Contents

Contents 1

8 Inferences Based on a Single Sample: Tests of Hypothesis 2

8.1 The Elements of a Test of Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
8.2 Formulating Hypotheses and Setting Up the Rejection Region . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
8.3 Observed Significance Levels: p-Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
8.4 Test of Hypothesis about a Population Mean: Normal (z) Statistic . . . . . . . . . . . . . . . . . . . . . . . . 10
8.5 Test of Hypothesis about a Population Mean: Student’s t-Statistic . . . . . . . . . . . . . . . . . . . . . . . . 13
8.6 Large-Sample Test of Hypothesis about a Population Proportion . . . . . . . . . . . . . . . . . . . . . . . . . 16

1
Chapter 8

Inferences Based on a Single Sample: Tests of

Hypothesis

Suppose you wanted to determine whether the mean level of a driver’s blood alcohol exceeds the legal limit after two drinks,
or whether the majority of registered voters approve of the president’s performance. In both cases, you are interested in
making an inference about how the value of a parameter relates to a specific numerical value. Is it less than, equal to, or
greater than the specified number? This type of inference, called a test of hypothesis, is the subject of this chapter.

8.1 The Elements of a Test of Hypothesis

Suppose we are interested in making an inference about the mean µ of a population. However, we are less interested in

estimating the value of µ than we are in testing a hypothesis about its value—that is, we want to decide whether the mean

of the ppopulation is less than, equal to, or greater than the specified number.

A statistical hypothesis is a statement about the numerical value of a population parameter.

We define two hypotheses:

1- The null hypothesis, denoted H0 , represents the hypothesis that will be assumed to be true unless the data provide

convincing evidence that it is false. This usually represents the “status quo” or some statement about the population

parameter that the researcher wants to test.

2- The alternative (research) hypothesis, denoted Ha (H1 ), represents the hypothesis that will be accepted only if the

data provide convincing evidence of its truth. This usually represents the values of a population parameter for which

the researcher wants to gather evidence to support.

The null and alternative hypotheses for instance can be stated as follows:

H0 : µ ≤ 2400

Ha : µ > 2400

Because the hypotheses concern the value of the population mean µ, it is reasonable to use the sample mean x̄ to make the

inference, just as we did when we formed confidence intervals for µ in Sections 7.2 and 7.3. In example above, “Convincing”

evidence in favor of the alternative hypothesis will exist when the value of x̄ exceeds 2,400 by an amount that cannot be

readily attributed to sampling variability. To decide, we compute a test statistic, i.e., a numerical value computed from the

sample.

2
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 3

The test statistic is a sample statistic, computed from information provided in the sample, that the researcher uses to

decide between the null and alternative hypotheses.

Figure 8.1: The sampling distribution of x̄, assuming µ = 2,400

A Type I error occurs if the researcher rejects the null hypothesis in favor of the alternative hypothesis when, in fact, H0

is true. The probability of committing a Type I error is denoted by α.

α = P (Type I error) = P (Rejecting the null hypothesis when in fact the null hypothesis is true)

In our example,

α = P (z > 1.645 when infact µ = 2, 400) = .05

We now summarize the elements of the test:

H0 : µ ≤ 2400

Ha : µ > 2400
x̄−2400
Test statistic: z = σx̄

Rejection region: z > 1.645, which corresponds to α = .05

The rejection region of a statistical test is the set of possible values of the test statistic for which the researcher will reject

H0 in favor of Ha .

Suppose: x̄ = 2, 460, n = 50, and s = 200 as in the case of estimation, we can use s to approximate σ when s is calculated

from a large set of sample measurements. Then:

x̄−2400 2460−2400
Test statistic: z = σx̄ ≈ 200
√
= 2.12
50

Because this value of z exceeds 1.645, it falls into the rejection region. That is, we reject the null hypothesis that µ = 2, 400

and conclude that µ > 2, 400.

How much faith can be placed in this conclusion?

