Making decisions based on sample data helps us evaluate claims about
(Introduction to Statistics & Data Analysis 3rd ed. pages 525-529/4th ed. pages 578-581)
HYPOTHESES AND TEST PROCEDURES
Correctly set up and carry out a hypothesis test about a population
• Correctly set up and carry out a hypothesis test about a population
• Describe Type I and Type II errors in context.
« Understand the factors that affect the power of a test.
In this section, we will look at the basics of setting up and carrying out
a hypothesis test using a univariate data set. Then we will use this
information to draw conclusions about some unknown population
parameter. Finally, anytime we make a decision based on sample data,
there is a risk of error, so we will discuss what types of errors might
be made when testing hypotheses.
HYPOTHESIS TESTING USING
A SINGLE SAMPLE
2) Practice MC
3) Practice Short Answer
4) Answer Key
EXAMPLE The marketing manager for an online computer game store
targets the company advertising toward males because he believes
that 75% of the company's purchases are made by men. The sales
manager claims that the proportion of purchases made by females has
the new tires have been manufactured as specified. After all, the tire
company wouldn't stay in business for long if they didn't provide what
the initial claim about the mean tread thickness that the auto company
believes to be fact. This initial assumption is called the null hypothesis.
We write the null hypothesis as:
H0:ft = 0.3125 .
H0 stands for "the null hypothesis"
ju is the population mean tread thickness for all tires of this type.
The auto manufacturer may suspect that there has been a change in
the mean thickness of the tire tread, so they decide to check several of
the tires. This leads the auto company to develop what is called an
alternative hypothesis. The alternative hypothesis is a competing
hypothesis and could be written in one of the following three ways:
Ha:p* 0.3125 in. or
fj. < 0.3125 in. or
H > 0.3125 in.
Ha stands for "the alternative hypothesis"
Because the auto manufacturer suspects that the mean tread
thickness has changed, but does not have a specific direction in mind,
they would use ju ^ 0.3125 in. as the alternative hypothesis.
No matter which alternative hypothesis the company uses, the
hypothesis testing procedure only allows us to favor this alternative if
there is strong evidence against the null hypothesis. This evidence
would come from sample data. We would evaluate what we see in the
the sample to determine if the sample mean tire tread is just too far
from what the null hypothesis specifies to be explained by just chance
differences from sample to sample. This same reasoning is used in all
hypothesis tests considered in the AP Statistics course.
The null hypothesis is usually written as
H0: some population characteristic = the hypothesized value
and the alternative hypothesis is written as one of the following:
Ha: some population characteristic * the hypothesized value
Ha: some population characteristic < the hypothesized value
Ha: some population characteristic > the hypothesized value
Now that we have an understanding of how to generate the null and
alternative hypotheses, a test procedure will be used to decide if we
should reject the null hypothesis. Test procedures are considered in
the next section. Once a decision is made after the test procedure is
performed, there is a chance that the final decision is wrong. In other
words, an error could have been made. There are two possible types
of errors and they are called a Type I error and a Type II error.
Either types of error may occur when making a decision either to
reject or to fail to reject the null hypothesis. For example, in the tire
tread problem, if a decision is made to reject the null hypothesis, this
could be a wrong decision that would cause the auto manufacturer to
conclude that the tires did not meet specifications. However, if the
decision was to fail to reject the null hypothesis, this could also be
wrong and the auto manufacturer could end up using tires that do not
meet specifications. In either case, there is a possible error exists that
is potentially damaging in some way.
