Multiple Comparison Procedures - Analysis of Variance Designs - page 144

Figure 7.18 The Least Squares screen is now configured for the Dunnett comparisons.

7.20 COMMUNICATING THE SIMPLE CONTRAST RESULTS

In reporting the results of the Simple contrasts, we would probably con-
struct a small table (call it Table 1 for the sake of the write-up) of means
and standard deviations rather than reporting them in the text. Given
that, one way to report the results is as follows:

The amount of preparation for the SAT in which students engaged appeared
to significantly affect their performance on the test, F (4, 30)

= 43 . 47, p <

2

. 85. The means and standard deviations of the groups are presented

in Table 1. A post-ANOVA set of simple contrasts compared the group with
zero months of study to each of the other groups. The results indicated that
all of the groups differed significantly from the comparison group; thus, some
amount of study resulted in better performance on the SAT than no study at
all.

05, η = .

Note that these Simple contrasts tell only part of the story revealed
by the Tukey tests. This is because the Tukey procedure examined all
possible pairwise comparisons and we were able to select which aspect
of the comparisons to report. With preset contrasts, you are examin-
ing only a particular feature of the results, and the conclusions you are
permitted to draw are therefore limited to what you learned about that
feature. This may give you some insight as to why many researchers are