

10.21

(a) At the .05 level, there is no evidence that there is a significant difference in the mean
ratings between TV and Internet services. A paired-samples t test revealed a tSTAT of
–1.37, which was below the upper critical limit. Because tSTAT = –1.37 or p-value = 0.204,
do not reject H0.
(b) The paired samples t-test assumes the mean difference scores are normally distributed.

Copyright ©2019 Pearson Education Ltd.

382 Chapter 10: Two-Sample Tests

10.21 (c)
cont.

The differences appear to be right skewed. However, the sample size is very small, which
makes it difficult to interpret the histogram for normality. The data also contains one
outlier. Removal of this one outlier leads to a different conclusion.

After removal of row 10 from the dataset, tSTAT = –3.51 with a p-value of 0.008. In this
case, one would reject H0.
(d) Using the complete dataset, the confidence interval for the mean difference between TV
and Internet ratings is −2.387 ≤ µD ≤ 0.587.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 383

10.22 (a) Because tSTAT = –6.9984 < 2.0423 reject H0. There is evidence to conclude that the mean
download speed at AT&T is lower than at Verizon Wireless.
(b) You must assume that the distribution of the differences between the ratings is
approximately normal.
(c)

(c) Although a histogram of differences reveals slight skewness to the left, there does not
appear to be a substantial violation of the normality assumption.
(d) The confidence interval is from –6.477 to –3.551.

10.23 (a) H 0 : µD = 0
H1 : µ D ≠ 0

Non-
Cancerous cancerous
Sample mean 27.33333333 21.33333
Sample variance 119.0952381 47.80952
Sample size 15 15
Pearson's correlation 0.542718654
Hypothesized difference 0
Total Degrees of Freedom 14
t-Test statistic 2.520504151
Lower critical value -2.14478669
Upper critical value 2.144786688
p-value 0.024476795
Reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

384 Chapter 10: Two-Sample Tests

10.23 (a) D−µ p-value = 0.0245

Test statistic:=
tSTAT = 2.5205
SD
n
cont. Decision: At the 0.05 level of significance, there is sufficient evidence of a difference in
the mean iron concentrations of cancerous and non-cancerous tissues. Since the p-value =
0.0245 < 0.05, reject H0.
(b) The paired sample t-test assumes that the differences are normally distributed.
(c) Boxplot of Difference:

Both the boxplot and the normal probability plot do not indicate severe departures from
normality.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 385

10.24 (a) Define the difference in bone marrow microvessel density as the density before the
transplant minus the density after the transplant and assume that the difference in density
is normally distributed.
H 0 : µD ≤ 0 vs. H1 : µ D > 0
Excel output:
t-Test: Paired Two Sample for Means
Before After
Mean 312.1429 226
Variance 15513.14 4971
Observations 7 7
Pearson Correlation 0.295069
Hypothesized Mean Difference 0
df 6
t Stat 1.842455
P(T<=t) one-tail 0.057493
t Critical one-tail 1.943181
P(T<=t) two-tail 0.114986
t Critical two-tail 2.446914

Test statistic: tSTAT = D − µ D = 1.8425

SD
n
(a) Decision: Since tSTAT = 1.8425 is less than the critical value of 1.943, do not reject H 0 .
There is not enough evidence to conclude that the mean bone marrow microvessel
density is higher before the stem cell transplant than after the stem cell transplant.
(b) p-value = 0.0575. The probability of obtaining a mean difference in density that gives
rise to a t test statistic that deviates from 0 by 1.8425 or more is .0575 if the mean density
is not higher before the stem cell transplant than after the stem cell transplant.
S 123.7005
(c) D ± t D= 86.1429 ± 2.4469 −28.26 ≤ µ D ≤ 200.55
n 7
You are 95% confident that the mean difference in bone marrow microvessel density
before and after the stem cell transplant is somewhere between –28.26 and 200.55.
(d) You must assume that the distribution of differences between the mean density of before
and after stem cell transplant is approximately normal.

10.25 The descriptive statistics show that the sample mean is about 16 minutes less after the training
program than before it. The means and the medians are similar but the kurtosis value of the
“before” sample is higher than for a normal distribution. However the sample size is large enough
to use the paired sample t-test.
before after
Sample mean 123.685 107.395
Sample variance 94.069 85.93279487
Sample size 40 40
Hypothesized difference 0
Total Degrees of Freedom 39
t-Test statistic 6.437
Lower critical value -2.023
Upper critical value 2.023
p-value 1.28E-07
Reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

386 Chapter 10: Two-Sample Tests

10.25 The p-value is very small, so we can conclude that the two population means are not the same.
cont. The training program seems to be effective on average.
PHStat output:
Paired t Test

Data
Hypothesized Mean Difference 0
Level of significance 0.05

Intermediate Calculations
Sample Size 38
DBar -2.7632
Degrees of Freedom 37
SD 6.6309
Standard Error 1.0757
t Test Statistic -2.5688

Lower-Tail Test
Lower Critical Value -1.6871
p -Value 0.0072
Reject the null hypothesis

The PHStat output for the paired t test indicates the p-value is 0.0072 < 0.05, and, hence, reject
H 0 that the mean Cola A Adindex is no less than Cola B (Test Cola) Adindex. There is enough
evidence that the cola video ad is significant in raising the Adindex of the test Cola.

10.26 (a) H0 : µ D ≥ 0
H1: µ D < 0
Decision rule: d.f. = 39. If tSTAT < –2.4258, reject H0.
D − µD
Test statistic: tSTAT = = –9.372
SD
n
Decision: Since tSTAT = –9.372 is less than the critical bound of –2.4258, reject H0. There
is enough evidence to conclude that the mean strength is lower at two days than at seven
days.
(b) You must assume that the distribution of the differences between the mean strength of the
concrete is approximately normal.
(c) p-value is virtually 0. The probability of obtaining a mean difference that gives rise to a
test statistic that is –9.372 or less when the null hypothesis is true is virtually 0.

10.27 (a) True, decision rule is p-value 0.047 < 0.05. Thus, reject H0.
(b) False, decision rule is p-value 0.023 > 0.01. Thus, accept H0.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 387

X 1 45 X 2 25 X 1 + X 2 45 + 25
10.28 (a)
= = 0.45, p=
p1
= 2 = = 0.50,= and p = = 0.467
n1 100 n2 50 n1 + n2 100 + 50
H0: π 1 = π 2 H1: π 1 ≠ π 2
Decision rule: If Z < – 2.58 or Z > 2.58, reject H0.