A Type I error is made if we reject the null hypothesis and the null
hypothesis is actually true. Although the hypothesis test, based on
probability, supports the decision, we are led to an incorrect inference
about the population. This would amount to having strong enough
evidence to conclude that the tires do not have a mean tread thickness
of 0.3125 in. The company would decide to return the tires, causing the
tire manufacturer to lose money. If the tires actually meet
specifications, the tire manufacturer lost money due to the decision
A Type II error is made if we fail to reject the null and in reality the
null hypothesis is not true and should have been rejected. This type of
error would amount to not having enough evidence to say the tires did
not meet specifications. In this instance, the company would
unknowingly use these tires. This error could mean that customers
(Introduction to Statistics & Data Analysis 3rd ed. pages 531-534/4th ed. pages 582-586)
ERRORS IN HYPOTHESIS TESTING
The null and alternative hypotheses are written in terms of
population characteristics. In this example, the alternative is written as
"less than" since the sales manager's claim is that the proportion of
purchases made by males is less than what the marketing manager
H a :p males <0.75
Hypothesis Testing Using a Single Sample * 215
Once we have verified that the sample size is large enough and that
3. When n is large enough, the sampling distribution of p is
It is important to check to make sure the sample size is large
enough before carrying out a one-proportion hypothesis test. To verify
that the sample size is large enough, check to make sure that
1- /< = P
Next, we take the hypotheses we developed and systematically test
them to decide whether or not to reject the null hypothesis. This
process is known as a test procedure and the same basic procedure is
used in the many different hypothesis tests. However, depending on
the type of data that we have and the question of interest, there are
different hypothesis tests. The first test we consider is a large-sample
hypothesis test for a population proportion.
In this case, we are looking at categorical data that come from a
single sample, such as the data on the proportion of customers who
are male. In this situation, the data consists of observations on a
categorical variable with two possible values—male or female.
Just as was the case with a confidence interval, the hypothesis test
is based on the properties of the sampling distribution of p, the
sample proportion. Recall that p is the sample proportion based on a
(Introduction to Statistics & Data Analysis 3rd ed. pages 537-548/4th ed. pages 589-599)
LARGE-SAMPLE HYPOTHESIS TESTS FOR A PROPORTION
Type II error is more problematic as it is something that we can't
control easily. The values of a and /? are related—the smaller we
make a , the larger /3 becomes, all other things being equal. For this
reason, we generally choose a significance level a that is the largest
value that is considered an acceptable risk of Type I error. This will
help control for the errors by keeping a small as well as controlling a
216 * Chapters
P(z < -2.16 if H0 is true)
Because the conditions were met, we know that if H0 were true, z
has an approximately normal distribution. Using what we know about
the normal distributions, we know that getting a z score more than +2
or less than -2 does not occur very often. In fact, we can compute the
probability of observing a z value as small as -2.16 given that the
distribution is standard normal:
Next, we calculate the value of the z test statistic.
= 0.704 so we can now substitute into our test statistic
423(1 - 0.75) = 105.75 > 10
423(0.75) = 317.25 > 10
enough to convince us that chance differences from sample to sample
could not account for this difference? This is the question that is
answered by a hypothesis test.
First, let's check the assumptions needed. The sample was a random
sample of customers, so that condition is met. The second condition is
that the sample size is large enough, so we check
EXAMPLE Suppose that the sales manager in the online computer game
customer example selects a random sample of 423 previous customers
and finds that 298 were males. The sample proportion is then
= 0.70. We can see that 0.70 is smaller than 0.75, but is it small
If the null hypothesis is true, then this z statistic will have a standard
normal distribution. If the value of the z statistic is something that
would be "unexpected" for a standard normal variable, we regard this
as evidence that the null hypothesis should be rejected.
where p = population proportion.
In the example where we wanted to test H0 : p = 0.75 versus
Ha :p<0.75, we consider whether the sample proportion is enough
smaller than the hypothesized proportion of 0.75 that the difference
can't be explained just by sampling variability. To do this, we calculate
the value of the z statistic using 0.75 (from the null hypothesis) as the
value for p: z0.75(1-0.75)
Hypothesis Testing Using a Single Sample * 217
P-value > a, H0 should be not be rejected.
P-value < a, H0 should be rejected.
Ha -. p > hypothesized
Ha -. p < hypothesized
Ha : p •£ hypothesized
Finally, we note that there are three different computations of the Pvalue to consider. The computation chosen will depend on which
inequality (<, >, or *) appears in the alternative hypothesis. Here are
the three possible situations that can occur.