The answer is α = .05—that is, we selected the level of risk, α, of making a Type I error when we constructed the test.

A Type II error occurs if the researcher accepts the null hypothesis when, in fact, H0 is false. The probability of committing

a Type II error is denoted by β.

CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 4

Figure 8.2: Location of the test statistic for a test of the hypothesis H0 : µ = 2,400

It is often difficult to determine β precisely. Rather than make a decision (accept H0 ) for which the probability of error β

is unknown, we avoid the potential Type II error by avoiding the conclusion that the null hypothesis is true. Instead, we will

simply state that the sample evidence is insufficient to reject H0 at α = .05. Because the null hypothesis is the “status-quo”

hypothesis, the effect of not rejecting H0 is to maintain the status quo. Therefore:

Because α is usually specified by the analyst, we will generally be able to reject H0 (accept Ha ) when the sample evidence

supports that decision. However, because β is usually not specified, we will generally avoid the decision to accept H0 ,

preferring instead to state that the sample evidence is insufficient to reject H0 when the test statistic is not in the rejection

region.

Following table summarizes the four possible outcomes (i.e., conclusions) of a test of hypothesis. The “true state of

nature” columns refer to the fact that either the null hypothesis H0 is true or the alternative hypothesis Ha is true.

Conclusions and Consequences for a Test of Hypothesis

True State of Nature

Conclusion H0 True Ha True
Accept H0 (Assume H0 True) Correct decision Type II error (probability β)
Reject H0 (Assume Ha True) Type I error (probability α) Correct decision

Note

Note that a Type I error can be made only when the null hypothesis is rejected in favor of the alternative hypothesis,

and a Type II error can be made only when the null hypothesis is accepted.

Note

The elements of a test of hypothesis are summarized as follows. Note that the first four elements are all specified before

the sampling experiment is performed. In no case will the results of the sample be used to determine the hypotheses; the

data are collected to test the predetermined hypotheses, not to formulate them.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 5

Elements of a Test of Hypothesis

1- Null hypothesis (H0 ): A theory about the specific values of one or more population parameters. The theory generally

represents the status quo, which we adopt until it is proven false. The theory is always stated as

H0 : parameter = value.

2- Alternative (research) hypothesis (Ha ): A theory that contradicts the null hypothesis. The theory generally represents

that which we will adopt only when sufficient evidence exists to establish its truth.

3- Test statistic: A sample statistic used to decide whether to reject the null hypothesis.

4- Rejection region: The numerical values of the test statistic for which the null hypothesis will be rejected. The rejection

region is chosen so that the probability is α that it will contain the test statistic when the null hypothesis is true,

thereby leading to α Type I error. The value of α is usually chosen to be small (e.g., .01, .05, or .10) and is referred to

as the level of significance of the test.

5- Assumptions: Clear statement(s) of any assumptions made about the population(s) being sampled.

6- Experiment and calculation of test statistic: Performance of the sampling experiment and determination of the numerical

value of the test statistic.

7- Conclusion:

a. If the numerical value of the test statistic falls in the rejection region, we reject the null hypothesis and conclude that

the alternative hypothesis is true. We know that the hypothesis-testing process will lead to this conclusion incorrectly

(Type I error) only 100α% of the time when H0 is true.

b. If the test statistic does not fall in the rejection region, we do not reject H0 . Thus, we reserve judgment about

which hypothesis is true. We do not conclude that the null hypothesis is true because we do not (in general) know the

probability β that our test procedure will lead to an incorrect acceptance of H0 (Type II error).