=Z STAT
(=
p1 − p2 ) − (π 1 − π 2 ) ( 0.45-0.50 ) − 0 = –0.58
1 1  1 1 
p (1 − p )  +  0.467 (1-0.467 )  + 
 n1 n2   100 50 
Decision: Since ZSTAT = –0.58 is between the critical bound of ±2.58, do not reject H0.
There is insufficient evidence to conclude that the population proportion differs for
group 1 and group 2
 p1 (1 − p1 ) p2 (1 − p2 )   .45 (.55 ) .5 (.5 ) 
(b) ( p1 − p2 ) ± Z + =−0.05 ± 2.5758  + 
 n1 n2   100 50 
−0.2727 ≤ π 1 − π 2 ≤ 0.1727

10.29 PHStat output:

Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.01
Group 1
Number of Items of Interest 46
Sample Size 502
Group 2
Number of Items of Interest 71
Sample Size 502

Intermediate Calculations
Group 1 Proportion 0.091633466
Group 2 Proportion 0.141434263
Difference in Two Proportions -0.0498008
Average Proportion 0.1165
Z Test Statistic -2.4590

Two-Tail Test
Lower Critical Value -2.5758
Upper Critical Value 2.5758
p -Value 0.0139
Do not reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

388 Chapter 10: Two-Sample Tests

10.29 (a) H0: π 1 = π 2 H1: π 1 ≠ π 2 where Populations: 1 = males, 2 = females

cont. Decision rule: If ZSTAT < – 2.5758 or ZSTAT > 2.5758, reject H0.
Z STAT = –2.4590
Decision: Since ZSTAT = –2.4590 is between the two critical bounds, do not reject H0.
There is insufficient evidence of a difference between males and females in the
proportion who said they prefer window tinting as a luxury upgrade at the 0.01 level of
significance.
(b) p-value = 0.0139. The probability of obtaining a difference in two sample proportions of
–0.0498 or more in either direction when the null hypothesis is true is 0.0139.
p (1 − p1 ) p2 (1 − p2 )
(c) ( p1 − p2 ) ± Z 1 +
n1 n2
PHStat output:
Confidence Interval Estimate
of the Difference Between Two Proportions

Data
Confidence Level 99%

Intermediate Calculations
Z Value -2.5758
Std. Error of the Diff. between two Proporti 0.0202
Interval Half Width 0.0520

Confidence Interval
Interval Lower Limit -0.1018
Interval Upper Limit 0.0022

–0.1018 ≤ π 1 − π 2 ≤ 0.0022
You are 99% confident that the difference between the proportion of males and females
who said they prefer window tinting as a luxury upgrade is between –0.1018 and 0.0022.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 389

10.29 (d) PHStat output:

cont.
Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.01
Group 1
Number of Items of Interest 60
Sample Size 502
Group 2
Number of Items of Interest 71
Sample Size 502

Intermediate Calculations
Group 1 Proportion 0.119521912
Group 2 Proportion 0.141434263
Difference in Two Proportions -0.02191235
Average Proportion 0.1305
Z Test Statistic -1.0307

Two-Tail Test
Lower Critical Value -2.5758
Upper Critical Value 2.5758
p -Value 0.3027
Do not reject the null hypothesis
(a) H0: π 1 = π 2 H1: π 1 ≠ π 2 where Populations: 1 = males, 2 = females
Decision rule: If ZSTAT < – 2.5758 or ZSTAT > 2.5758, reject H0.
Z STAT = –1.0307
Decision: Since ZSTAT = –1.0307 is between the two critical bounds, do not reject
H0. There is insufficient evidence of a difference between males and females in the
proportion who said they prefer window tinting as a luxury upgrade at the 0.01 level
of significance.
(b) p-value = 0.3027. The probability of obtaining a difference in two sample proportions
of –0.0219 or more in either direction when the null hypothesis is true is 0.3027.
p (1 − p1 ) p2 (1 − p2 )
(c) ( p1 − p2 ) ± Z 1 +
n1 n2

Copyright ©2019 Pearson Education Ltd.

390 Chapter 10: Two-Sample Tests

10.29 (d) (c) PHStat output:

cont.
Confidence Interval Estimate
of the Difference Between Two Proportions

Data
Confidence Level 99%

Intermediate Calculations
Z Value -2.5758
Std. Error of the Diff. between two Proporti 0.0212
Interval Half Width 0.0547

Confidence Interval
Interval Lower Limit -0.0766
Interval Upper Limit 0.0328

–0.0766 ≤ π 1 − π 2 ≤ 0.0328
You are 99% confident that the difference between the proportion of males and
females who said they prefer window tinting as a luxury upgrade is between
–0.0766 and 0.0328.

10.30 (a) H0: π 1 ≤ π 2 H1: π 1 > π 2

Population 1 = following VOD D+4 viewing, 2 = general TV viewing
(b) PHStat output:
Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.05
Group 1
Number of Items of Interest 147
Sample Size 250
Group 2
Number of Items of Interest 35
Sample Size 201

Intermediate Calculations
Group 1 Proportion 0.588
Group 2 Proportion 0.174129353
Difference in Two Proportions 0.413870647
Average Proportion 0.4035
Z Test Statistic 8.9045

Upper-Tail Test
Upper Critical Value 1.6449
p -Value 0.0000
Reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 391

10.30 (b) Decision rule: If ZSTAT > 1.6449, reject H0.

cont. Test statistic:
( p − p2 ) − (π 1 − π 2 ) = 8.9045 p-value is essentially 0.
Z STAT = 1
1 1
p (1 − p )  + 
 n1 n2 
Decision: Since p-value < 0.05, reject H0. There is sufficient evidence to conclude that
the population proportion of those who viewed the brand on VOD D4 were more likely to
visit the brand website.
(c) Yes, the result in (b) make it appropriate to claim that ad impact is stronger following
VOD D4+ than following general TV viewing.

10.31 (a)

At the .05 level, there is evidence that a higher percentage of female shoppers bought on
impulse compared to male shoppers. The ZSTAT of –2.54 is below the lower critical value.
Because ZSTAT = –2.54 or p-value = 0.011, reject H0.
(b) The p-value of 0.011 means that the probability of obtaining an equal or larger value than
the observed test statistic, when H0 is true, is 1.1%.

