Repeated Measures ANOVA
Repeated measures ANOVA online with sphericity check and Greenhouse Geisser correction.
When to use it
Use when the same subjects are measured under multiple conditions or at multiple time points (within-subject design). The most efficient design when individual differences are large because each subject acts as their own control.
When NOT to use it
Do not use when missing data are common; consider a linear mixed-effects model instead. Do not use when sphericity is heavily violated and conditions are strongly ordered (use multilevel models or MANOVA). For complex multi-factor within-by-between designs, mixed models are more flexible.
What you get
Within-subject ANOVA table with sources Subject (random), Condition, Error; Mauchly's test of sphericity (W and chi-square); Greenhouse-Geisser and Huynh-Feldt corrected df and p when sphericity is violated; partial eta-squared effect size; per-condition means with SE; pairwise contrasts with Bonferroni or Tukey adjustment.
How to interpret the output
If Mauchly's W has p above 0.05, sphericity is acceptable and the uncorrected F applies. If p is at or below 0.05, use the Greenhouse-Geisser-corrected df (more conservative) or Huynh-Feldt (slightly less conservative when GG epsilon is at least 0.75). Partial eta-squared above 0.14 is conventionally a large effect (Cohen 1988).
Common pitfalls
- Reporting the uncorrected F when Mauchly is significant.
- Confusing within-subject and between-subject sources.
- Listwise deletion of subjects with any missing time point (use a mixed model).
- Treating Mauchly as a perfect test; with small n it has low power, so always report both corrected and uncorrected p.
Try it in StatVeda
The Repeated Measures ANOVA engine runs entirely in the browser. No signup, no install, no data sent to a server. Paste your data, hit Run, copy the output.
Open Repeated Measures ANOVAReferences
- Greenhouse, S. W. and Geisser, S. (1959). On methods in the analysis of profile data. Psychometrika, 24(2), 95 to 112.
- Huynh, H. and Feldt, L. S. (1976). Estimation of the Box correction for degrees of freedom from sample data in randomized block and split-plot designs. Journal of Educational Statistics, 1(1), 69 to 82.
- Tabachnick, B. G. and Fidell, L. S. (2013). Using Multivariate Statistics, 6th edition. Pearson, Boston.