Chapter 09 - Hypothesis Testing

Chapter 9 Hypothesis Testing

Solutions
1.
a. The alternative hypothesis H A needs to be the exact opposite of what is
stated in the null hypothesis H 0 . Here H A should not have the equality
sign.
b. The null hypothesis H 0 regarding the population mean  must include
some form of the equality sign. The alternative hypothesis H A should not
have the equality sign.

c. The hypothesized value should be the same under the null and alternative
hypotheses; here, 0.40 ≠ 0.42.

d. Hypothesis testing is about a population parameter (for instance,  or p )

and not a sample statistic .

9-1
Chapter 09 - Hypothesis Testing

2.
a. Invalid. The population parameter  should be used instead of the sample
statistic .

b. Valid.
c. Valid.
d. Invalid. The null hypothesis H 0 regarding the population mean  must
include some form of the equality sign.

9-2
Chapter 09 - Hypothesis Testing

e. For a given sample, a reduction in the probability of Type I error increases

the probability of Type II error, and vice versa.
f. If I have already decided on the Type I error, a bigger sample size will only
reduce the Type II error.
g. For a given sample, if I make it difficult to reject the null hypothesis, it will
reduce the Type I error and, therefore, increase the Type II error.
h. For a given sample, if I make it easy to reject the null hypothesis, it will
increase the Type I error
3.
a. Incorrect. We never accept the null hypothesis.
b. Correct. The null hypothesis is rejected whenever it is not supported by the
sample data.
c. Incorrect. We cannot establish a claim if the null hypothesis is not rejected.
d. Correct.

4.
a.
b.
c.

5.
a. The consequence of a Type I error is to conclude that the mean weight is
different from 18 ounces, when it is equal to 18 ounces. Type II error is to
fail to reject that the mean weight is 18 ounces, when the mean weight is not
18 ounces.
b. A Type I error is to conclude that the stock price increases on more than
60% of the trading days, when it does not. A Type II error is not to conclude

that the stock price increases on more than 60% of the trading days when it `
does.
c. A Type I error is to conclude that Americans sleep for less than 7 hours a day,
when they actually sleep for at least 7 hours. A Type II error is not to
conclude that Americans sleep for less than 7 hours a day, when they do.

9-3
Chapter 09 - Hypothesis Testing

d.
e.
f.

6.
a. A Type I error is to conclude that the majority of voters support the
candidate, when they do not. Type II error is not to conclude that the
candidate has the majority when the candidate does.
b. A Type I error is to conclude that the average pizza is less 10 inches, when it
is at least 10 inches. Type II error is not to conclude that the pizza is less
than 10 inches when it is.

c. A Type I error is to conclude that the tablet does not contain 250mg of
ibuprofen when it does. A Type II error is not to conclude that the tablet
does not contain 250mg of ibuprofen when it does not equal 250mg.

7. Here the null hypothesis suggests that the person is telling the truth and the
alternative hypothesis suggests that the person is not telling the truth.
a. Type I error, 0.005.
b. Type II error, 1 ‒ 0.95 = 0.05.

9-4
Chapter 09 - Hypothesis Testing

c. The consequence of a Type I error is to conclude that the individual is telling

a lie, when the person is actually telling the truth. A Type II error is to
conclude that the individual is telling the truth, when the person is actually
lying.
d. If we do not reject the null hypothesis, we do not prove that the individual is
telling the truth.

8. Here the null hypothesis suggests that the person is free of disease and the
alternative hypothesis suggests that the person has the disease.
a. Type I error,
b. Type II error, .

c. The consequence of Type I error is to conclude that the individual has the
disease, when the person does not have it. A Type II error is to conclude that
the individual does not have the disease whereas the person does have it.
d. If we do not reject the null hypothesis, we do not prove that the individual is
free of disease.
9.
a. With n = 25, The value of the test statistics,

p-value

b. We do not reject H 0 because the p-value .

c. The value of the test statistics, the p-value

d. We reject H 0 because the p-value

10.

With n = 25, the value of the test statistic, We

do not reject H 0 because

With n = 100, the . Here, we reject H 0 because

.

9-5
Chapter 09 - Hypothesis Testing

11.
a. The decision rule is to
reject H 0

b. The value of the test statistic,

c. We reject H 0 because

12.

the p-value
H 0 because the p-value If we reject H 0
because the p-value .

13.

a. p-value
.

b. p-value
.

c. p-value
.

d. p-value
.

14.
a. , the critical values are ―1.96 and 1.96.
The decision rule is to reject H 0 if

b. The value of the test statistic, Since z = 1.84

falls between ―1.96 and 1.96, we do not reject H 0 .