When writing a solution to a hypothesis test problem, make sure to
explicitly compare the P-value to a This shows you know how to
link the P-value and a in making your decision. Then always state
your conclusion in the context of the problem.
Once you have calculated the value of the test statistic and the
associated P-value, compare the P-value to the value of a .
If the P-value is small, we reject the null hypothesis. How small does
the P-value have to be in order to reject H0 ? This will depend on how
large a risk of a Type I error we are willing to assume. This level of risk
is preset by the researcher and is known as the significance level of the
test. It is denoted by a . For example, if we set a = 0.05 r this means we
know that a result as extreme as what we saw in the sample could
happen as often as 5% of the time if the null is true, but we can live
with a Type I error occurring for about 5 out of 100 of all possible
P-value—the probability of getting a test statistic value at least as
extreme as what was observed for the sample, if the null
hypothesis is actually true
Test statistic—a value computed from the sample data that is used
to make a decision to either reject H0 or fail to reject H0
Two key parts of a hypothesis test are the test statistic and the Pvalue.
. ' ' • • '
has a t distribution with df = n - 1
In either case, the forms of the hypotheses of interest look very
much like those of the proportions test. The difference is that now the
t = -— , when we DON'T know a
has approximately a standard normal distribution or
z = -— when we know a
Now that we have a handle on a hypothesis test for proportions, let's
consider tests based on numerical data. With numerical data, we are
usually interested in making inferences about a population mean. In
Chapter 8 on confidence intervals, there were two types of intervals
for estimating means, a z interval and a t interval. Which of these
intervals is used in a particular situation is determined by whether we
know the population standard deviation (a) or whether we have only
the sample standard deviation (sj. In either case, we found that if n is
large enough or if we know the population is roughly normally
distributed, then either
(Introduction to Statistics & Data Analysis 3rd ed. pages 550-55S/4th ed. pages 602-61 0)
HYPOTHESIS TESTS FOR A MEAN
p Read as p-hat, is used to denote the sample proportion.
p This is the notation used to denote the population proportion when
we have categorical data.
P-value This is a probability computed from the value of the test
statistic. All hypothesis tests involve the use of a P-value to make a
Using incorrect notation can lower your score on free-response
questions. Remember to keep the different "p" notations straight.
Hypothesis Testing Using a Single Sample * 219
ASSUMPTIONS The assumptions that must be satisfied are the same as
those with confidence intervals.
We must have a random sample and the sample size must be large
or the population distribution of tread thickness must be
approximately normal. The problem states that the sample was a
random sample, so that condition is met. The sample size is large (n is
Ha: jj. > 0.3125 (since the company suspects the mean
tread is greater than the requirement of 0.3125)
set the significance level (a) at 0.05.
H0: fi = 0.3125
EXAMPLE Trustworthy Tires sells their leading high performance tire to
a car manufacturer. The car manufacturer requires that the tires have
a mean tread thickness of 0.3125. The car manufacturer thinks tires
received from Trustworthy may not be meeting this requirement and
that the mean tread may be greater than 0.3125, because they have
been finding that in some cars the tires are hitting parts of the wheel
well area on bumps, hi a random sample of 32 tires, they found the tire
treads to have a mean of 0.3625 in. and a standard deviation of 0.094
in. Since both these values came from the sample, we use the notation
for sample statistics: x = 0.3360 and sx = 0.094. For this situation, let's
look at the components that are common to all hypothesis tests.
P-value: computed as area under the z curve
If a is NOT known
_x- hypothesized value
P-value: computed as area under the t curve with df = n -1
It is very rare that the population standard deviation is known, so
we will focus mainly on the t statistic. Revisiting the earlier tire tread
scenario, we can write the hypotheses, check assumptions, calculate
the test statistic and P-value, and give a conclusion in context.
The test statistic is based on either the z or the t statistic shown
above, depending on whether or not we know a.
