200 Participants completed 3 tasks (easy, medium, difficult) and a cognitive ability test. All variables are continuous (for the task I have the time it took each participant to complete each task, and the cognitive test's score). To sum, I have 1 within variable (task difficulty) and 1 between variable (cognitive ability). The task time in each task level are positive skewed.

My hypothesis is that there will be an interaction between task difficulty and cognitive ability. What would be more appropriate – Repeated measures ANOVA or Mixed models?

Many thanks


1 Answer 1


Let me know if I'm understanding your question incorrectly -- you measured the cognitive ability of your participants first, then have them complete 3 tasks. You then get the time and scores on all the tests. This experiment design seems like a typical repeated measures setup.

If you are using just the scores as response variables, then repeated measures ANCOVA would make sense, with independent variables task difficulty and cognitive ability (I'm assuming the cognitive ability is some IQ test and therefore a covariate).

If you want to test the effects of the IV on both time and test scores, I would recommend repeated-measures MANCOVA, as the time and test scores are likely to be correlated.

  • $\begingroup$ Thanks. I'm not looking at the test scores, only the test time. $\endgroup$ Feb 9, 2018 at 7:56
  • $\begingroup$ In that case repeated-measures ANCOVA should be appropriate. $\endgroup$
    – Asy
    Feb 9, 2018 at 23:09
  • $\begingroup$ Thanks again. But then I have no IV, only a covariate (cognitive ability) and a DV (time) - Is this possible? $\endgroup$ Feb 10, 2018 at 15:43
  • $\begingroup$ Your task difficulty should be the categorical IV. $\endgroup$
    – Asy
    Feb 10, 2018 at 21:01
  • $\begingroup$ As far as I know, this is not possible in SPSS: For RM, SPSS need the data in wide format, so I can't associate the dummy variables for task difficulty and the other variables. $\endgroup$ Feb 12, 2018 at 18:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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