10.32 (a) PHStat output:

Z Test for Differences in Two Proportions

Data Confidence Interval Estimate

Hypothesized Difference 0 of the Difference Between Two Proportions
Level of Significance 0.01
Group 1 Data
Number of Items of Interest 930 Confidence Level 99%
Sample Size 1000
Group 2 Intermediate Calculations
Number of Items of Interest 230 Z Value -2.5758
Sample Size 1000 Std. Error of the Diff. between two Proporti 0.0156
Interval Half Width 0.0401
Intermediate Calculations
Group 1 Proportion 0.93 Confidence Interval
Group 2 Proportion 0.23 Interval Lower Limit 0.6599
Difference in Two Proportions 0.7 Interval Upper Limit 0.7401
Average Proportion 0.58
Z Test Statistic 31.71351646

Two-Tail Test
Lower Critical Value -2.5758
Upper Critical Value 2.5758
p -Value 0.0000
Reject the null hypothesis

Copyright ©2019 Pearson Education Ltd.

392 Chapter 10: Two-Sample Tests

10.32 (a) H0: π 1 = π 2 H1: π 1 ≠ π 2

cont. Population: 1 = Superbanked; 2 = Unbanked.
Decision rule: If Z STAT > 2.5758 or p-value < 0.05, reject H0.
Test statistic:
( p − p2 ) − (π 1 − π 2 ) = 31.7135
Z STAT = 1
1 1
p (1 − p )  + 
 n1 n2 
Decision: Since p-value < 0.05, reject H0. There is evidence of a difference in the
proportion of Superbanked and Unbanked with respect to the proportion that use credit
cards.
(b) p-value is virtually 0. The probability of obtaining a difference in proportions that gives
rise to a test statistic below –31.7135 or above +31.7135 is 0.0000 if there is no
difference in the proportion of Superbanked and Unbanked who use credit cards.
 p1 (1 − p1 ) p2 (1 − p2 )   0.93(1 − 0.93) 0.2391 − 0.23 
(c) ( p1 − p2 ) ± Z
 + = 0.7 ± 2.5758  + 
 n1 n2   1000 1000 
0.6599 ≤ π 1 − π 2 ≤ 0.7401
You are 99% confident that the difference in the proportion of Superbanked and
Unbanked who use credit cards is between 0.6599 and 0.7401.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 393

10.33 (a)

At the .05 level, there is evidence that a higher percentage of B2B marketers use
LinkedIn compared to B2C marketers. The ZSTAT of 11.07 is well above the upper critical
value. Because ZSTAT = 11.07 or p-value = 0.000, reject H0.
(b) The p-value is 0.000, which indicates that the probability of obtaining an equal or larger
value than the observed test statistic, when H0 is true, is 0.000%.

Copyright ©2019 Pearson Education Ltd.

394 Chapter 10: Two-Sample Tests

10.33 (c)
cont.

At the .05 level, there is no evidence of a difference in the use of YouTube between the
B2B and the B2C marketers. The ZSTAT of –0.24 is above the lower critical value with a
p-value of 0.807. Because ZSTAT = –0.24 or p-value = 0.807, do not reject H0.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 395

10.34 (a) H0: π 1 = π 2 H1: π 1 ≠ π 2 Where Population: 1 = Co-browsing,

2 = Non-co-browsing
Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.05
Group 1
Number of Items of Interest 81
Sample Size 129
Group 2
Number of Items of Interest 65
Sample Size 176

Intermediate Calculations
Group 1 Proportion 0.627906977
Group 2 Proportion 0.369318182
Difference in Two Proportions 0.258588795
Average Proportion 0.4787
Z Test Statistic 4.4662

Two-Tail Test
Lower Critical Value -1.9600
Upper Critical Value 1.9600
p -Value 0.0000
Reject the null hypothesis
Decision rule: If p-value < 0.05, reject H0.
Decision: Since p-value = 0.0000 < 0.05, reject H0. There is sufficient evidence of a
difference between co-browsing organizations and non-co-browsing organizations in the
proportion that use skills-based routing to match the caller with the right agent.
(b) p-value is 0.0000. The probability of obtaining a difference in proportions that gives rise
to a test statistic that deviates from 0 by 4.4662 or more in either direction is 0.0000 if
there is not a difference between co-browsing organizations and non-co-browsing
organizations in the proportion that use skills-based routing to match the caller with the
right agent.

10.35

Copyright ©2019 Pearson Education Ltd.

396 Chapter 10: Two-Sample Tests

10.35 (a) At the .05 level, there is evidence that a significantly higher proportion of film and video
cont. projects were crowd funded compared to technology projects. The ZSTAT of –3.79 is
below the lower critical value. Because ZSTAT = –3.79 or p-value = 0.000, reject H0.
(b) The p-value is 0.000, which means that the probability of obtaining an equal or larger
value than the observed test statistic, when H0 is true, is less than 0.000%.
(c)

The lower limit is –0.1829 and the upper limit is –0.0567 for the mean difference in
proportion crowdfunded between the technology and film projects.

10.36 (a) 2.20

(b) 2.57
(c) 3.50

10.37 (a) α =0.05, n1 =16, n2 =21, F0.05 = 2.20

(b) α =0.01, n1 =16, n2 =21, F0.01 = 3.09

10.38 (a) Population B: S 2 = 25

(b) 1.5625

S12 161.9
10.39 FSTAT
= = = 1.2109
S 2 2 133.7

10.40 df numerator = 24, df denominator = 24,

10.41 α =0.05, n1 =25, n2 =25, F0.05/ 2 = = 2.27

10.42 FSTAT 1.2109 < 2.27, do not reject H0.

Because =

10.43 The F test for the ratio of two variances is sensitive to departures from normality. The F test
should not be used because both populations are skewed and do not meet the assumption of
normality. The Levene test or a nonparametric test should be used in this situation.

10.44 (a) FSTAT 1.2995 < 3.18, do not reject H0.

