Comparing more than two means: one factor ANOVA with independent samples. Multiple comparison testing with the Newman-Keuls test - Improving the User Experience through Practical Data Analytics - page 154

FIGURE 6.10
Post Hoc dialog box; SPSS.
6.7 APPLICATION OF THE S-N-K TO THIS RESULT
You'll remember our F-test results were highly signiicant with a p -value of 0.000 (see
Figure 6.8 ). Now, we apply the S-N-K test to this result. After clicking on Post Hoc,
obtaining Figure 6.10 , we then click the S-N-K box (see dashed arrow in Figure 6.10 ),
and then click “Continue” (bent arrow in Figure 6.10 ). This brings us back to Figure 6.9 ,
and we simply click on “OK.” We will get the output in Figure 6.8 , but now we will also
get the output representing the S-N-K test. The latter is shown in Figure 6.11 .
Note that we can identify that the result is, indeed, for the S-N-K test (see arrow
in Figure 6.11 ). We now discuss the interpretation of Figure 6.11 .
6.8 DISCUSSION OF THE RESULT
Figure 6.11 displays a table called “Homogeneous Subsets.” Basically, this table is telling
us that there are three sets of means that are signiicantly different from one another. First,
the true (sophistication) mean of age-group 1 is judged as different from (lower than)