8.2 Formulating Hypotheses and Setting Up the Rejection Region

Steps for Selecting the Null and Alternative Hypotheses

1- Select the alternative hypothesis as that which the sampling experiment is intended to establish. The alternative

hypothesis will assume one of three forms:

a. One-tailed, upper-tailed (Ha : µ > µ0 )

b. One-tailed, lower-tailed (Ha : µ < µ0 )

c. Two-tailed (Ha : µ 6= µ0 )
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 6

2- Select the null hypothesis as the status quo, that which will be presumed true unless the sampling experiment conclu-

sively establishes the alternative hypothesis. The null hypothesis will be specified as that parameter value closest to the

alternative in one-tailed tests and as the complementary (or only unspecified) value in two-tailed tests. (H0 : µ = µ0 )

A one-tailed test of hypothesis is one in which the alternative hypothesis is directional and includes the symbol “<” or “>”.

A two-tailed test of hypothesis is one in which the alternative hypothesis does not specify departure from H0 in a particular

direction and is written with the symbol“6=.”

Example 8.2.1. it is producing machine bearings with a mean diameter of .5 inch. If the mean diameter of the bearings is

larger or smaller than .5 inch, then the process is out of control and must be adjusted. Formulate the null and alternative

hypotheses for a test to determine whether the bearing production process is out of control.

Solution

The hypotheses must be stated in terms of a population parameter. Here, we define µ as the true mean diameter (in

inches) of all bearings produced by the metal lathe. If either µ > .5 or µ < .5, then the lathe’s production process is out of

control.

H0 : µ = 0.5 (i.e., the process is in control)

Ha : µ 6= 0.5 (i.e., the process is out of control)

Note

Whenever a claim is made about the value of a particular population parameter and the researcher wants to test the

claim, believing that it is false, the claimed value will represent the null hypothesis.

Rejection Region

The rejection region for a two-tailed test differs from that for a one-tailed test. When we are trying to detect departure

from the null hypothesis in either direction, we must establish a rejection region in both tails of the sampling distribution of

the test statistic. The following figure depicts three possibilities:

Figure 8.3: Rejection regions corresponding to one- and two-tailed tests

Note that the smaller α you select, the more evidence (the larger z) you will need before you can reject H0 .
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 7

Rejection Regions for Common Values of α, Large n

Alternative Hypotheses
Lower-Tailed Upper- Two-Tailed
Tailed
α = 0.1 z < −1.28 z > 1.28 z < −1.645 or z > 1.645
α = 0.05 z < −1.645 z > 1.645 z < −1.96 or z > 1.96
α = 0.01 z < −2.33 z > 1.28 z < −2.575 or z > 2.575

Example 8.2.2. The effect of drugs and alcohol on the nervous system has been the subject of considerable research.

Suppose a research neurologist is testing the effect of a drug on response time by injecting 100 rats with a unit dose of the

drug, subjecting each rat to a neurological stimulus, and recording its response time. The neurologist knows that the mean

response time for rats not injected with the drug (the “control” mean) is 1.2 seconds. She wishes to test whether the mean

response time for drug-injected rats differs from 1.2 seconds. Set up the test of hypothesis for this experiment, using α = .01.

Solution The key word mean in the statement of the problem implies that the target parameter is µ, the mean response

time for all drug-injected rats.

H0 : µ = 1.2 (Mean response time is 1.2 seconds)

Ha : µ 6= 1.2 (Mean response time is less than 1.2 or greater than 1.2 seconds
x̄−1.2
Test statistic: σx̄

Rejection region: z < −2.575 or z > 2.575

Figure 8.4: Two-tailed rejection region: α = .01

Assumptions: Since the sample size of the experiment is large enough (n > 30), the Central Limit Theorem will apply,

and no assumptions need be made about the population of response time measurements. The sampling distribution of the

sample mean response of 100 rats will be approximately normal, regardless of the distribution of the individual rats’ response

times.

Note that the test is set up before the sampling experiment is conducted.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 8

8.3 Observed Significance Levels: p-Values

According to the statistical test procedure described in Section 8.2, the rejection region and, correspondingly, the value of

α are selected prior to conducting the test, and the conclusions are stated in terms of rejecting or not rejecting the null

hypothesis. A second method of presenting the results of a statistical test is based on the observed significance level (or

p-value) for the test.

