Scribd
Upload
22 views3 pages

Module 8 Part 6

Uploaded by

Fatemeh Yarahmadi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
22 views3 pages

Module 8 Part 6

Uploaded by

Fatemeh Yarahmadi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Hypothesis Tests for Paired Samples

INTRODUCTION

In a previous lesson, we tested claims about a population mean. In this lesson we learn how to test a claim
about a population mean of paired differences. We will see that the process for testing claims about a
population mean of paired differences is the same as the process for testing claims about a population
mean.

Assessing a Psychological Treatment

Psychologists often argue over the best way to help people with anxiety. Two possible treatments are
counseling sessions and pharmaceutical treatments (in other words, medicine that helps to reduce anxiety).

Suppose that a team of psychologists has recommended that people who experience anxiety attacks should
receive regular counseling. The psychologists claim that the counseling sessions provide a therapeutic, or
healing, experience. The psychologists promote this type of therapy rather than pharmaceutical treatments.

To test the effectiveness of counseling, a researcher randomly selects six people who experience anxiety
attacks. These six people are interested in the counseling treatment. The researcher gives each person an
anxiety test before and after the person has a counseling session with a therapist. Higher scores on the test
indicate a higher level of anxiety in the person taking the test.

The test scores of the six people are shown in the table below. At the 5% level of significance, can we
confirm that the counseling sessions reduce anxiety? Assume the population of test score differences is
normally distributed.

Person 1 2 3 4 5 6
Pre-Counseling Score 66 71 80 78 85 90
Post-Counseling Score 64 65 79 81 81 91

Step 1: Determine the Hypotheses

1 Are the data paired? Explain why you think they are or are not paired.

ȋͳȌ Answer: Yes, there are two observations for each person in the sample.

Statway® Version 3.1, © 2018 WestEd. All rights reserved.


2 If counseling does not make a difference in anxiety levels, what should the mean of the differences in
test scores be?

3 If counseling helps reduce anxiety and does make a difference, would we expect the difference to be
positive or negative? (Hint: The difference is the post-counseling score minus the pre-counseling score.)
Explain your answer.

4 State the null and alternative hypotheses.

H0:

Ha:

5 Is this a right-tailed, left-tailed, or two-tailed test?

Step 2: Collect the Data

6 Are the criteria for approximate normality met for the sampling distribution of means of sample
differences?

7 Determine the following:

● The mean of the sample differences,


● The standard deviation of the sample differences, and
● The number of data pairs in the sample, n.

8 Is the mean of sample differences consistent with the alternative hypothesis?


ȋʹȌ Answer: The mean of the differences should be zero.

ȋ͵Ȍ Answer: The mean of differences should be negative, because we expect the scores to be lower at
post-counseling than at pre-counseling.

ȋͶȌ Answer: H0 : μ = 0, Ha: μ < 0.

ȋͷȌ Answer: A left-tailed test.

ȋ͸Ȍ Answer: Yes. While the number of paired differences is not greater than 30, we have assumed
that they are members of a normal population.

ȋ͹Ȍ Answer: The sample mean is −1.5, the sample standard deviation is approximately 3.27, and the
sample size is 6.

ȋͺȌ Answer: Yes.


Step 3: Assess the Evidence

9 Calculate the test statistic for the observed mean of sample differences.

10 Sketch the T-distribution. Identify the position of the observed test statistic. Shade the area that
represents the P-value.

11 Use the test statistic to determine the P-value.

Step 4: State a Conclusion

12 How does the P-value compare to the significance level? (Hint: Is it greater than, less than, or equal to
the significance level?) Should we reject or fail to reject the null hypothesis?

13 What can we conclude about the alternative hypothesis?

14 State a conclusion in context. (Hint: Provide the significance level, write what you can conclude about
the alternative hypothesis, and put this information in the context of the psychological treatment.)

ȋͻȌ Answer: T = −1.12.

ȋͳͲȌ Answer: Students should sketch a bell curve with the test statistic in the left tail. The area to the
left of T = −1.12 should be shaded.

ȋͳͳȌ Answer: 0.160.

ȋͳʹȌ Answer: P-value > 0.05. We fail to reject the null hypothesis.

ȋͳ͵Ȍ Answer: There is not enough evidence to support the alternative hypothesis.

ȋͳͶȌ Answer: At the 5% level of significance, there is not sufficient evidence to conclude that the
mean difference is not 0. The evidence does not allow us to conclude that average anxiety levels
are lower following the counseling sessions.

You might also like