from the same subject. Some subjects showed generally
more beta-endorphins at both times than others, and we
wanted to eliminate this subject-to-subject variability
that had nothing to do with stress. In fact, there isn’t
much of a relationship between the two measures, but
we can’t fairly ignore it after the fact.7.19 The means for males and females were 2.73 and 2.79
respectively, producing a t 5 .485 on 90 df. We can-
not reject the null hypothesis that males and females are
equally satisfied. A matched-sample tis appropriate
because it would not seem reasonable to assume that the
sexual satisfaction of a husband is independent of that
of his wife.
7.21 The correlation between the scores of husbands and
wives was .334, which is significant, and which con-
firms the assumption that the scores would be related.
7.23 The important question is what would the sampling dis-
tribution of the mean (or differences between means)
look like, and with 91 pairs of scores that sampling
distribution would be substantially continuous with a
normal distribution of means.
7.25 t 5 2.545, p,.05, which tells us that the quality of life
was better for the intervention group.
7.27
7.29 (a) Null hypothesis: there is not a significant differ-
ence in test scores between those who have read
the passage and those who have not.
(b) Alternative hypothesis: there is a significant differ-
ence between the two conditions.
(c) t 5 8.89 on 43 dfif we pool the variances. This
difference is significant.
(d) We can conclude that students do better on this
test if they read the passage on which they are
going to answer questions.
7.31 Girls in the Control group lost an average of 0.45
pounds, while girls in the Cognitive Behavior Therapy
group gained 3.01 pounds. A ton two independent
groups 52 1.68 on 53 df, which is not significant. Cog-
nitive Behavior Therapy did not lead to significantly
greater weight gain. (Variances were homogeneous.)
7.33 If those means had actually come from independent
samples, we could not remove differences due to cou-
ples, and the resulting twould have been somewhat
smaller.
7.35 The correlation was fairly low.
7.37 (a) I would assume that the experimenters expected
that there would be more stories exhibiting posi-
tive parent-child relationships among the mothers
or children in the Normal group.
(b) The means were 3.55 and 2.1 for the normal and
schizophrenic groups, respectively, with t 5 2.66
on 38 df, which is significant. The experimental
hypothesis in (a) was supported.
CI.95=0.24...m...5.80.7.39 It is just as likely that having a schizophrenic child
might lead to deterioration in parent-child relationships.
Because we can’t assign children to groups at random,
we cannot speak confidently about causation.
7.41 Confidence limits on Exercise 7.40: CI.95 5 1.153
(2.131)(1.965 52 3.04 ,,5.34. Because the confi-
dence limits include 0, these results are in agreement
with the previous nonsignificant result.
7.43 t 5 2.134. Because the variances are very dissimilar,
we could run a conservative test by using the smaller of
n 1 – 1 and n 2 – 1 df, which would produce a nonsignifi-
cant result. We would have a nonsignificant result even
if we used the full n 11 n 2 – 2 df.
7.45 If you take the absolute differences between the obser-
vations and their group means and run a ttest compar-
ing the two groups on the absolute differences, you
obtain t 5 0.625. Squaring this you have F 5 0.391,
which makes it clear that Levene’s test in SPSS is oper-
ating on the absolute differences. (The tfor squared dif-
ferences would equal 0.213, which would give an Fof
0.045.)
7.47 Computer exercise
7.49 The effect size (d) was 0.62 using the standard deviation
of weights before therapy. This indicates a gain of
roughly 2 3 of a standard deviation over the course of
therapy.
7.51 (a) The scale of measurement is important because if
we rescaled the categories as 1, 2, 4, and 6, for ex-
ample, we would have quite different answers.
(b) The first exercise asks if there is a relationship
between the satisfaction of husbands and wives.
The second simply asks if males (husbands) are
more satisfied, on average, than females (wives).
(c) You could adapt the suggestion made in the text
about combining the ton independent groups and
the ton matched groups.
(d) I’m really not very comfortable with the ttest
because I am not pleased with the scale of meas-
urement. An alternative would be a ranked test, but
the number of ties is huge, and that probably wor-
ries me even more.Chapter 8
8.1 (a) 0.250 (b) 2.50 (c) .71
8.3 n 5 99, 126, and 169 (I have rounded up since N is al-
ways an integer.)
8.5 This is a graphic.
8.7 (a) For power 5 .50, 5 1.95 and n 515 .21 16
(b) For power 5 .80, 5 2.80 and n 5 31.36 32
8.9 d 5 .50, 5 1.46, power 5 .31
8.11 t( 52 1.98) is numerically equal to , although tis
calculated from statistics and dis calculated fromddd Ld L>m6Answers 741