Because =
(b) FSTAT 1.2995 < 2.62, do not reject H0.
Because =

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 397

10.45 (a) H0: σ old

2
= σ new
2

H1: σ old
2
≠ σ new
2

old new
Sample mean 2.620 4.477
Sample variance 1.897 1.292
Sample size 15 13
Degrees of Freedom 14 12
F Test statistic 1.469
p-value 0.255
Upper critical value 2.637

The F statistic is less than the upper critical value, so do not reject the null hypothesis.
There is not enough evidence of a difference in the variability of GDP growth rates.
(b) The p-value is 0.255, which is the probability of obtaining a test statistic as high or more
extreme than 1.469, given that the null hypothesis is true.
(c) We have to assume that both the populations are normal.
(d) Based on (a) and (b), since no significant difference was revealed, a pooled-variance t-
test should be used.

10.46 (a) H0: σ POL

2
= σ HUN
2

H1: σ POL
2
≠ σ HUN
2

POL HUN
Sample mean 4.060 5.018
Sample variance 1.037 1.027
Sample size 50 50
Total Degrees of Freedom 49 49
F Test statistic 1.010
p-value 0.486
Upper critical value 1.762

The p-value is large so we do not reject the null hypothesis. There is no evidence of a
difference in the variation of the two populations.
(b) The p-value in (a) is 0.486, which is the probability of obtaining a test statistic equal to or
more extreme than 1.010, given that the null hypothesis is true.
(c) Both populations need to be normal.
(d) One should use the pooled-variance t-test.

Copyright ©2019 Pearson Education Ltd.

398 Chapter 10: Two-Sample Tests

10.47 (a) H0: σ O2 = σ Y2

H1: σ O2 ≠ σ Y2

BBS_Older BBS_Younger
Sample mean 37.65 47.5
Sample variance 69.71315789 35
Sample size 20 20
Total Degrees of
Freedom 19 19
F Test statistic 1.991804511
p-value 0.142098744
Upper critical value 2.526450934

The F statistic is less than the upper critical value, so do not reject the null hypothesis.
There is not enough evidence of a difference in the variability of the BBS values.
(b) The p-value is 0.1421, which is the probability of obtaining a test statistic equal to or
more extreme than 1.9918, given that the null hypothesis is true.
(c) The test assumes that the two populations are both normally distributed

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 399

10.47 (c)
cont.

We cannot assume that the datasets are normally distributed.

(d) Based on (a), (b) and (c), it is appropriate to use a pooled-variance t-test to compare the
means of the two age groups.

Copyright ©2019 Pearson Education Ltd.

400 Chapter 10: Two-Sample Tests

10.48 (da) H0: σ nc2 = σ c2

H1: σ nc2 ≠ σ c2

non-captain captain
Sample mean 27.689 32.406
Sample variance 14.593 12.636
Sample size 704 32
Total Degrees of Freedom 703 31
F Test statistic 1.155
p-value 0.322
Upper critical value 1.782

The F statistic is less than the upper critical value, so do not reject the null hypothesis.
There is not enough evidence of a difference in the variability of ages.
(db) The p-value is 0.322.Iit is the probability of obtaining a test statistic equal to or more
extreme than 1.155, given that the null hypothesis is true.
(dc) We have to assume that both populations are normal. Based on the sample, the ages are
fairly close to a normal distribution in this case.
(dd) Based on (a) and (b), a pooled-variance t-test should be used

10.49

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 401

10.49
cont.

(a) At the .05 level, there is evidence that there is a difference between the variances in the
time spent accessing the Internet between the males and females. The FSTAT of 12.47 is
well above the critical value. Because FSTAT = 12.4665 or p-value = 0.0000, reject H0.
(b) On the basis of the results in (a), the separate variance t test would be the appropriate
choice for these data.

Copyright ©2019 Pearson Education Ltd.

402 Chapter 10: Two-Sample Tests

10.50 (a) H0: σ T2 = σ U2

H1: σ T2 ≠ σ U2

Taxi Uber

Sample mean 7.423 4.568

Sample variance 3.741 1.345
Sample size 40 40
Total Degrees of Freedom 39 39
F Test statistic 2.782
p-value 0.001
Upper critical value 1.891

The p-value is small, so we reject the null hypothesis. There is evidence of a difference
between variances of the two populations.
(b) Both populations need to be normally distributed.
(c) From the boxplot and the normal probability plots, the delay times appear to be
approximately normally distributed.
(d) As there is a difference between the variances, one should use the separate-variance t-
test.

10.51 Box plot, or scatter graph

10.52 The pooled variance t-test should be used when the populations are approximately normally
distributed and the variances of the two populations are assumed equal.

10.53 The F test can be used to examine differences in two variances when each of the two populations
is assumed to be normally distributed.

10.54 No, the sample size can be different.

10.55 Repeated measurements represent two measurements on the same items or individuals, while
paired measurements involve matching items according to a characteristic of interest.

10.56 The lecturer should use pooled variance t-test because the random samples are independently
selected from two populations.

10.57 If you cannot assume that both populations are normally distributed, you have two choices. You
can use a nonparametric procedure, such as the Wilcoxon rank sum test that does not depend on
the assumption of normality for the two populations, or you can use a normalizing transformation
on each of the outcomes and then use the pooled-variance t test.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 403

10.58

(a) At the .05 level, there is insufficient evidence that there is a statistically difference in
variances for the two types of jobs. Because FSTAT = 1.36 or p-value = 0.228, do not reject
H0.

Copyright ©2019 Pearson Education Ltd.

404 Chapter 10: Two-Sample Tests

10.58 (b) Based on the results from (a), one would fail to reject the null hypothesis and would
cont. choose the pooled variance t test.

(c) At the .05 level, there is evidence that the mean salary for Black belt jobs are
significantly higher than the mean salary for Green belt jobs. The tSTAT of 3.97 is above
the critical value. The p-value is 0.000. Because tSTAT = 3.97 or p-value = 0.000, reject H0.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 405

10.59
Europe Other
Sample mean 418.80 415.00
Sample variance 29409.29 24502.40
Sample size 10 11
Total Degrees of Freedom 9 10
F 1.20
p-value 0.39
F critical 3.78

(a) The F test statistic is lower than the critical value, so we did not find evidence of a
difference in the variances between the two locations.

