Unit 9 · Same people, several times

How to run a repeated-measures ANOVA in SPSS

Run a one-way repeated-measures ANOVA in SPSS, check Mauchly’s test, use Greenhouse-Geisser and report partial eta squared.

Free tutorialPro practiceAbout 5 minutes6 practice questions

Quick answer: How do you run a repeated-measures ANOVA in SPSS?

  1. Click Analyze › General Linear Model › Repeated Measures.
  2. Name the within-subjects factor, enter the number of levels, click Add and then Define.
  3. Move the repeated variables into the slots in order.
  4. Under Options… tick Estimates of effect size; then read Mauchly’s test and the within-subjects table.

Report it in APA style: Recall changed over time, F(2, 48) = 38.29, p < .001, ηp² = .61.

Last updated November 2026 · Not sure this is the right test? Try the Which statistical test? tool.

What is a repeated-measures ANOVA?

A repeated-measures ANOVA compares three or more means from the same people, such as memory recall on day 1, week 1 and month 1. It extends the paired t-test to more than two conditions.

When to use it

  • The same participants measured under three or more conditions or time points.
  • A Scale outcome.
  • If data are ordinal or clearly non-normal, use the Friedman test.

How it works

Because the same people appear in every condition, differences between people can be removed from the error term, making the test more powerful. The key extra assumption is sphericity: the variances of the differences between every pair of conditions should be similar. Mauchly’s test checks it. If it’s violated, the Greenhouse-Geisser correction reduces the degrees of freedom (multiplying them by epsilon) to give a more cautious p-value. Pairwise comparisons with a Bonferroni correction show which conditions differ.

Assumptions

  • Scale outcome, roughly normal in each condition.
  • Sphericity (only relevant with three or more conditions).
  • Independent participants.

Worked example

Recall declined over time, F(2, 48) = 38.29, p < .001, ηp² = .61, and all three time points differed (Bonferroni, all p ≤ .002).

Key terms

Within-subjects factor
A factor where every person is measured at every level.
Sphericity
Equal variances of the differences between conditions.
Greenhouse-Geisser
A correction for violated sphericity.
Partial eta squared (ηp²)
Effect size: proportion of variance explained, excluding other effects.

Common mistakes

  • Ignoring a significant Mauchly’s test.
  • Running separate paired t-tests without correction.
  • Forgetting order effects: counterbalance where possible.

How to do it in SPSS

Define the within-subjects factor

Repeated-measures ANOVA compares three or more scores from the same people. First tell SPSS what the repeated factor is called and how many levels it has.

Repeated Measures Define Factor(s)
Within-Subject Factor Name:1
time
Number of Levels:2
3
AddChangeRemove
Factors
time(3)
Measure Name:
recall
Define3ResetCancelHelp
  1. 1
    Factor nameA name for the thing that changes, such as time or condition. It can’t match an existing variable name.
  2. 2
    Number of levelsHow many measurements each person has: here Day 1, Week 1 and Month 1.
  3. 3
    DefineOpens the main dialog, where you drag recall_d1, recall_w1 and recall_m1 into the slots in order.

Reading repeated-measures output

SPSS prints many tables. Check sphericity first, then read the matching row of the within-subjects table.

Output1 [Document1] – SPSS Statistics Viewer
FileEditViewDataTransformAnalyzeGraphsUtilitiesExtensionsWindowHelp
Output
Log
General Linear Model
Mauchly's Test of Sphericity
Tests of Within-Subjects Effects
Pairwise Comparisons
Mauchly's Test of Sphericity
EffectMauchly's WApprox.
Chi-Square
dfSig.Greenhouse-
Geisser ε
time.8124.792.0911.842
Tests of Within-Subjects Effects
SourceType III Sum
of Squares
dfMean
Square
FSig.Partial Eta
Squared
timeSphericity Assumed2422.002211.0038.293.000.6154
Greenhouse-Geisser422.001.684250.5938.29.000.615
Error(time)Sphericity Assumed264.48485.51
Greenhouse-Geisser264.4840.426.54

Means: Day 1 = 18.40 · Week 1 = 15.20 · Month 1 = 12.60 (N = 25)

Pairwise Comparisons (Bonferroni)
(I) time(J) timeMean Difference (I−J)Std. ErrorSig.5
123.20*.62.000
135.80*.71.000
232.60*.66.002

*. The mean difference is significant at the .05 level.

  1. 1
    Mauchly's testTests sphericity (equal variances of the differences). Sig. .091 is above .05, so the assumption holds.
  2. 2
    Which rowSphericity holds, so read Sphericity Assumed. If Mauchly’s were significant, read Greenhouse-Geisser instead.
  3. 3
    FRecall changed significantly over time, F(2, 48) = 38.29, p < .001.
  4. 4
    Partial eta squaredηp² = .61: time explains a large share of the variation.
  5. 5
    Pairwise comparisonsFrom Options › Compare main effects with Bonferroni. All three time points differ.

The SPSS syntax

Prefer typing to clicking? Open File › New › Syntax, paste this in, select it and choose Run › Selection. Every dialog's Paste button writes the same kind of code.

GLM recall_d1 recall_w1 recall_m1
  /WSFACTOR=time 3 Polynomial
  /MEASURE=recall
  /METHOD=SSTYPE(3)
  /EMMEANS=TABLES(time) COMPARE ADJ(BONFERRONI)
  /PRINT=DESCRIPTIVE ETASQ
  /WSDESIGN=time.

Practice data

Download the file used in this lesson and run it yourself in SPSS. Your results will closely match the screens above, though a few numbers may differ slightly.

  • recall.savWords recalled at day 1, week 1 and month 1 (N = 25)

Now practise it interactively →

Reading is only half of it. Tap through the real SPSS menus, read the output and pick the right write-up in 6 quick questions, with instant feedback on every answer.

Practice for this lesson is part of Pro, with every unit, unlimited hearts and saved progress.

More in Same people, several times