Exercises 221Control Cognitive Therapy
2 0.5 3.3 1.7 2 9.1
2 9.3 11.3 0.7 2.1
2 5.4 0.0 2 0.1 2 1.4
12.3 2 1.0 2 0.7 1.4
2 2.0 2 10.6 2 3.5 2 0.3
2 10.2 2 4.6 14.9 2 3.7
2 12.2 2 6.7 3.5 2 0.8
11.6 2.8 17.1 2.4
2 7.1 0.3 2 7.6 12.6
6.2 1.8 1.6 1.9
2 0.2 3.7 11.7 3.9
2 9.2 15.9 6.1 0.1
8.3 2 10.2 1.1 15.4
2 4.0 2 0.7
20.9Mean 2 0.45 3.01
St Dev. 7.99 7.31
Variance 63.82 53.41
Run the appropriate test to compare the group means. What would you conclude?7.32 Calculate the confidence interval on m 1 2 m 2 for the data in Exercise 7.31.
7.33 In Exercise 7.19 we saw pairs of observations on sexual satisfaction for husbands and wives.
Suppose that those data had actually come from unrelated males and females, such that the
data are no longer paired. What effect do you expect this to have on the analysis?
7.34 Run the appropriate ttest on the data in 7.19 assuming that the observations are independ-
ent. What would you conclude?
7.35 Why isn’t the difference between the results in 7.34 and 7.19 greater than it is?
7.36 What is the role of random assignment in Everitt’s anorexia study referred to in Exercise
7.31, and under what conditions might we find it difficult to carry out random assignment?
7.37 The Thematic Apperception Test presents subjects with ambiguous pictures and asks them
to tell a story about them. These stories can be scored in any number of ways. Werner,
Stabenau, and Pollin (1970) asked mothers of 20 Normal and 20 Schizophrenic children to
complete the TAT, and scored for the number of stories (out of 10) that exhibited a positive
parent-child relationship. The data follow:
Normal 8463144642
Schizophrenic2113272131
Normal 2114332634
Schizophrenic0242330122
a. What would you assume to be the experimental hypothesis behind this study?
b. What would you conclude with respect to that hypothesis?
7.38 In Exercise 7.37, why might it be smart to look at the variances of the two groups?
7.39 In Exercise 7.37, a significant difference might lead someone to suggest that poor parent-child
relationships are the cause of schizophrenia. Why might this be a troublesome conclusion?
7.40 Much has been made of the concept of experimenter bias, which refers to the fact that even
the most conscientious experimenters tend to collect data that come out in the desired direc-
tion (they see what they want to see). Suppose we use students as experimenters. All the ex-
perimenters are told that subjects will be given caffeine before the experiment, but one-half
of the experimenters are told that we expect caffeine to lead to good performance and one-
half are told that we expect it to lead to poor performance. The dependent variable is the