Europe Other
Sample mean 418.8 415
Sample variance 29409.3 24502.4
Sample size 10 11
Pooled variance 26826.7
Hypothesized difference 0
Total Degrees of Freedom 19
t-Test statistic 0.053
Lower critical value -2.093
Upper critical value 2.093
p-value 0.958
Do not reject null hypothesis

(b)
(c)
(d) We did not find evidence that the Formula One ticket prices are different in Europe than
in the rest of the world. The test statistic is between the two critical values and the p-
value is high, so we do not reject the null hypothesis.

Based on our data, we can conclude that both the mean and standard deviation of ticket prices are
similar in Europe and the rest of the world. In both locations, one can find both relatively cheap
and expensive Grand Prix races.

Copyright ©2019 Pearson Education Ltd.

406 Chapter 10: Two-Sample Tests

10.60 (a)

Using a .01 level of significance, there is insufficient evidence that there is a difference in
the variances in time spent online between males and females. Because FSTAT = 1.36 or
p-value = 0.127, do not reject H0.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 407

10.60 (b) It is most appropriate to use the pooled-variance t test to test for differences in mean
cont. online time between males and females.

Copyright ©2019 Pearson Education Ltd.

408 Chapter 10: Two-Sample Tests

10.60 (b) At the .01 significance level, the tSTAT of –9.76 is well below the lower critical value.
cont. Because tSTAT = –9.76 or p-value = 0.000, one would reject the null hypothesis that there
is no difference between the mean online time between males and females. The results
indicate that men spent significantly more time online compared to women.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 409

10.60 (c) Using a .01 level of significance, there is evidence that there is a difference in the
cont. variances in time spent playing online games between males and females. Because
FSTAT = 1.78 or p-value = 0.005, reject H0.

(d) A separate variance t test revealed that at the .01 significance level, there is evidence that
men and women differ significantly in amount of time spent playing online games. Men
spent significantly more time playing online games than women. Because FSTAT = 26.40
or p-value = 0.000, reject H0.

Copyright ©2019 Pearson Education Ltd.

410 Chapter 10: Two-Sample Tests

10.61 • teaching quality

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 411

10.61
cont.

From the boxplots and normal probability plots, we see that the distribution of the teaching
quality rating is quite normal (although a slight left skew can be seen on the histogram) for the
international and national students at the Faculty of Business and Economics of the University of
Pécs. Hence, the results from the F test on the difference in variances to determine whether the
pooled-variance t test or separate-variance t test is more appropriate for the difference in means
are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 4.84 4.52
Sample
variance 1.606531 2.377143
Sample size 50 50
Total Degrees
of Freedom 49 49
F 0.675824
p-value 0.0869
F critical 0.567476

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

Copyright ©2019 Pearson Education Ltd.

412 Chapter 10: Two-Sample Tests

10.61 H0: µ INT = µ NAT

cont. H1: µ INT ≠ µ NAT
international national
Sample mean 4.84 4.52
Sample variance 1.606531 2.377143
Sample size 50 50
Pooled variance 1.991837
Hypothesized difference 0
Total Degrees of Freedom 98
t-test statistic 1.133687
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.259692
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean teaching quality rating between the
international and national students is different.

• offered scholarships

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 413

10.61
cont.

From the boxplots, normal probability plots and on the histogram we see that the distribution of
the offered scholarships rating is quite normal for both international and national students at this
university. Hence, the results of the F test on the difference in variances to determine whether the
pooled-variance t test or separate-variance t test is more appropriate for the difference in means
are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 4.64 4.32
Sample variance 1.990204 2.303673
Sample size 50 50
Degrees of
Freedom 49 49
F 0.863926
p-value 0.305339
F critical 0.567476

Copyright ©2019 Pearson Education Ltd.

414 Chapter 10: Two-Sample Tests

10.61
cont.

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

H0: µ INT = µ NAT

H1: µ INT ≠ µ NAT

international national
Sample mean 4.64 4.32
Sample variance 1.990204 2.303673
Sample size 50 50
Pooled variance 2.146939
Hypothesized
difference 0
Degrees of Freedom 98
t-Test statistic 1.091969
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.277524
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean offered scholarships rating between the
international and national students at this university is different.

• affordable tuition fee

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 415

10.61
cont.

Copyright ©2019 Pearson Education Ltd.

416 Chapter 10: Two-Sample Tests

10.61
cont.
From the boxplots and normal probability plots, we see that the distribution of the affordable
tuition fee rating is quite normal (although a slight left skew can be seen on the histogram) for the
international and national students at this university. Hence, the results from the F test on the
difference in variances to determine whether the pooled-variance t test or separate-variance t test
is more appropriate for the difference in means are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 5.66 4.48
Sample
variance 2.39224 2.458776
Sample size 50 50
Total Degrees
of Freedom 49 49
F 0.97294
p-value 0.46195
F critical 0.56748

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

H0: µ INT = µ NAT

H1: µ INT ≠ µ NAT

international national
Sample mean 5.66 4.48
Sample variance 2.39224 2.458776
Sample size 50 50
Pooled variance 2.42551
Hypothesized
difference 0
of Freedom 98
t-Test statistic 3.78835
Lower critical value -1.9845
Upper critical value 1.98447
p-value 0.00026
Do reject null hypothesis

Since the p-value < 0.05, do reject H0.

There is enough evidence to conclude that the mean affordable tuition fee rating between the
international and national students at this university is different.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 417

10.61
cont.
• affordable housing

Copyright ©2019 Pearson Education Ltd.

418 Chapter 10: Two-Sample Tests

10.61
cont.

From the boxplots and normal probability plots, we see that the distribution of the affordable
housing rating is quite normal (although a slight left skew can be seen on the histogram) for the
international and national students at this university. Hence, the results from the F test on the
difference in variances to determine whether the pooled-variance t test or separate-variance t test
is more appropriate for the difference in means are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 4.98 4.76
Sample
variance 2.958776 2.308571
Sample size 50 50
Total Degrees
of Freedom 49 49
F 1.281648
p-value 0.19407
F critical 1.762189

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 419

10.61
cont.
H0: µ INT = µ NAT
H1: µ INT ≠ µ NAT

international national
Sample mean 4.98 4.76
Sample variance 2.958776 2.308571
Sample size 50 50
Pooled variance 2.633673
Hypothesized
difference 0
of Freedom 98
t-Test statistic 0.677816
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.499486
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean affordable housing rating between the
international and national students at this university is different.