If a is known
x - hypothesized value
220 * Chapter 9
Time to Pain Relief (in minutes)
SAMPLE PROBLEM 1 A well-known brand of pain relief tablets is
advertised to begin relief within 24 minutes. To test this claim, a
random sample of 18 subjects suffering from the same types of
headache pain record when they first notice relief after taking the pain
relief tablet. The data gathered from this study are shown.
Note that the steps in a test about a population mean are the same
as for the test for a population proportion. What distinguishes the two
tests is the type of data (numerical for means and categorical for
proportions), the specific assumptions that must be checked, and the
test statistic used for the test.
When carrying out a test for means, whether or not you know the
standard deviation of the population is what determines if you should
use a z test or t test. Only use the z test when you are sure the
standard deviation given is from the population.
CONCLUSION In this example, with a = 0.05 we see that the P-value is not
less than a. We fail to reject the null hypothesis and conclude that
there is not convincing evidence that the mean tread thickness of tires
is greater than 0.3125 in.
Note that because the inequality in the null hypothesis was >, the Pvalue is the area to the right of 1.41 under the t curve with df = 31.
with df = n -1
= 32-1 = 31
P - value = 0.08
x - hypothesized value
Hypothesis Testing Using a Single Sample
a = 0.05 (given]
With ]i representing the population mean time to relief,
CONCLUSION Since 0.027 < 0.05, we reject the null hypothesis in favor of
the alternative hypothesis. There is convincing evidence that the mean
time to pain relief is greater than 24 minutes.
As a final note, should you be in the unusual situation where the
population standard deviation (CT), is given, you would use the z test
statistic and the association P-value would be determined using the
standard normal distribution. Otherwise, the process would be the
same as the process illustrated in the example above.
P-value = 0.027
x = 24.78, sx = 1.59, df = 17
TEST STATISTIC A t test will be used since we don't know the population
22 23 24 25 26 27 28 29 30
Random sample: The problem states the 18 subjects are a random
Large sample or normal population distribution: Since there were
only 18 subjects, we need to be willing to assume that the distribution
of relief times in the population is approximately normal. A dotplot of
the sample relief times is shown below. Because the dotplot is
approximately symmetric and there are no outliers, it is reasonable to
think that the population distribution is approximately normal.
While the AP curriculum does not require you to calculate power, you
are expected to know the factors that affect the power of a test. There
are three factors that are generally considered when thinking about
power. First, Increasing « will raise the power. Although this seems
like a fast fix, it is dangerous because the probability of a Type I error
will also increase. Another way to raise power is simply to increase the
sample size, although this isn't always practical. The other things that
(Introduction to Statistics & Data Analysis 3rd ed. pages 562-567/4th ed. pages 613-621)
POWER AND PROBABILITY OF TYPE II ERROR
P > a, H0 should not be rejected.
P < a, H0 should be rejected.
Conclusion (be sure you state it in context too)
Give the P-value associated with the value of the test statistic
Show your z or t calculation
The assumptions vary by test, but you should always state AND
CHECK the assumption appropriate for the test you are performing.
Identify the test procedure by name or by formula. Check All
Assumptions (or Conditions)
Ho and Ha (correctly written)
State the Hypotheses and define any symbols used in the
For a standard hypothesis test, you will be required to provide
complete answers to four key parts in every test. Get in the habit of
thinking in terms of these four pieces. If you chunk the information
under each of these categories, you will have a better chance of
remembering all aspects of any hypothesis test. The four pieces are
Hypothesis Testing Using a Single Sample * 223
(c) One way to increase power would be to increase the significance
level of the test. However, this will also increase the chance of a
Type I error. Also, increasing sample size will increase power.
(b) A Type II error in this situation would occur by concluding there
wasn't enough evidence to say that the proportion of customers
who are male is less than 0.75 when this proportion really is less
than 0.75. In this case, the company would probably continue to
target males in its advertising, which might result in a loss of
potential sales to female customers.
