7.21 For the data in Exercise 7.19, create a scatterplot and calculate the correlation between
husband’s and wife’s sexual satisfaction. How does this amplify what we have learned
from the analysis in Exercise 7.19. (I do not discuss scatterplots and correlation until
Chapter 9, but a quick glance at Chapter 9 should suffice if you have difficulty. SPSS will
easily do the calculation.)
7.22 Construct 95% confidence limits on the true mean difference between the Sexual Satisfac-
tion scores in Exercise 7.19, and interpret them with respect to the data.
7.23 Some would object that the data in Exercise 7.19 are clearly discrete, if not ordinal, and that
it is inappropriate to run a ttest on them. Can you think what might be a counter argument?
(This is not an easy question, and I really asked it mostly to make the point that there could
be controversy here.)
7.24 Give an example of an experiment in which using related samples would be ill-advised be-
cause taking one measurement might influence another measurement.
7.25 Sullivan and Bybee (1999) reported on an intervention program for women with abusive
partners. The study involved a 10-week intervention program and a three-year follow-up,
and used an experimental (intervention) and control group. At the end of the 10-week inter-
vention period the mean quality of life score for the intervention group was 5.03 with a stan-
dard deviation of 1.01 and a sample size of 135. For the control group the mean was 4.61
with a standard deviation of 1.13 and a sample size of 130. Do these data indicate that the
intervention was successful in terms of the quality of life measure?
7.26 In Exercise 7.25 Calculate a confidence interval for the difference in group means. Then cal-
culate a d-family measure of effect size for that difference.
7.27 Another way to investigate the effectiveness of the intervention described in Exercise 7.25
would be to note that the mean quality of life score before the intervention was 4.47 with a
standard deviation of 1.18. The quality of life score was 5.03 after the intervention with a
standard deviation of 1.01. The sample size was 135 at each time. What do these data tell
you about the effect of the intervention? (Note: You don’t have the difference scores, but as-
sume that the standard deviation of difference scores was 1.30.)
7.28 For the control condition for the experiment in Exercise 7.25 the beginning and 10-week
means were 4.32 and 4.61 with standard deviations of 0.98 and 1.13, respectively. The sam-
ple size was 130. Using the data from this group and the intervention group, plot the change
in pre- to post-test scores for the two groups and interpret what you see.
7.29 In the study referred to in Exercise 7.13, Katz et al. (1990) compared the performance on
SAT items of a group of 17 students who were answering questions about a passage after
having read the passage with the performance of a group of 28 students who had not seen
the passage. The mean and standard deviation for the first group were 69.6 and 10.6,
whereas for the second group they were 46.6 and 6.8.
a. What is the null hypothesis?
b. What is the alternative hypothesis?
c. Run the appropriatet test.
d. Interpret the results.
7.30 Many mothers experience a sense of depression shortly after the birth of a child. Design a
study to examine postpartum depression and, from material in this chapter, tell how you
would estimate the mean increase in depression.
7.31 In Exercise 7.25, we saw data from Everitt that showed that girls receiving cognitive behav-
ior therapy gained weight over the course of that therapy. However, it is possible that they
just gained weight because they got older. One way to control for this is to look at the
amount of weight gained by the cognitive therapy group (n 5 29) in contrast with the
amount gained by girls in a Control group (n 5 26), who received no therapy. The data on
weight gain for the two groups is shown below.220 Chapter 7 Hypothesis Tests Applied to Means