• careers service

Copyright ©2019 Pearson Education Ltd.

420 Chapter 10: Two-Sample Tests

10.61
cont.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 421

10.61
cont.
From the boxplots, normal probability plots, and on the histogram we see that the distribution of
the careers service rating was quite normal for the international and national students at this
university. Hence, the results from the F test on the difference in variances to determine whether
the pooled-variance t test or separate-variance t test is more appropriate for the difference in
means are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 4.56 4.48
Sample
variance 2.537143 2.989388
Sample size 50 50
Degrees of
Freedom 49 49
F 0.848717
p-value 0.284045
F critical 0.567476

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

H0: µ INT = µ NAT

H1: µ INT ≠ µ NAT

international national
Sample mean 4.56 4.48
Sample variance 2.537143 2.989388
Sample size 50 50
Pooled variance 2.763265
Hypothesized
difference 0
Degrees of Freedom 98
t-Test statistic 0.240629
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.810345
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean careers service rating between the
international and national students at this university is different.

Copyright ©2019 Pearson Education Ltd.

422 Chapter 10: Two-Sample Tests

10.61
cont.
• graduate employment

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 423

10.61
cont.

From the boxplots, normal probability plots, and on the histogram we see that the distribution of
the graduate employment rating is quite normal for the international and national students at this
university. Hence, the results from the F test on the difference in variances to determine whether
the pooled-variance t test or separate-variance t test is more appropriate for the difference in
means are reliable.

H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 4.6 4.98
Sample
variance 2.612245 2.02
Sample size 50 50
Degrees of
Freedom 49 49
F 1.293191
p-value 0.185636
F critical 1.762189

Since the p-value > 0.05, do not reject H0. At the 5% level of significance, there is insufficient
evidence to conclude that the two variances are not the same. Hence, a pooled-variance t test is
more appropriate.

Copyright ©2019 Pearson Education Ltd.

424 Chapter 10: Two-Sample Tests

10.61
cont.
H0: µ INT = µ NAT
H1: µ INT ≠ µ NAT

international national
Sample mean 4.6 4.98
Sample variance 2.612245 2.02
Sample size 50 50
Pooled variance 2.316122
Hypothesized
difference 0
Degrees of Freedom 98
t-Test statistic -1.24845
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.214839
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean graduate employment rating between the
international and national students at this university is different.

• study result

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 425

10.61
cont.

From the boxplots, normal probability plots, and on the histogram, we see that the distribution of
study results is quite normal for the international and national students at this university. Hence,
the results from the F test on the difference in variances to determine whether the pooled-variance
t test or separate-variance t test is more appropriate for the difference in means are reliable.

Copyright ©2019 Pearson Education Ltd.

426 Chapter 10: Two-Sample Tests

10.61
cont.
H0: σ INT
2
= σ NAT
2

H1: σ INT
2
≠ σ NAT
2

international national
Sample mean 3.326 3.404
Sample
variance 0.794208 0.461616
Sample size 50 50
Degrees of
Freedom 49 49
F 1.720494
p-value 0.030181
F critical 1.762189

Since the p-value > 0.05 (the built-in MS Excel package performs a one-sided test), do not reject
H0. At the 5% level of significance, there is insufficient evidence to conclude that the two
variances are not the same. Hence, a pooled-variance t test is more appropriate.

H0: µ INT = µ NAT

H1: µ INT ≠ µ NAT

international national
Sample mean 3.326 3.404
Sample variance 0.794208 0.461616
Sample size 50 50
Pooled variance 0.627912
Hypothesized
difference 0
Degrees of Freedom 98
t-Test statistic -0.49217
Lower critical value -1.98447
Upper critical value 1.984467
p-value 0.6237
Do not reject null hypothesis

Since the p-value > 0.05, do not reject H0.

There is not enough evidence to conclude that the mean study result between the international and
national students at this university is different.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 427

10.62 (a) H0: µ ≤ 10 minutes. Introductory computer students required no more than a mean of 10
minutes to write and run a program in VB.NET.
H1: µ > 10 minutes. Introductory computer students required more than a mean of 10
minutes to write and run a program in VB.NET.
Decision rule: d.f. = 8. If tSTAT > 1.8595, reject H0.
X – µ 12 – 10
Test statistic:= tSTAT = = 3.3282
S 1.8028
n 9
Decision: Since tSTAT = 3.3282 is greater than the critical bound of 1.8595, reject H0.
There is enough evidence to conclude that the introductory computer students required
more than a mean of 10 minutes to write and run a program in VB.NET.
(b) H0: µ ≤ 10 minutes. Introductory computer students required no more than a mean of 10
minutes to write and run a program in VB.NET.
H1: µ > 10 minutes. Introductory computer students required more than a mean of 10
minutes to write and run a program in VB.NET.
Decision rule: d.f. = 8. If tSTAT > 1.8595, reject H0.
X – µ 16 – 10
Test statistic:= tSTAT = = 1.3636
S 13.2004
n 9
Decision: Since tSTAT = 1.3636 is less than the critical bound of 1.8595, do not reject H0.
There is not enough evidence to conclude that the introductory computer students
required more than a mean of 10 minutes to write and run a program in VB.NET.
(c) Although the mean time necessary to complete the assignment increased from 12 to 16
minutes as a result of the increase in one data value, the standard deviation went from 1.8
to 13.2, which in turn brought the t-value down because of the increased denominator.
(d) H0: σ IC 2 = σ CS 2 H1: σ IC 2 ≠ σ CS 2
Decision rule: If FSTAT > 3.8549, reject H0.
SCS 2 2.02
Test statistic: F= STAT = = 1.2307
S IC 2 1.80282
Decision: Since FSTAT = 1.2307 is lower than the critical bound 3.8549, do not reject H0.
There is not enough evidence to conclude that the population variances are different for
the Introduction to Computers students and computer majors. Hence, the pooled variance
t test is a valid test to see whether computer majors can write a VB.NET program (on
average) in less time than introductory students, assuming that the distributions of the
time needed to write a VB.NET program for both the Introduction to Computers students
and the computer majors are approximately normal.
(d) H0: µ IC ≤ µCS The mean amount of time needed by Introduction to Computers students
is not greater than the mean amount of time needed by computer majors.
H1: µ IC > µCS The mean amount of time needed by Introduction to Computers students
is greater than the mean amount of time needed by computer majors.

