Purpose of the statistic

The one-way, repeated measures analysis of variance is exactly analogous to the one-way between subjects ANOVA except that the groups contain either the same subjects or individuals who have been explicitly matched in some way. It is used to ascertain how likely these within subject mean differences would be to occur by chance alone. Studies that might employ such a design include the multiple (i.e., three or more times) measurement of a single group of individuals across time (e.g., adding a long term follow-up assessment to a single group, pretest-posttest design) or, less commonly, a situation in which the same group of individuals is exposed to three or more different conditions.

A within subject design is extremely efficient in comparison to one which employs different subjects in each group, requiring far fewer subjects when even a moderate correlation can be obtained among its repeated observations. A one-way, repeated measures (RM) ANOVA, then, is used when:

1. there is a single, independent variable which is defined as group membership in three or more groups or as three or more separate points in time (recalling that if only two groups are involved a paired t-test can be employed, which is inferentially identical to a two-group RM ANOVA),

2. the dependent variable is measured in such a way that it can be described by a mean (i.e., it is continuous in nature and not categorical),

3. […]