What is Repeated Measures ANOVA?
Repeated measures ANOVA is a specialized statistical test used when you take multiple measurements from the same participants across different time points or conditions. It is the extension of the paired t-test to three or more measurement occasions.
When to Use Repeated Measures ANOVA
- Pre-test, post-test, and follow-up designs (same participants measured at multiple time points)
- Crossover trials where participants receive all treatments
- Within-subjects factorial designs
- Learning and development studies tracking change over time
Key Assumptions
- Normality: The dependent variable should be approximately normally distributed at each measurement point
- Sphericity: The variances of the differences between all pairs of measurement conditions should be equal. Tested with Mauchly's test. If violated, use Greenhouse-Geisser or Huynh-Feldt correction.
- No significant outliers at any time point
Step-by-Step Guide in SPSS
- Go to Analyze → General Linear Model → Repeated Measures
- In "Within-Subject Factor Name", type a name for your factor (e.g., "Time")
- Enter the number of levels (time points) in "Number of Levels"
- Click Add then Define
- Assign your measurement variables to the within-subjects factors
- Click Options and select: Descriptive statistics, Estimates of effect size, Observed power
- For post-hoc tests, click EM Means and select "Compare main effects"
- Click OK
Interpreting the Output
Key elements to report:
- Mauchly's Test: If p < .05, sphericity is violated; apply correction
- Tests of Within-Subjects Effects: F-statistic, degrees of freedom, p-value
- Partial Eta Squared (η²p): Effect size (small: .01, medium: .06, large: .14)
- Pairwise Comparisons: Which specific time points differ significantly
APA Reporting Example
"A one-way repeated measures ANOVA was conducted to evaluate the effect of time on [outcome measure]. The results showed a statistically significant effect of time, F(2, 58) = 12.45, p < .001, η²p = .30, indicating that scores changed significantly across the three time points."