Lecture 6
Lecture 6
• Any scores in the sample that differ from the sample mean are believed
to include what statisticians call error.
One –way ANOVA…
An example of within-group error.
One –way ANOVA…
Illustrations of between-groups and within-groups
deviations.
Post-Hoc Tests
• If we are comparing three group means, we might find a relatively
large average difference between these group means even if two of
the three group means are identical. Therefore, a statistically
significant F value tells us only that somewhere there is a
meaningful difference between my group means. But it does not
tell us which groups differ from each other significantly. To do this,
we must conduct post-hoc tests.
• There are a variety of post-hoc tests available. Some are more
conservative, making it more difficult to find statistically significant
differences between groups, whereas others are more liberal.
• All post-hoc tests use the same basic principle. They allow you to
compare each group mean to each other group mean and
determine if they are significantly different while controlling for
the number of group comparisons being made
Post-Hoc Tests…
• The difference between the various types of post-hoc tests is what
each test uses for the standard error.
• For the purposes of demonstration, we will consider the Tukey HSD
(HSD stands for Honestly Significantly Different) post-hoc test.
• This is a fairly liberal test, meaning that it is more likely to produce
statistically significant differences than some other tests (e.g., the
Scheffe).
• The Tukey test compares each group mean to each other group mean
by using the familiar formula described for t tests. Specifically, it is the
mean of one group minus the mean of a second group divided by the
standard error
Factorial Analysis of Variance
16