Unit 10 · Factorial designs

How to run a mixed ANOVA in SPSS

Run a mixed (split-plot) ANOVA in SPSS with a within-subjects and a between-subjects factor, and interpret the interaction.

Free tutorialPro practiceAbout 5 minutes6 practice questions

Quick answer: How do you run a mixed ANOVA in SPSS?

  1. Click Analyze › General Linear Model › Repeated Measures and define the time factor.
  2. Move the repeated variables into Within-Subjects Variables.
  3. Move the group variable into Between-Subjects Factor(s), then click OK.
  4. Read the time × group interaction.

Report it in APA style: The therapy group improved more than the control group, F(1, 58) = 18.20, p < .001, ηp² = .24.

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

What is a mixed ANOVA?

A mixed ANOVA combines a within-subjects factor (people measured more than once) with a between-subjects factor (different groups). The classic example is a trial measuring anxiety before and after in a therapy group and a control group.

When to use it

  • Pre-post designs with a control group.
  • Any design with one repeated factor and one grouping factor.

How it works

You get a main effect of time, a main effect of group, and the time × group interaction. In intervention studies, the interaction is usually the key result: it tests whether the groups changed differently over time. The within-subjects effects appear in one table and the between-subjects effect in another.

Assumptions

  • Scale outcome, roughly normal.
  • Sphericity for the within factor (if it has three or more levels).
  • Equal variances across groups (Levene’s test) and equal covariance matrices (Box’s M).

Worked example

Anxiety fell more in the therapy group (24.1 to 16.3) than in the control group (23.8 to 22.6), time × group F(1, 58) = 18.20, p < .001, ηp² = .24.

Key terms

Between-subjects factor
Different people at each level.
Within-subjects factor
The same people at every level.
Box’s M
Tests equality of covariance matrices across groups; very sensitive, often judged at p < .001.

Common mistakes

  • Focusing on the main effect of time when the question is whether therapy beat control.
  • Analysing each group separately with paired t-tests instead of testing the interaction.

How to do it in SPSS

Mixed ANOVA set-up

A mixed ANOVA combines a repeated factor (before and after) with a between-groups factor (therapy vs control).

Repeated Measures
Variables
Participant ID [id]
Age in years [age]
Within-Subjects Variables (time):1
anx_pre(1)
anx_post(2)
Between-Subjects Factor(s):2
group
Covariates:
Model…Contrasts…Plots…3Post Hoc…EM Means…Save…Options…4
OKPasteResetCancelHelp
  1. 1
    Within-Subjects VariablesThe repeated measurements, in time order.
  2. 2
    Between-Subjects Factor(s)The variable that splits people into separate groups.
  3. 3
    Plots…time on the Horizontal Axis, group in Separate Lines. The chart shows whether the groups changed differently.
  4. 4
    Options…Tick Descriptive statistics, Estimates of effect size and Homogeneity tests (Levene’s and Box’s M).

Reading mixed ANOVA output

The within-subjects table holds the effects involving time. The between-subjects table holds the group effect.

Output1 [Document1] – SPSS Statistics Viewer
FileEditViewDataTransformAnalyzeGraphsUtilitiesExtensionsWindowHelp
Output
Log
General Linear Model
Tests of Within-Subjects Effects
Tests of Between-Subjects Effects
Tests of Within-Subjects Effects
SourceType III Sum
of Squares
dfMean
Square
FSig.Partial Eta
Squared
time1305.041305.0424.60.000.298
time * group2225.681225.6818.20.000.239
Error(time)719.205812.40
Tests of Between-Subjects Effects
SourceType III Sum
of Squares
dfMean
Square
FSig.Partial Eta
Squared
group3124.001124.003.10.084.051
Error2320.005840.00

Means: therapy 24.1 → 16.3 · control 23.8 → 22.6 (n = 30 each)

  1. 1
    Main effect of timeAveraged over both groups, anxiety fell from pre to post.
  2. 2
    time * groupThe key result in most studies: the therapy group improved more than the control group, F(1, 58) = 18.20, p < .001, ηp² = .24.
  3. 3
    Main effect of groupAveraged over both time points, the groups didn’t differ significantly (p = .084). That’s expected when they started at the same level.

With only two time points, sphericity can’t be violated, so Mauchly’s test shows a dot instead of a p-value.

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 anx_pre anx_post BY group
  /WSFACTOR=time 2 Polynomial
  /METHOD=SSTYPE(3)
  /PLOT=PROFILE(time*group)
  /PRINT=DESCRIPTIVE ETASQ HOMOGENEITY
  /WSDESIGN=time
  /DESIGN=group.

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.

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 Factorial designs