9-6
Chapter 09 - Hypothesis Testing

c. , the critical values are ―1.645 and 1.645.

The decision rule is to reject H 0 if .
d. The value of the test statistic,

, we reject H 0 .

15.
a.

b. The value of the test statistic, The p-

value .
c. Since the p-value . Customers at this
store spend more than $130.
d.
We reach the same conclusion as
in part c.

16.
a.

b. The value of the test statistic, . The p-value

c. . The braking distance in the

sample data is not significantly different from 120 miles.
d.

Since z = ―1.64 falls between ―2.575 and 2.575, . We

reach the same conclusion as in c.

17.
a.

The value of the test statistic, The p-value =

. Since the p-value
The researcher’s claim is
not supported by the sample data.
b. We reach
the same conclusion as in a.

18.
a.

9-7
Chapter 09 - Hypothesis Testing

b. The value of the test statistic, The p-value

c. Since the p-value The manager’s

claim is not supported by the sample data.
d. ; the decision rule is
to reject . We
reach the same conclusion as in c.

19.
a.

b. The value of the test statistic, .

The decision rule is to reject if

c. Based on the sample data, the average content of bottles is not significantly
different from 16 ounces at the 5% significance level. Thus, the accuracy of
the bottling process is not compromised.

20.
a.
b. ; the p-value

c. We do not reject since the p-value . At the 5%

significance level, we cannot conclude that the average price in Mission
Viejo is more than $500 thousand.

21.
a.
b. the p-value
.
c. Since the p-value At the 5%
significance level, we conclude that that the average hourly wage is less than
$22.

22.
a.
b. the p-value

9-8
Chapter 09 - Hypothesis Testing

c. Since the p-value At the 5%

significance level, the average weekly stock price of Home Depot is
significantly different from $30.

23.
a. The p-
value

b. the p-value

c. the p-value

d. the p-value

24.
a. 0.05 < p-value < 0.10. At , we do not reject since the p-value
At , we reject since the p-value
b. 0.01 < p-value < 0.025. At , we do not reject since the p-value
At , we reject since the p-value
c. 0.01 < p-value < 0.025. At , we do not reject since the p-value
At , we reject since the p-value
d. 0.05 < p-value < 0.10. At , we do not reject since the p-value
At , we reject since the p-value
25.

a. p-value
2  0.025 < p-value < 2  0.05; 0.05 < p-value < 0.10.

b. p-value
2  0.025 < p-value < 2  0.05; 0.05 < p-value < 0.10.

c. p-value
2  0.01 < p-value < 2  0.025; 0.02 < p-value < 0.05.

9-9
Chapter 09 - Hypothesis Testing

d. p-value
2  0.005 < p-value < 2  0.01; 0.01 < p-value < 0.02.

26.
a. 0.05 < p-value < 0.10; At , we do not reject since the p-value
At , we reject since the p-value
b. 0.05 < p-value < 0.10; At , we do not reject since the p-value
At , we reject since the p-value
c. 0.02 < p-value < 0.05; At , we do not reject since the p-value
At , we reject since the p-value
d. 0.01 < p-value < 0.02; At , we do not reject since the p-value
At , we reject since the p-value

27.
a. .
b. the critical values are ―2.898 and 2.898.
c. .
d.

28.

a. p-value 0.025 <

p-value < 0.05. Since the p-value

b. p-value 0.05 <

p-value < 0.10. Since the p-value >

29.
a.
, we r
b.
t , we do not r

30.

9-10
Chapter 09 - Hypothesis Testing

a. p-value
. Since the p-value
The sample data suggest that the population
mean is different from 16.

b.
same conclusion as in a.

31.

p-value
Since the p-value

32. ;

, we do not r

33.
a.

b. .
c. p-value 0.05 < p-value < 0.10. Since the p-value
The sample data do not suggest that the
machine is working improperly.
d. ; the critical values are ―2.030 and 2.030. The
decision rule is to reject if
. Same conclusion as in c.

9-11
Chapter 09 - Hypothesis Testing

34.
a.

b. .
It is necessary to assume that the population is normally distributed given
that the small sample size of 7 observations is less than 30.
c. . The
average waiting time in the sample data is not significantly more than 5
minutes at the 10% level. There is no need to hire an additional employee.
d. p-value ; since the p-value
Same conclusion as in c.

35.
a. Given a sample size of only 6 observations, it is necessary to assume that the
price of small computers is normally distributed.
b.

c. .
d.

The sample evidence

does not support the reporter’s claim at the 5% significance level. There
seems to be no cause for concern for larger computer companies.

36.
a.
b. It is not necessary to assume that the underlying population is normally
distributed. The sample size is larger than 30, so the sample mean is
normally distributed according to Central Limit Theorem.
c.
d.
The bottling process has not fallen out of adjustment.

