the poor recall to be expected from passive studying. Good recall, whether it be lists of
words or of complex statistical concepts, requires active and “deep” processing of the ma-
terial, which is in turn facilitated by noting associations between the to-be-learned material
and other material that you already know. You have probably noticed that sitting in class
and dutifully recording everything that the instructor says doesn’t usually lead to the grades
that you think such effort deserves. Now you know a bit about why.11.5 Writing Up the Results
Reporting results for an analysis of variance is somewhat more complicated than reporting
the results of a ttest. This is because we not only want to indicate whether the overall Fis
significant, but we probably also want to make statements about the differences between
individual means. We won’t discuss tests on individual means until the next chapter, so this
example will be incomplete. We will come back to it in Chapter 12. An abbreviated version
of a statement about the results follows.
In a test of the hypothesis that memory depends upon the level of processing of the ma-
terial to be recalled, participants were divided into five groups of ten participants each.
The groups differed in the amount of processing of verbal material required by the in-
structions, varying from simply counting the letters in the words to be recalled to form-
ing mental images evoked by each word. After going through the list of 27 words three
times, participants were asked to recall as many items on the list as possible. A one-
way analysis of variance revealed that there were significant differences among the
means of the five groups (F(4,45) 5 9.08, p,.05).Visual inspection of the group
means revealed that the level of recall generally increased with the level of processing
required, as predicted by the theory. (Note: Further discussion of these differences will
have to wait until Chapter 12.)11.6 Computer Solutions
Most analyses of variance are now done using standard computer software, and Exhibit 11.1
contains examples of output from SPSS. Other statistical software will produce similar results.
In producing the SPSS printout that follows, I used the One-Wayselection from the
Compare Meansmenu.330 Chapter 11 Simple Analysis of Variance
Exhibit 11.1 SPSS One-Way Printout (continues)