The observed significance level, or p-value, for a specific statistical test is the probability (assuming H0 is true) of observing

a value of the test statistic that is at least as contradictory to the null hypothesis, and supportive of the alternative hypothesis,

as the actual one computed from the sample data.

Steps for Calculating the p-Value for a Test of Hypothesis

1. Determine the value of the test statistic z corresponding to the result of the sampling experiment.

2. a. If the test is one-tailed, the p-value is equal to the tail area beyond z in the same direction as the alternative

hypothesis. Thus, if the alternative hypothesis is of the form > , the p-value is the area to the right of, or above, the

ob- served z-value. Conversely, if the alternative is of the form < , the p-value is the area to the left of, or below, the

observed z-value.

b. If the test is two-tailed, the p-value is equal to twice the tail area beyond the observed z-value in the direction

of the sign of z—that is, if z is positive, the p-value is twice the area to the right of, or above, the observed z-value.

Conversely, if z is negative, the p-value is twice the area to the left of, or below, the observed z-value. See the following

figures:

Figure 8.5: Finding the p-value for a one-tailed test

Figure 8.6: Finding the p-value for a two-tailed test: p-value =2( p2 )
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 9

Reporting Test Results as p-Values: How to Decide Whether to Reject H0

1. Choose the maximum value of α that you are willing to tolerate.

2. If the observed significance level (p-value) of the test is less than the chosen value of α, reject the null hypothesis.

Otherwise, do not reject the null hypothesis.

Note: Some statistical software packages (e.g., SPSS) will conduct only two-tailed tests of hypothesis. For these packages,

you obtain the p-value for a one-tailed test as shown in the box:

Reported p-value
a. p = 2
(
Ha is of form > and z is positive
if (8.1)
Ha is of form < and z is negative

b. p = 1 − ( Reported2 p-value )

(
Ha is of form > and z is negative
if (8.2)
Ha is of form < and z is positive

Example 8.3.1. Consider the one-tailed test of hypothesis, H0 : µ = 100 versus Ha : µ > 100.

a. Suppose the test statistic is z = 1.44. Find the p-value of the test and the rejection region for the test when α = .05.

Then show that the conclusion using the rejection region approach will be identical to the conclusion based on the

p-value.

b. Now suppose the test statistic is z = 3.01; find the p-value and rejection region for the test when α = .05. Again, show

that the conclusion using the rejection region approach will be identical to the conclusion based on the p-value.

Solution

a. For z = 1.44, the observed significance level of the test is:

p-value = P (z > 1.44) = 1 − P (z < 1.44) = 1 − .925 = .075

This p-value is shown on Figure 8.7. Since α = .05 and the test is upper-tailed, the rejection region for the test is

z > 1.645. This rejection region is also shown in Figure 8.7. Observe that the test statistic (z = 1.44) falls outside the

rejection region, implying that we fail to reject H0 . Also, α = .05 is less than p-value = .075. This result also implies that

we should fail to reject H0 . Consequently, both decision rules agree—there is insufficient evidence to reject H0 .
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 10

Figure 8.7: Test for Example 8.3.1a—Fail to reject H0

b. For z = 3.01, the observed significance level of the test is:

p-value = P (z > 3.01) = 1 − P (z < 3.01) = 1 − .9987 = .0013

The test statistic (z = 3.01) falls within the rejection region, leading us to reject H0 . And, α = .05 now exceeds the

p-value (.0013), which also implies that we should reject H0 .