(a) A Type I error would result if it were concluded that the
proportion of customers who are male is less than 0.75, when in
fact this proportion is 0.75. A possible consequence of this error
would be that the company might change its strategy of targeting
males in its advertising, which might result in a decrease in sales.
SOLUTION TO PROBLEM 2
(c) What can be done to increase the power of this test?
(b) Identify the Type II error in this scenario and provide a possible
consequence of this error.
(a) Identify the Type I error in this scenario and provide a possible
consequence of this error;
H 0 :p = 0.75
H a :p<0.75
proportion of males was less than the 0.75 claimed by the marketing
manager. The appropriate hypotheses for this situation were:
Hypotheses: In either symbols or words, you need to clearly state
both the null hypothesis and the alternative hypothesis.
Once data has been gathered and an appropriate hypothesis test
carried out, the findings are typically shared with others interested in
the outcome. In communicating results in journals and newspapers, it
is not common to provide the same level of detail that you would want
to provide in a solution to a hypothesis testing question on the AP
exam. Some of the important things to conclude when reporting
(Introduction to Statistics & Data Analysis 3rd ed. pages 571-574/4th ed pages 623-625)
INTERPRETATION OF RESULTS IN HYPOTHESIS TESTING
You will be able to write the null and alternative hypothesis for a
test about a population mean or a population proportion.
« You will be able to describe Type I and Type II errors in context.
• You will be able to describe a possible consequence of each type
error in context.
• You will be able to carry out a test of hypotheses about a population
• You will be able to carry out a test of hypotheses about a population
• You will be able to interpret the result of a hypothesis test in
HYPOTHESIS TESTING USING A SINGLE SAMPLE: STUDENT
OBJECTIVES FOR THE AP EXAM
In writing a conclusion for a hypothesis test, remember you are
always either rejecting or failing to reject the null hypothesis. This
means you either have convincing evidence in favor of the alternate
hypothesis or that there is not enough evidence. You never say that
you accept the null hypothesis because that implies strong evidence
for the null hypothesis.
In many cases, the reported results only include a statement such as
P-value < 0.05. This is common and tells the reader that the results of
the test yielded a P-value smaller than 0.05, hence statistically
significant. Journals may also use a standard method of coding. * significant, would mean their P-value was <0.05, ** = very significant,
means P-value < 0.01.
As you review published reports, be sure to look for the four key
components you would report and ask yourself some questions about
these pieces. What were the hypotheses they tested? Did they use an
appropriate test for these? What was the associated P-value and what
significance level was used? Also, were the conclusions reached
consistent with the results of the test?
include a comparison of the P-value to a. Stating that you reject the
H0 is not sufficient.
Hypothesis Testing Using a Single Sample * 225
4. A Type II error occurs in which of the following situations?
(A) H0 is rejected and the null hypotheses is true.
(B) HO is not rejected and the null hypotheses is false.
(C) Ha is not rejected and the null hypotheses is false.
3. A Type I error occurs in which of the following situations?
(A) Ha is rejected and the null hypotheses is true.
(B) H0 is not rejected and the null hypotheses is false.
(C) H0 is rejected and the null hypotheses is true.
(D) The P-value is too small to reject the null hypothesis.
(E) The « level is too small and so the null hypothesis is rejected.
2. A concrete learner is a student who learns best when various types
of hands-on or manipulative activities are used to illustrate
abstract concepts. Researchers have long believed that 60% of all
students remain concrete learners until they are between 16 and 21
years of age. Each student in a random sample of 32 students age
17 to 19 was evaluated, and it was found that 24 of the 32 were
concrete learners. Would it be appropriate to use the z test for a
population proportion to test to determine if the proportion of
concrete learners in this age group is less than 0.60?
(A) Yes. Since 32(0.6) = 19.2 and 32(1 - 0.6) = 12.8, and we can
proceed with the test.
(B) Yes. Since 32 is larger than 30, the sample is sufficiently large
and we can proceed with the test.
(C) Yes. Since we know from the sample was a random sample, we
can proceed with the test.
(D) No. Since 32(0.05) = 1.6, we do not have a large enough sample
to proceed with test.