Copyright ©2019 Pearson Education Ltd.

 428 Chapter 10: Two-Sample Tests

10.62 (d) PHStat output:

cont.
Pooled-Variance t Test for the Difference Between Two
Means
(assumes equal population variances)
Data
Hypothesized Difference 0
Level of Significance 0.05
Population 1 Sample
Sample Size 9
Sample Mean 12
Sample Standard Deviation 1.802776
Population 2 Sample
Sample Size 11
Sample Mean 8.5
Sample Standard Deviation 2

Intermediate Calculations
Population 1 Sample Degrees of Freedom 8
Population 2 Sample Degrees of Freedom 10
Total Degrees of Freedom 18
Pooled Variance 3.666667
Difference in Sample Means 3.5
t Test Statistic 4.066633

Upper-Tail Test
Upper Critical Value 1.734064
p-Value 0.000362
Reject the null hypothesis
Decision rule: d.f. = 18. If tSTAT > 1.7341, reject H0.
Test statistic:
(nIC − 1) ⋅ S IC 2 + (nCS − 1) ⋅ SCS 2 9 ⋅ 1.80282 + 11 ⋅ 2.02
=S p2 = = 3.6667
(nIC − 1) + (nCS − 1) 8 + 10

=tSTAT
(=
X − X ) − (µ − µ )
IC CS IC CS 12.0 − 8.5
= 4.0666
 1 1  1 1 
2
Sp  +  3.6667  + 
 nIC nCS   9 11 
Decision: Since tSTAT = 4.0666 is greater than 1.7341, reject H0. There is enough evidence
to support a conclusion that the mean time is higher for Introduction to Computers
students than for computer majors.
(e) p-value = 0.0052. If the true population mean amount of time needed for Introduction to
Computer students to write a VB.NET program is indeed no more than 10 minutes, the
probability for observing a sample mean greater than the 12 minutes in the current sample
is 0.0052, which means it will be a quite unlikely event. Hence, at a 95% level of
confidence, you can conclude that the population mean amount of time needed for
Introduction to Computer students to write a VB.NET program is more than 10 minutes.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 429

10.62 (e) As illustrated in part (d) in which there is not enough evidence to conclude that the
cont. population variances are different for the Introduction to Computers students and
computer majors, the pooled variance t test performed is a valid test to determine whether
computer majors can write a VB.NET program in less time than in introductory students,
assuming that the distributions of the time needed to write a VB.NET program for both
the Introduction to Computers students and the computer majors are approximately
normal.

10.63 (a)

A F test for the ratio of two variances revealed a significant difference between the
variances of the social shoppers and the other online shoppers. This difference was
significant at the .01 significance level.

Copyright ©2019 Pearson Education Ltd.

430 Chapter 10: Two-Sample Tests

10.63 (b)
cont.

A separate-variance t test revealed a significance difference at the .05 level between the
mean amount spent by social and other online shoppers. Social shoppers spent
significantly more than other online shoppers. Because tSTAT = 5.73 or p-value = 0.000,
one would reject the null hypothesis that there is no significance difference in mean spent
between social and other online shoppers. The Minitab output did not include the exact
p-value due to the use of only decimal places.
(c) The confidence interval for the mean difference in amount spent is
6.95 ≤ µ1 − µ2 ≤ 14.19.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 431

10.64

Copyright ©2019 Pearson Education Ltd.

432 Chapter 10: Two-Sample Tests

10.64
cont.

From the box plots and normal probability plots, both sets of data seem to have rather
symmetrical distributions that are quite normally distributed.

The following F test for any evidence of difference between two population variances suggests
that there is insufficient evidence to conclude that the two population variances are significantly
different at the 5% level of significance.

Saudi Arabia United Arab Emirates

Sample mean 39513.35 43056.02
Sample variance 3.15E+08 5.44E+08
Sample size 44 44
Degrees of
Freedom 43 43
F Test statistic 0.579167
p-value 0.076807
Lower critical value 0.545649

Since both sets of data are drawn from independent populations, the most appropriate test for any
difference in the amounts of electric power consumed by the two countries’ families is the
pooled-variance t test.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 433

10.64
cont.
H0: µUSA = µUAE
H1: µUSA ≠ µUAE

Saudi Arabia United Arab Emirates

Sample mean 39513.35455 43056.01818
Sample variance 314928609.1 543761151.4
Sample size 44 44
Pooled variance 429344880.3
Hypothesized
difference 0
Degrees of Freedom 86
-
t-Test statistic 0.801933056
-
Lower critical value 1.987934206
Upper critical value 1.987934206
p-value 0.424802057
Do not reject the null hypothesis

Since the p-value > 0.05, at the 5% level of significance, we do not have enough evidence to
reject the null hypothesis of no difference in the amounts of electric power consumed by the two
countries. We can conclude that there is no evidence of significant difference in the amounts.

10.65 Population 1 = Wing A, 2 = Wing B

H0: σ 12 = σ 22 The population variances are the same.
H1: σ 12 ≠ σ 22 The population variances are different.
Decision rule: If FSTAT > 2.5265, reject H0.
(1.4172 ) = 1.0701
2
S12
Test statistic: FSTAT
= =
(1.3700 )
2
S2 2
Decision: Since FSTAT = 1.0701 is lower than the critical bound of Fα / 2 = 2.5265, do not reject H0.
There is not enough evidence to conclude that there is a difference between the variances in Wing
A and Wing B. Hence, a pooled-variance t test is more appropriate for determining whether there
is a difference in the mean delivery time in the two wings of the hotel.
H0: µ1 = µ2 H1: µ1 ≠ µ2
Decision rule: d.f. = 38. If tSTAT < – 2.0244 or tSTAT > 2.0244, reject H0.
Test statistic:
( n − 1) S12 + ( n2 − 1) S2 2 = ( 20 − 1)(1.3700 ) + ( 20 − 1)(1.4172 ) = 1.9427
2 2

S p2 = 1
( n1 − 1) + ( n2 − 1) ( 20 − 1) + ( 20 − 1)
tSTAT =
(X 1 − X 2 ) − ( µ1 − µ2 )
=
(10.40-8.12 ) − 0
= 5.1615
1 1  1 1 
Sp  +  2
1.9427  + 
 n1 n2   20 20 
Decision: Since tSTAT = 5.1615 is greater than the upper critical bound of 2.0244, reject H0. There
is enough evidence of a difference in the mean delivery time in the two wings of the hotel.

