I am evaluating pre and post test data for 2 groups using repeated measures ANOVA. Given that I am missing a few data cells, the final n is reduced in the analysis. I'd like to know if I can run separate repeated measures ANOVA for each subtest (or combinations that have the same number of full data sets) rather than include them all in the same model to maximize the data analyzed. If I ran separate ANOVAs, are there implications for my interpretation of the output? Thank you for any advice.


You're better off doing something that allows you to do the equivalent analysis, but doesn't get upset about the missing data (a structural equation model or a multilevel model), or doing imputation first to fill in the missings.

Perhaps tell us more about your model, your data, and what program you're using to analyze it.

| cite | improve this answer | |
  • $\begingroup$ Thanks for the response. So this study explores the outcome for math ability. We used many measures of math that sample slightly different aspects of math ability (e.g., calculations, timed calculations, etc.). (The measures are not consistently correlated - i.e., 2 out of the set are significant.) Because I care more about whether each measure independently changed, I thought I could preserve the group size if I ran the repeated measures ANOVA for each measure alone as opposed to putting them all into the same model and losing participant data sets. I am using SPSS for the analysis. $\endgroup$ – redracer Mar 30 '13 at 20:09
  • $\begingroup$ What are the repeated measures? Have you tested people twice? $\endgroup$ – Jeremy Miles Mar 30 '13 at 21:48
  • $\begingroup$ Yes, thanks for clarifying. Each person was tested twice (pre and post) on the same standardized measures (e.g., WJ Math Fluency). $\endgroup$ – redracer Mar 30 '13 at 22:37
  • $\begingroup$ I would do each measure alone - you're not interested in the differences between the measures (or are you?). $\endgroup$ – Jeremy Miles Mar 30 '13 at 22:43
  • $\begingroup$ I am interested in each measure independently, so running the analysis with each measure alone would be a good solution. I wasn't sure if that had implications for how I interpreted the data post-hoc (i.e., correcting for multiple comparisons) not because I cared about the differences between measures, but because I was running several ANOVAs. It sounds like in this case, I would not need to adjust the alpha value in evaluating the results, and actually instead of doing a repeated measures ANOVA I could just run paired sample T-tests instead. Thanks again. $\endgroup$ – redracer Mar 31 '13 at 22:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.