(E) No. While 32 is larger than 30, it is so close to 30 and we don't
know if the population distribution is normal.
1. A psychologist reports that the result of a hypothesis test was
statistically significant at the 0.05 level. Which of the following is
consistent with this statement?
(A) The P-value calculated was smaller than the significance level
(B) The P-value calculated was larger than the significance level of
(C) The significance level calculated was larger than 0.05.
(D) The significance level calculated was smaller than 0.05.
(E) There was not enough information to make a decision.
6. An animal rights group has been very supportive of a new silicon
product that caps the nails on cats as an alternative to surgically
declawing the pets. The company who makes the caps claims they
last for an average of 69 days before needing to be replaced.
Before publically endorsing the product, the animal rights group
plans to collect data to see if there is convincing evidence that the
mean time before replacement is needed is actually less than what
the company claims. Which of the following would be an
appropriate pair of hypotheses for the animal rights group to test?
(A) H0: ju = 69 days, Ha: ju > 69 days
(B) H0: V = 69 days, Ha: ju < 69 days
(C) H0: // = 69 days, Ha: n * 69 days
(D) H0: x = 69 days, Ha: x > 69 days
(E) H0: x = 69 days, Ha: x < 69 days
5. A graduate student at a private university wanted to study the
amount of money that students at his university carried with them.
A recent study reported that the average amount of money carried
by college students is $31. He decides to collect data and carry out
a test to see if there is evidence that the average is higher for
students at his university. Which of the following describes a Type
II error in this context?
(A) This would lead to the incorrect idea that students at his
university, on average, spend more money each month than
students at other universities.
(B) This would lead to the incorrect idea that students at his
university carry, on average, more than $31.
(C) This would lead to the incorrect idea that students at his
university carry, on average, less than $31.
(D) This would lead to the incorrect idea that there was no reason
to believe that students at his university carry, on average,
more than $31.
(E) This would lead to the correct idea that students at other
campuses carry, on average, less than $31.
Hypothesis Testing Using a Single Sample * 227
OK _ 9 Q
, with d f = 3 9
, with df = 39
(E) t =
(D) t =
, with df = 40
~ , with df = 40
(C) t = =^^-, with df = 39
(B) t ==
(A) z =
7. Neutering dogs is a common surgical practice. The mean time to
recover from the general anesthetic used is 28 hours. A
veterinarian believes that since changing to a new anesthetic, the
mean recovery time is shorter than before. To investigate, she
selects a random sample of 40 surgeries done with the new
anesthetic and finds that the mean recovery time was 25 hours an
the standard deviation was 2.5. She plans to use this sample data t
test to see if there is evidence that the recovery time is shorter wit
the new anesthetic. Which of the following is the correct test
statistic for this study?
\) z 83
9. Bicycles purchased from a discount store come unassembled. The
assembly instructions that come with the bicycle claim that the
average assembly time is 30 minutes. A consumer group has
received complaints from people who say that the assembly time
was greater than the time claims. They decide to purchase 40 of
these bikes and have asked 40 different people to assemble them.
The consumer group believed that it was reasonable to regard
these 40 people as representative of the population of people who
might purchase this bike. For this sample, they found that the
assembly times had a mean of 34.2 minutes and a standard
deviation of 8.6 minutes. Is there convincing evidence that the
claimed average assembly time is too low at the 0.05 significance
(A) No, z = 0.49 P-value = 0.312.
(B) Yes, t = 3.09, df = 39, P-value = 0.002.
(C) Yes, t = 3.05, df = 39, P-value = 0.004.
(B) z =
\) 82z = - '
\) z =83
A recently published study reported that 63% of the nation's
students have some type of structured homework study time. A
school surveyed each student in a random sample of 83 students
who attend the school and found that only 52% reported having a
structured homework time. This data was used to carry out a
hypothesis test to determine if there was evidence that the
proportion of students at the school who had structured
homework time was less that the proportion reported in the
national study. Which of the following would be the test statistic
for this test?