Copyright ©2019 Pearson Education Ltd.

434 Chapter 10: Two-Sample Tests

10.66 H0: π 1 = π 2 H1: π 1 ≠ π 2 where Populations: 1 = Males, 2 = Females

Decision rule: If p-value < 0.05, reject H0.

Gender:
PHStat output:
Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.05
Group 1
Number of Items of Interest 50
Sample Size 300
Group 2
Number of Items of Interest 96
Sample Size 330

Intermediate Calculations
Group 1 Proportion 0.166666667
Group 2 Proportion 0.290909091
Difference in Two Proportions -0.12424242
Average Proportion 0.2317
Z Test Statistic -3.6911

Two-Tail Test
Lower Critical Value -1.9600
Upper Critical Value 1.9600
p -Value 0.0002
Reject the null hypothesis
Decision: Since the p-value is smaller than 0.05, reject H0. There is enough evidence of a
difference between males and females in the proportion who order dessert.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 435

10.66 Beef Entrée:

cont. PHStat output:
Z Test for Differences in Two Proportions

Data
Hypothesized Difference 0
Level of Significance 0.05
Group 1
Number of Items of Interest 74
Sample Size 197
Group 2
Number of Items of Interest 68
Sample Size 433

Intermediate Calculations
Group 1 Proportion 0.375634518
Group 2 Proportion 0.15704388
Difference in Two Proportions 0.218590638
Average Proportion 0.2254
Z Test Statistic 6.0873

Two-Tail Test
Lower Critical Value -1.9600
Upper Critical Value 1.9600
p -Value 0.0000
Reject the null hypothesis
Decision: Since the p-value = 0.0000 is smaller than 0.05, reject H0. There is enough evidence of a
difference in the proportion who order dessert based on whether a beef entrée has been ordered.

10.67
Normal Probability Plot
3900

3850

3800
Vermont

3750

3700

3650

3600

3550
-4 -3 -2 -1 0 1 2 3 4
Z Value

Copyright ©2019 Pearson Education Ltd.

436 Chapter 10: Two-Sample Tests

10.67
cont.
Normal Probability Plot
3300

3250

3200
Boston

3150

3100

3050

3000
-4 -3 -2 -1 0 1 2 3 4
Z Value

The normal probability plots suggest that the two populations are not normally distributed so an F
test is inappropriate for testing the difference in two variances. The sample variances for Boston
and Vermont shingles are 1204.992 and 2185.032, respectively. It appears that a separate
variance t test is more appropriate for testing the difference in means.
H 0 : µ B = µV Mean weights of Boston and Vermont shingles are the same.
H1 : µ B ≠ µV Mean weights of Boston and Vermont shingles are different.
Separate-Variances t Test for the Difference Between Two Means
(assumes unequal population variances)
Data
Hypothesized Difference 0
Level of Significance 0.05
Population 1 Sample
Sample Size 368
Sample Mean 3124.214674
Sample Standard Deviation 34.71299377
Population 2 Sample
Sample Size 330
Sample Mean 3704.042424
Sample Standard Deviation 46.74432189

Intermediate Calculations
Numerator of Degrees of Freedom 97.9257
Denominator of Degrees of Freedom 0.1625
Total Degrees of Freedom 602.7216
Degrees of Freedom 602
Separate Variance Denominator 3.1457
Difference in Sample Means –579.8277503
Separate-Variance t Test Statistic –184.3210

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 437

10.67
cont.
Two-Tail Test
Lower Critical Value –1.9639
Upper Critical Value 1.9639
p-Value 0.0000
Reject the null hypothesis
Since the p-value is essentially zero, reject H 0 . There is sufficient evidence to conclude that the
mean weights of Boston and Vermont shingles are different.

10.68 The normal probability plots suggest that the two populations are not normally distributed.
An F test is inappropriate for testing the difference in the two variances. The sample variances for
Boston and Vermont shingles are 0.0203 and 0.015, respectively. Because tSTAT = 3.015 > 1.967
or p-value = 0.0028 < α = 0.05, reject H0. There is sufficient evidence to conclude that there is a
difference in the mean granule loss of Boston and Vermont shingles.

10.69 Because an F test for the ratio of two variances revealed a significant difference at the .05 level
between the variances of the two types of smartphone batteries, a separate-variance t test is
appropriate for these data.

Copyright ©2019 Pearson Education Ltd.

438 Chapter 10: Two-Sample Tests

10.69 Because FSTAT = 5.44 or p-value = 0.000, reject H0.

cont.

The mean battery life of the existing battery design was significantly longer than the life of the
new battery design. Because tSTAT = 33.80 or p-value = 0.000, one would reject the null hypothesis
that the mean battery life is the same for the two types of battery designs. The existing design
would be preferred. The Minitab output did not include the exact p-value due to the use of only
decimal places.

10.70 An analysis of the data in 10.67 revealed that a pallet of Vermont shingles weighed significantly
more than a pallet of Boston shingles. Assuming that a heavier weight shingle is associated with
higher quality, the Vermont shingle might be perceived as a higher quality shingle compared to
the Boston shingle. The below figure shows the results from a separate variance t-test.

Copyright ©2019 Pearson Education Ltd.

 Solutions to End-of-Section and Chapter Review Problems 439

10.70
cont.

An analysis of the data in 10.68 revealed that the Vermont shingles were associated with less
granule loss compared to the Boston shingles following accelerated-life testing. Shingles with
less weight loss are assumed to have a longer life expectancy. Both shingles would be expected to
outperform the length of the warranty period because their weight losses were well below the .8
gram threshold. However, the Vermont shingles would be expected to have a longer life
expectancy given that they loss less weight relative to the Boston shingles. The below figure
shows the results from a pooled-variance t-test.

Taken together, the results from 10.67 and 10.68 suggest that the Vermont shingle may be a
higher quality shingle based on pallet weight and life expectancy as determined by granule loss
associated with accelerated-life testing. These conclusions suggest that the manufacturer may be
able to charge more for the Vermont shingle compared to the Boston shingle.

Copyright ©2019 Pearson Education Ltd.