Figure 8.8: Test for Example 8.3.1b—Reject H0

8.4 Test of Hypothesis about a Population Mean: Normal (z) Statistic

When testing a hypothesis about a population mean µ, the test statistic we use will depend on whether the sample size n

is large (say,n > 30) or small and whether we know the value of the population standard deviation, σ. In this section, we

consider the large-sample case. Therefore, according to Chapter 7, the test statistic for a test based on large samples will

be based on the normal z-statistic and the sample standard deviation s provides a good approximation to σ in canse it is

unknown.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 11

Large-Sample Test of Hypothesis about µ Based on a Normal (z) Statistic

Test statistic: σ known

x̄−µ0
zc = √σ
n

Test statistic: σ unknown

x̄−µ0
zc = √s
n

Lower-Tailed Tests

H0 : µ = µ0

Ha : µ < µ0

Rejection region: z < −zα

p-value: P (z < zc )

Upper-Tailed Tests

H0 : µ = µ0

Ha : µ > µ0

Rejection region: z > zα

p-value: P (z > zc )

Two-Tailed Tests

H0 : µ = µ0

Ha : µ 6= µ0

Rejection region: |z| > z α2

p-value: if zc is positive 2P (z > zc ) if zc is negative 2P (z < zc )

Decision

Reject H0 if p − value < α or if test statistic (zc ) falls in rejection region ( the result is “statistically significant”) where:
α
P (z > zα ) = α, P (z > z α2 ) = 2

Conditions Required for a Valid Large-Sample Hypothesis Test for µ

1. A random sample is selected from the target population.

2. The sample size n is large (e.g.,n ≥ 30).

CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 12

Possible Conclusions for a Test of Hypothesis

1. If the calculated test statistic falls in the rejection region, reject H0 and conclude that the alternative hypothesis Ha

is true. State that you are rejecting H0 at the α level of significance. Remember that the confidence is in the testing

process, not the particular result of a single test.

2. If the test statistic does not fall in the rejection region, conclude that the sampling experiment does not provide

sufficient evidence to reject H0 at the α level of significance. [Generally, we will not “accept” the null hypothesis unless

the probability β of a Type II error has been calculated.

Example 8.4.1. Refer to the neurological response-time test setup in Example 8.2.2. The sample of 100 drug-injected rats

yielded the results (in seconds) in Data Set: DRUGRAT. At α = .01, use these data to conduct the test of hypothesis,

H0 : µ = 1.2 (Mean response time is 1.2 seconds)

Ha : µ 6= 1.2 (Mean response time is less than 1.2 or greater than 1.2 seconds

Solution

Based on the datd set: x̄ = 1.0517, s = .4982. Now we substitute these sample statistics into the test statistic and obtain
x̄−µ0 1.0517−1.2
zc = √s
= 0.4982
√
= −2.98
n 100

Rejection region: |z| > z 0.01

2
then |z| > 2.575.

Figure 8.9: Two-tailed rejection region: α = .01

The figure above shows taht z = −2.98 falls in the lower-tail rejection region, which consists of all values of z < −2.575.

Therefore, this sampling experiment provides sufficient evidence to reject H0 and conclude, at the α = 0.01 level of significance,

that the mean response time for drug-injected rats differs from the control mean of 1.2 seconds. It appears that the rats

receiving an injection of the drug have a mean response time that is less than 1.2 seconds.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 13

The following output is obtained by Minitab.

Figure 8.10: MINITAB lower-tailed test of mean LOS

Minitab: Stat → Basic Statistics → 1 − SampleZ

One or more Samples, each in a column → select Time

Known standard deviation: 0.4982 (calculate StDev prior to this step using Basic Statistic)

check “Perform hypothesis test”

Hypothesized mean: 1.2

→ options

confidence level: 99

Alternative hypothesis: Mean6= Hypothesized mean

→ ok → ok

Note

The level of α is determined before the sampling experiment is performed. If we decide that we are willing to tolerate a

1% Type I error rate, the result of the sampling experiment should have no effect on that decision. In general, the same data

should not be used both to set up and to conduct the test.

Question

Analyze the hypothesises using confidence interval.

