One-Way Repeated Measures ANOVA Guide
What This Test Measures
A one-way repeated measures ANOVA compares several related means. The same subjects are measured under each condition. This design removes much subject noise. It is common in learning tests, medical follow ups, product ratings, and timed performance studies.
The test separates total variation into three parts. Condition variation shows change across the repeated levels. Subject variation shows stable differences between people or units. Error variation is the remaining unexplained part. The F ratio compares condition mean square with error mean square.
Why The Calculator Helps
Manual repeated analysis is slow. Each subject has a row. Each condition has a column. The calculator reads that matrix and builds the full ANOVA table. It also reports means, standard deviations, effect sizes, and paired comparisons. These extra values help you move beyond one p value.
The result is most useful when the data are complete. Every subject should have a score for every condition. Scores should be measured on a meaningful numeric scale. Large outliers can distort means and sums of squares. Review the example table before entering your own values.
Reading The Output
Start with the descriptive table. Check whether condition means move in the expected direction. Then read the ANOVA table. A small p value means at least one condition mean differs. Partial eta squared shows how much explainable repeated variation belongs to the condition effect.
Sphericity matters because repeated scores are correlated. When the assumption looks weak, use Greenhouse-Geisser or Huynh-Feldt adjusted p values. These corrections reduce the degrees of freedom. They make the test more cautious.
Practical Notes
ANOVA tells you that a difference exists. It does not show which pairs differ. Use the paired comparison table for that question. Bonferroni adjusted p values are conservative. They are useful when many condition pairs are checked.
Report the number of subjects, the number of conditions, F, degrees of freedom, p value, and effect size. Add the correction used when sphericity is doubtful. Keep raw data stored safely. Reproducible tables make the analysis easier to audit, share, and explain.
Use charts only after checking assumptions. Clear notes help readers understand design, repeated levels, exclusions, and the chosen alpha before conclusions are drawn.