37. We will use the p-value approach.

p-value
0.05 < p-value < 0.10. Since the p-value
We cannot conclude at the 5% level that the mean forecast of all NABE
members is greater than 3%.

9-12
Chapter 09 - Hypothesis Testing

38. We will use the p-value approach.

Based on the performance of 12 cars, we use and df = 11

to compute p-value
0.005 < p-value < 0.01. Since the p-value , At the
5% level of significance, we can conclude that the average clock time of all
cars is less than 6 seconds.
39.

a. The p-value 0.025 < p-value < 0.05. Since the p-value
, The mean mortgage rate for the population is
higher than 4.2%.
b.
We reach the same conclusion as in part a.

40. We will conduct the test with the p-value approach.

p-value
0.001 < p-value < 0.005. Since the p-value ,
The sample data indicate that the mean drop of all home
prices in San Diego is higher than the 7%.

41.
a.
b. Given the small sample size n = 7 is less than 30, it is necessary to assume
that the population from which the sample is taken is normally distributed

c.
We use Excel to compute the p-value as .
d. Since the p-value The P/E ratio in
the footwear industry does not differ from the overall average of 14.9%, at
the 10% level of significance.

9-13
Chapter 09 - Hypothesis Testing

42.
a.

b.
We use Excel to compute the p-value as
c. Since the p-value The MPG in the
sample data is not significantly different from 95.
43.
a.
b. Given the small sample size (n= 26), it is necessary to assume that the
population from which the sample is taken is normally distributed.

c. .
d. We use Excel to compute the p-value as
e. Since the p-value . The average
monthly debt payments are greater than $900 at the 5% significance level.

44.

a. the p-
value
.

b. ; the
p-value
.

c. the p-value
.

d. the p-value
.

9-14
Chapter 09 - Hypothesis Testing

45.
a. the p-
value the p-value

b. Since the p-value = 0.3594 is greater than

c. Since the p-value = 0.2810 is greater than

d. the p-value
the p-value

46.

a. the p-
value

b. he p-
value .

c. the p-value
.

d. the p-value
.

47.
a. Since the p-value = 0.7642 is greater than

b. Since the p-value = 0.394 is less than

c. Since the p-value = 0.4180 is greater than

d. the p-value
the p-value

9-15
Chapter 09 - Hypothesis Testing

48.
a.
b.

c.

49.

a.
Since the p-value

b.
Since the p-value

50.
a.

b.

9-16
Chapter 09 - Hypothesis Testing

The decision rule is

51.

the p-
value

Since the p-value ,

we do not reject At the 1% significance level, the population proportion
is not different from 0.70.

52.

The p-value

At the 5% significance level, we cannot conclude that more than 50% of the
observations in a population are below 10.

53.
a.

b. the
p-value

9-17
Chapter 09 - Hypothesis Testing

c. Since the p-value , we reject the . The law is

effective at reducing the proportion of teens who talk on their cell phones
while driving at the 5% significance level.
d. With , the critical value, The decision rule is to
reject We
reach the same conclusion as in part c.

54.
a.

b.
The p-value
c. Since the p-value , we reject . At the 5% significance
level, the economist’s concern is supported by the sample data.

55.

The bartender uses correct proportions only in 10 out of 30, so the incorrect
proportions are used in 20 out of 30.

With
, the critical
value,
The manager’s suspicion is supported by the sample data; at the 5%
significance level, the bartender does not use the correct proportions in more
than 50% of margaritas.

9-18
Chapter 09 - Hypothesis Testing

56.
a.

The p-value = Since the p-value =

0.1093 > 0.05 =
b. Since the p-value , we do not reject .
c. Based on the sample data, we at both the 5% and 10%
significance levels. The production company’s expectation of more than 30%
of viewers returning to the theaters for the same movie is not supported by the
sample data.

57.
a.

The p-value Since the p-value = 0.0367

. The sample data indicate that older Asian workers
feel less fulfilled than their British counterparts.
b.

9-19
Chapter 09 - Hypothesis Testing

The p-value Since the p-value =

0.0102 . The data shows that younger workers in
Asia feel more fulfilled than their British counterparts.

58.

The p-value Since the p-value =

0.1038 . At the 5% significance level, the
politician’s claims are not supported by the data.

59.

9-20
Chapter 09 - Hypothesis Testing

The p-value Since the p-value =

0.1562 . The sample evidence does not
support the claim that more than 75% of financial institutions that offer
online banking facilities are prone to fraud.

60.

The p-value Since the p-value =

0.1660 . The sample evidence does not
support the claim that more than 60% of seniors have made serious
adjustments to their lifestyle.