8.5 Test of Hypothesis about a Population Mean: Student’s t-Statistic

Recall from Section 7.3 that when we are faced with making inferences about a population mean from the information in

a small sample so that two problems emerged. Therefore, we defined and used the t- statistic which follows t-distribution.

Therefore, as the test statistic of a small-sample test of a population mean, we use the t-statistic.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 14

Small-Sample Test of Hypothesis about µ Based on a Student’s t-Statistic

Test statistic:

x̄−µ0
tc = √s
n

Lower-Tailed Tests

H0 : µ = µ0

Ha : µ < µ0

Rejection region: t < −tα

p-value: P (t < tc )

Upper-Tailed Tests

H0 : µ = µ0

Ha : µ > µ0

Rejection region: t > tα

p-value: P (t > tc )

Two-Tailed Tests

H0 : µ = µ0

Ha : µ 6= µ0

Rejection region: |t| > t α2

p-value: if tc is positive 2P (t > tc ) if tc is negative 2P (t < tc )

Decision

Reject H0 if p − value < α or if test statistic (tc ) falls in rejection region ( the result is “statistically significant”) where:
α
P (t > tα ) = α, P (t > t α2 ) = 2

Conditions Required for a Valid Small-Sample Hypothesis Test for µ

1. A random sample is selected from the target population.

2. The population from which the sample is selected has a distribution that is approximately normal.

Example 8.5.1. A major car manufacturer wants to test a new engine to determine whether it meets new airpollution

standards. The mean emission µ of all engines of this type must be less than 20 parts per million of carbon. Ten engines are

manufactured for testing purposes, and the emission level of each is determined. The data (in parts per million) are listed in

Data Set: ENGINE. Do the data supply sufficient evidence to allow the manufacturer to conclude that this type of engine

meets the pollution standard? Assume that the manufacturer is willing to risk a Type I error with probability α = .01.
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 15

Solution

The elements of this small-sample one-tailed test are as follows:

H0 : µ = 20 (Mean emission level is 20 ppm.)

Ha : µ < 20 (Mean emission level is less than 20 ppm.)

x̄−µ0 17.57−20
Test statistic: tc = √s
= 2.95
√
= −2.60
n 10

Rejection region: For α = .01 and df = n - 1 = 9, the one-tailed rejection region (see Figure 8.10) is t < −t0.01 = −2.821.

Figure 8.11: A t-distribution with 9 df and the rejection region for Example 8.5.1.

Since the calculated t falls outside the rejection region (see Figure 8.10)), the manufacturer cannot reject H0 . There is

insufficient evidence to conclude that µ < 20 parts per million and that the new type of engine meets the pollution standard.

The same conclusion can be reached calculating the p-value.

p-value: P (t < −2.60) = 0.0143691

The p-value (0.0143691) is greater than α = 0.01. Thus, the two methods agree and we cannot reject H0 : µ = 20 in favor

of Ha : µ < 20.

The following output is obtained by Minitab.

Figure 8.12: MINITAB analysis of 10 emission levels

Minitab: Stat → Basic Statistics → 1 − Sample t

One or more Samples, each in a column → select E-LEVEL

CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 16

check “Perform hypothesis test”

Hypothesized mean: 20

→ options

confidence level: 99

Alternative hypothesis: Mean< Hypothesized mean

→ ok → ok

Question

Check whether the assumptions are met.

8.6 Large-Sample Test of Hypothesis about a Population Proportion

Inferences about population proportions (or percentages) are often made in the context of the probability p of “success” for

a binomial distribution. We saw how to use large samples from binomial distributions to form confidence intervals for p in

Section 7.4. We now consider tests of hypotheses about p. Recall that the sample proportion p̂ is really just the sample

mean of the outcomes of the individual binomial trials and, as such, is approximately normally distributed (for large samples)

according to the Central Limit Theorem. Thus, for large samples we can use the standard normal z as the test statistic:

Test statistic:

Sample proportion - Null hypothesized proportion (p̂−p0 ) (p̂−p0 )

zc = Standard deviation of sample proportion = σp̂ =√ p0 q0
n

Lower-Tailed Tests

H0 : p = p0

Ha : p < p0

Rejection region: z < −zα

p-value: P (z < zc )

Upper-Tailed Tests

H0 : p = p0

Ha : p > p0

Rejection region: z > zα

p-value: P (z > zc )

Two-Tailed Tests

H0 : p = p0

H1 : p 6= p0

Rejection region: |z| > z α2

CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 17

p-value: if zc is positive 2P (z > zc ) if zc is negative 2P (z < zc )

Decision

Reject H0 if p − value < α or if test statistic (zc ) falls in rejection region ( the result is “statistically significant”) where:
α
P (z > zα ) = α, P (z > z α2 ) = 2

Conditions Required for a Valid Large-Sample Hypothesis Test for p

1. A random sample is selected from a binomial population.

2. The sample size n is large. (This condition will be satisfied if both np0 ≥ 15 and nq0 ≥ 15.)

Example 8.6.1. Consider a method currently used by doctors to screen women for breast cancer. The method fails to detect

cancer in 20% of the women who actually have the disease. Suppose a new method has been developed that researchers hope

will detect cancer more accurately. This new method was used to screen a random sample of 140 women known to have

breast cancer. Of these, the new method failed to detect cancer in 12 women. Does this sample provide evidence that the

failure rate of the new method differs from the one currently in use? (α = 0.05)

We can now place the problem in the context of a test of hypothesis:

H0 : p = p0 (Cancer detection rate = .2)

Ha : p 6= p0 (Cancer detection rate differs from .2)

12
p̂ = 140 = .086, p0 = 0.2, q0 = 0.8 thus:

np0 = 140 × 0.2 = 28 and nq0 = 140 × 0.8 = 112

Considering the results above and that the sample was selected at random, we can conclude the conditions are met.
(p̂−p0 )
Test statistic: zc = √ p0 q0 =
√.086−.2
0.2×0.8
= −3.35
n 140

Rejection region: |z| > z α2 which is:

z > z0.025 = 1.96 or z < −z0.025 = −1.96

And depicted in figure 8.13:

You can see that the test statistic falls in rejection region. Therefore, we reject the null hypothesis, concluding at the

.05 level of significance that the true failure rate of the new method for detecting breast cancer differs from .20. The same

conclusion can be reached calculating the p-value.

p-value: 2P (z < −3.35) = 0.001

The p-value (0.001) is less than α = 0.05. Thus, the two methods agree and we reject H0 : p = p0 in favor of H0 : p 6= p0 .
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 18

Figure 8.13: Rejection region for breast cancer example

The following output is obtained by Minitab.

Figure 8.14: MINITAB analysis of breast cancer example

Minitab: Stat → Basic Statistics → 1Proportion

Summarized data

Number of events: 12

Number of trials: 140

check “Perform hypothesis test”

Hypothesized proportion: 0.2

→ options

confidence level: 95

Alternative hypothesis: Proportion6= Hypothesized proportion

Method: Normal approximation (select Exact for small samples)

→ ok → ok
CHAPTER 8. INFERENCES BASED ON A SINGLE SAMPLE: TESTS OF HYPOTHESIS 19

Small samples

Since most surveys and studies employ large samples, the large-sample testing procedure based on the normal (z) statistic

presented here will be appropriate for making inferences about a population proportion. However, in the case of small

samples (where either np0 or nq0 is less than 15), tests for a population proportion based on the z-statistic may not be

valid—especially when conducting one-tailed tests. A test of proportions that can be applied to small samples utilizes the

binomial, rather than the normal, distribution. These are called exact binomial tests due to the fact that the exact (rather

than approximate) p-value for the test is computed based on the binomial distribution.

The core content of the slides are from the textbook of this course;

STATISTICS (13th Edition)

by

JAMES McCLAVE and TERRY SINCICH