The grouping information table highlights the significant
and nonsignificant comparisons. The table includes the following information:
· Method
– the
multiple comparison method used to construct the family of confidence
intervals from which the grouping table is generated.
· Comparisons
– each
grouping information table compares levels of one factor, or combined
levels of multiple factors.
· N
– the
sample size for each factor level.
· Mean
– fitted means
sorted in descending order.
· Grouping
– contains
columns of letters that group the factor levels. Levels that share a letter
are not significantly different. Conversely, if they do not share a letter,
the level means are significantly different.
To determine a likely range for the difference between two means, examine
the hypothesis test results or the interval plot for differences of means.
Example Output |

|
Grouping Information Using the Sidak Method and 95% Confidence
Subject N Mean Grouping
3 11 3.11333 A
4 9 2.84000 B
2 13 2.80667 B
1 12 2.15000 C
Means that do not share a letter are significantly different. |
Interpretation |

|
For the salary analysis, all pairwise comparisons were
requested for the Subject factor.
· Subject
3's mean (3.11333)
is significantly different from subjects 4, 2, and 1 because this subject
does not share any letters in the Grouping column. Teaching engineering
paid significantly higher mean salaries than any other subject in this
study.
· Subjects
4 and 2 share a grouping letter (B)
and, therefore, their means (2.80667
and 2.84000) are not significantly different from each other. However,
these two subjects are significantly different from subjects 3 and 1 because
they do not share letters with them.
· Subject
1 is significantly different from subjects 3, 4, and 2 because this subject
does not share any letters in the Grouping column. Teaching humanities
paid significantly lower mean salaries than any other subject in this
study.