Hypothesis Testing Using a Single Sample *
13. A local group claims that more than 60% of the teens driving after
10 p.m. are exceeding the speed limit. They plan to collect data in
hopes that a hypothesis test will provide convincing evidence in
support of their claim. Which of the following is true about the
hypotheses the group should test?
(A) The null hypothesis states that less than 60% of the teens are
III. The standard deviation of the statistic p is a^ - ^/npCl - p).
(A) I only
(B) n only
(C) III only
(D) II and III
(E) I, H, and III
12. Which of the following statements are true?
I. The null hypothesis for test about a population proportion
written as H0:p = hypothesized value.
n. For the z test to be an appropriate test for a population
proportion, the following condition must be met: np > 10 and
11. Which of the following is closest to the P-value associated with a
two-tailed t test with 20 degrees of freedom if the value of the test
statistic is 2.0?
10. The prom committee is thinking about changing the location of the
prom. The new location is more expensive to rent, and for the
increased cost to be reasonable, they would want to be fairly
certain that more than 46% of the senior class would attend the
prom. A survey of a random sample of 52 seniors found that 25
would attend if the site changed. Which of the following pairs of
hypotheses should the prom committee test?
(A) H0: ju = 46%, Ha: // > 46%
(B) H 0 :// = 46%,H a:> «*48%
(C) H0: p = 0.46, Ha:p> 0.46
(D) H0: p = 0.46, Ha:p * 0.46
(E) H0: p = 0.48, Ha:p< 0.46
230 <• Chapter 9
1. A bridal gown industry publication claims that nationwide the
average amount spent for a wedding gown is $1,012. A local bridal
shop in an urban community has noticed their more expensive
gowns are not selling well. Instead, the brides seem to be selecting
15. Suppose that the mean height of women in the United States is
64.5 in. with a standard deviation of 2.5 in. A clothing designer
feels that women who use her products may actually be taller on
average. She selects a random sample of 70 women from all
women who have previously purchased her clothing. What is the
population of interest, and what test would the designer use to test
(A)The population is all women in the United States and the
appropriate test is a t test with df = 70.
(B) The population is all women in the United States and the
appropriate test is a z test with = 2.5.
(C)The population is all women who have previously purchased the
designer's clothing and the appropriate test is a t test with df =
(D) The population is all women who have previously purchased
the designer's clothing and the appropriate test is a t test with
(E) The population is all women who have previously purchased
the designer's clothing and the appropriate test is a z test with
14. A study by a geological research team found that a new piece of
equipment designed to measure the forces of an earthquake is not
effective. They based this conclusion on data from a sample of 40
pieces of equipment and they carried out a test with a = 0.05. The
manufacturer of the equipment claims this study was flawed and
that their equipment is good. The research team is considering
carrying out a second study with the intention of increasing the
power of the test. Which of the following would ensure an increase
in the power of the test?
(A) Move the equipment to three randomly chosen new locations.
(B) Change a = 0.05 to a = 0.02.
(C) Carry out a two-sided test instead of a one-sided test.
(D) Increase the sample size to 60 pieces of equipment being
(E) Decrease the sample size to 20 pieces of equipment being
Hypothesis Testing Using a Single Sample * 231
2. A local school district believes that the proportion of seniors who
are absent from school on the last day of school may be increasing.
Over the past 5 years, 39% of the seniors have missed the last day.
This year, the school district is considering a new reward program
sponsored by local businesses where seniors who were at school
on the last day would be entered in a drawing for an iPad. To see if
this program might reduce the proportion of seniors who miss
school on the last day, a random sample of 398 seniors from the
school district was surveyed. Each student in the sample was asked
if they planned to attend on the last day of class given the
possibility of winning an iPad. Only 129 of the 398 seniors
indicated that they would miss the last day of school. The school
district would like to know if there is convincing evidence that the
new program would reduce the number of seniors absent on the
last day of school.
(a) What hypotheses should the school district test?
