Repeated-measures and mixed ANOVA
Compares means measured repeatedly on the same people (time points or conditions), optionally with between-subjects groups (mixed design). Gives sphericity tests, corrected F tests, multivariate tests and pairwise comparisons.
Menu: Analyze → General Linear Model → Repeated Measures…
When should I use it?
- Does blood pressure change across four visits?
- Does the change over time differ between a treatment and a control group (time × group interaction)?
What are the assumptions?
- Each row is one participant with the repeated scores in separate columns.
- Sphericity: equal variances of all pairwise differences (Mauchly's test); if violated, use the Greenhouse-Geisser or Huynh-Feldt rows, or the multivariate tests.
- Roughly normal differences; for mixed designs, equal covariance matrices across groups.
How do I run it in StatLab?
- Name the within-subject factor (e.g. time) and type its number of levels; add a second factor if the design is crossed (e.g. 2 × 3).
- Move the repeated variables in order — with two factors the last factor changes fastest (a1b1, a1b2, a1b3, a2b1…).
- Add between-subjects factors for a mixed design.
- Keep Bonferroni comparisons under EM Means.
How do I read the output?
| Output | Meaning |
|---|---|
| Mauchly's Test of Sphericity | Sig. < .05 means sphericity is violated; the epsilon columns give the corrections. |
| Tests of Within-Subjects Effects | F for the within factor and its interactions, with Sphericity Assumed, Greenhouse-Geisser, Huynh-Feldt and Lower-bound rows. |
| Multivariate Tests | An alternative that does not need sphericity. |
| Tests of Within-Subjects Contrasts | Linear, quadratic… trends across ordered levels. |
| Tests of Between-Subjects Effects | Group effect on the average of the repeated measures. |
| Pairwise Comparisons | Which levels differ, Bonferroni adjusted. |
Worked example
On the practice survey dataset (File → Sample dataset), this command — which the dialog's Paste button also produces — runs the procedure:
PROC glmRM /wname=time /measure=MEASURE_1 /deps=(pre post) /factors=(group) /covs=() /compare=bonferroni /desc=YES /eta=YES /power=NO.
Output tables: Within-Subjects Factors · Between-Subjects Factors · Descriptive Statistics · Multivariate Tests · Mauchly's Test of Sphericity · Tests of Within-Subjects Effects · Tests of Within-Subjects Contrasts · Tests of Between-Subjects Effects · time · Pairwise Comparisons.
How would I report it in APA style?
A mixed repeated-measures ANOVA was conducted (N = 150), with Programme group as between-subjects factor. time: F(1, 147) = 38.32, p < .001, partial η² = .21; time * group: F(2, 147) = 0.49, p = .616, partial η² = .01; group: F(2, 147) = 0.02, p = .980, partial η² = .00.
Computed by StatLab's validated procedure code when this page was built. The example only illustrates the steps; choose variables that fit your own research question.
What are common mistakes?
- Entering the repeated variables in the wrong order.
- Reporting the Sphericity Assumed row when Mauchly's test is significant.
- Analysing long-format data (one row per measurement) — restructure to one row per person first.