If a term is statistically significant in Fit General Linear Model or Fit Mixed Effects Model, you know that some pairs of group means are different, but you do not know which pairs differ. Use the grouping information table to determine whether the mean differences between specific pairs of groups are statistically significant. Groups that do not share a letter are significantly different.
The number of comparisons that Minitab displays depends on whether you chose Pairwise or With a control when completing the dialog box. In either case, assessing the statistical significance of the differences remains the same.
For more information on comparison methods, go to Using multiple comparisons to assess the practical and statistical significance.
Subject | N | Mean | Grouping | ||
---|---|---|---|---|---|
3 | 11 | 3.11333 | A | ||
4 | 9 | 2.84000 | B | ||
2 | 13 | 2.80667 | B | ||
1 | 12 | 2.15000 | C |
In these results, the table shows that only Subject 2 and 4 share a letter. The difference between these two groups is not statistically significant. All of the other subjects do not share a letter, which indicates that those differences are statistically significant.