(b) Identify the appropriate test and verify that any conditions
needed for the test are met.
(c) Describe Type I and Type II errors in the context of this
A. When a researcher says the results were statistically
significant, it means the P-value was less than the set significance
level (Introduction to Statistics & Data Analysis 3rd ed. pages 571574/4th ed. pages 623-625).
A. In a one-sample proportions test, one condition that is needed
is np > 10 and n(l - p) > 10. Notice these both use p and not p
from the sample (Introduction to Statistics & Data Analysis 3rd ed.
pages 537-548/4th ed. pages 589-599).
C. By definition, a Type I error occurs when the null hypothesis is
rejected when it should not be rejected. This might happen when
the P-value < significance level (Introduction to Statistics & Data
Analysis 3rd ed. pages 531-534/4th ed. pages 582-586).
B. A Type II error will occur anytime you fail to reject the null
when in fact the null is false. This might occur if the P-value is not
smaller than a (Introduction to Statistics & Data Analysis 3rd ed.
C. Since we don't know a, it must be a t test. Also, while df = 39,
n = 40 is the value that is used in the test statistic calculation
(Introduction to Statistics & Data Analysis 3rd ed. pages 550558/4th ed. pages 602-610).
D. In calculating the test statistic, the denominator uses the
hypothesized value and sample size of 83 (Introduction to Statistics
& Data Analysis 3rd ed. pages 537-548/4th ed. pages 589-599).
B. A t test with 39 degrees of freedom would be used
(Introduction to Statistics & Data Analysis 3rd ed. pages 550558/4th ed. pages 602-610).
14. D. The easiest ways to increase power are either to increase the
sample size or use a larger significance level (Introduction to
Statistics & Data Analysis 3rd ed. pages 562-567/4th ed. pages
13. E. Answer choices A, B, and D can be eliminated since the null
hypothesis must include the equal case. Because the group wants
to show support for then claim that more than 60% are speeding,
the alternative hypothesis would be p > 0.60 (Introduction to
Statistics & Data Analysis 3rd ed. pages 525-529/4th ed. pages
(Introduction to Statistics & Data Analysis 3rd ed. pages 537548/4th ed. pages 589-599).
random variable. The standard deviation of p is
12. B. Notice that choice I is written using p instead of p. All
hypotheses are stated in terms of the population value, which
would be p. Choice III is the standard deviation for a binomial
11. D. The P-value is approximately 0.06, which is closest to 0.05
(Introduction to Statistics & Data Analysis 3rd ed. pages 550558/4th ed. pages 602-610).
10. C. This is a test about a population proportion. The question of
interest is whether the population proportion p is greater than
0.46 (Introduction to Statistics & Data Analysis 3rd ed. pages 525529/4th ed. pages 578-581).
B. Since the group is concerned only if the caps last less than 69
days on average, the alternative hypothesis would be u < 69 days
(Introduction to Statistics & Data Analysis 3rd ed. pages 525529/4th ed. pages 578-581).
Hypothesis Testing Using a Single Sample * 233
(b) The appropriate test is a 1-sampJe z test where
(a) H0 •. p = 0.39 where p = proportion of seniors skipping
Ha -. p < 0.39
(Introduction to Statistics & Data Analysis 3rd ed. pages 550558/4th ed. pages 602-610).
Since the F-value of 0.21 is not smaller than 0.05, there is not
convincing evidence that the average amount spent on a wedding
gown at this shop is less than the national average of $1,012
p = 0.21
x = $985
sx = $235
n = 50
Since 50 > 30, the sample is large enough for the one sample t test
to be appropriate.
The problem states this was a random sample of wedding gown
H a ://<$1012
a = 0.05
234 * Chapter9
(Introduction to Statistics & Data Analysis 3rd ed. pages 531534/4th ed. pages 582-586).
A Type II error would be that the iPad drawing would in fact
reduce the proportion of seniors who miss the last day of
school, but the school district is not convinced of this and
does not implement the drawing.
Hypothesis Testing Using a Single Sample * 235