61.
a.

b. .
c.

9-21
Chapter 09 - Hypothesis Testing

The sample evidence supports the

manufacturer’s claim that the battery life is more than 10 hours.
d. The p-value 0.025 < p-value < 0.05. Since the p-value
We reach the same conclusion as in c.

62.
a.

b. .

c.

d. The claim by the weight

loss clinic is not supported by the sample data.

63.

9-22
Chapter 09 - Hypothesis Testing

At the 1% significance
level, we cannot conclude that drivers’ waiting time exceeds 20 minutes.
Thus, there is no need for the city council to spend additional money to
reduce the amount of traffic.

64.
a.

b.

c.

d.
At the 5% significance level, we cannot conclude that there is a change in the
mortgage rate from the previous week.

65.
a.
b. We use Excel to find the p-value as .
c. Since the p-value

9-23
Chapter 09 - Hypothesis Testing

d. At the 5% significance level, the sample evidence does not support that the
mean return is greater than 12%.

66.
a.

b.
c. We use Excel to find the p-value as

d. Since the p-value = we do not reject

Monthly sales do not differ from $130 thousand at the 5% significance
level.
67.
a.

b. .
c. The p-value

9-24
Chapter 09 - Hypothesis Testing

d. Since the p-value Based on the

sample evidence, there is no need for the retailer to improve its services at
the 5% significance level.
68.

The p-value
Since the p-value At the 5%
significance level, the sample evidence suggests that the percentage of
Americans who feel that the country is headed in the right direction is not
below 35%.

9-25
Chapter 09 - Hypothesis Testing

The p-value
Since the p-value At the 5%
significance level, the sample evidence suggests that the percentage of
Americans who feel that the country is headed in the right direction is below
35%.

69.
a.

b. .

c.
T
d. At the 1% significance level,
the percentage of individuals who want to watch the new show is not more
than 50%. As a result, the television network should keep its current
lineup.

70.
a.

b. .

9-26
Chapter 09 - Hypothesis Testing

The p-value
c. Since the p-value At the 5%
significance level, the sample data does support Pew Research 2010
findings. In other words, the percentage of Americans who only use cell
phones does not differ from 23%.

Case Study 9.1

1. Low Income (Family Income=$40,000)

9-27
Chapter 09 - Hypothesis Testing

At the 5% significance
level, the sample evidence does not support the spokesman’s claim that
families with an income of $40,000 spend less than $6,500 to send their
children to an elite college.

2. Medium Income (Family Income=$80,000)

At the 5% significance
level, the sample evidence does not support the spokesman’s claim that
families with an income of $80,000 spend less than $20,000 to send their
children to an elite college.

9-28
Chapter 09 - Hypothesis Testing

High Income (Family Income=$120,000)

At the 5% significance
level, the sample evidence does not support the spokesman’s claim that
families with an income of $120,000 spend less than $35,000 to send their
children to an elite college.

3. Overall, the claim by the college spokesperson is not substantiated by the

sample data.

9-29
Chapter 09 - Hypothesis Testing

Case Study 9.2

1.

Probability of Performance on Test

Best (5) Average(3 and 4) Poor(1 and 2)
(after) 0.0984 0.4016 0.5000
(before) 0.08 0.38 0.54

After the monetary incentive, the percentage of students who score 5 in the given sample is
10%, which is more than the historical average of 8%. Similarly, the percentage of students
who score 3 and 4 is 40%, which is higher than the historical average of 38%. Finally 50%
of the students score 1 and 2 in the sample, which is lower than the historical average of
54%. It appears that there are differences in performance before and after the monetary
incentive, but are these differences significant? There is a need to test the relevant
hypotheses.

9-30
Chapter 09 - Hypothesis Testing

2. SCORE = 5:

.
The p-value Since the p-value
The percentage of students scoring
5, after the introduction of the monetary incentive, is not significantly higher than
8% (the historical average).

3. SCORE = 1 and 2:

The p-value 0.1867. Since the p-value

The percentage of students scoring 1
and 2, after the introduction of the monetary incentive, is not significantly lower
than 54% (the historical average).

9-31
Chapter 09 - Hypothesis Testing

4. Overall, the monetary incentive did not make any significant impact in improving
students’ performance in Chicago at the 5% significance level.

Case Study 9.3

1.

At the 5% significance level,

the well-being score of Hawaiians is not significantly higher than the national
average of 65.9.

9-32
Chapter 09 - Hypothesis Testing

2.

. The p-
value . Since the p-value = 0.0192 < 0.05 =
At the 5% significance level, we conclude that the proportion of
Hawaiians who score below 50 is less than 40%.

9-33