I am analyzing an experiment run with 162 participants using a 2×2 mixed design ANOVA. The experiment has two independent variable with two levels each, and two dependent variables. I need to know the significance level of each level between each dependent variable.

Is there any reason to run post-hoc test in this context? SPSS complains about variables having less than 3 levels and, for this reason, doesn't allow me to run Post-hoc tests.

Question do i need the post-hoc results to find the effect of each level of the independent variables on each dependent variables , If yes, how to do it in SPSS? if not then how to find such results?

  • $\begingroup$ You probably also "need" to know effect sizes as well :P $\endgroup$ – probabilityislogic May 2 '16 at 7:33
  • $\begingroup$ I KNOW ... but how by using spss?!! $\endgroup$ – sara morsy May 2 '16 at 9:11

If you have a Main Effect there's no need to do a post hoc test. If the two groups differ (main effect) then a post hoc test tells you nothing more (although they all use different criteria). Do you understand what a main effect means(I'm not trying to be rude/condescending)?

By the way, the analysis is in Analyze, General Linear Model, Repeated Measures. In options you may want to select Effect Size Estimate for partial eta squared to be given.

  • $\begingroup$ i actually don't get what do you mean by "main effect"? ... well, if i'm getting you right.. then yes the two groups differ... For example, one of my independent variables is "qualification"; qualified and not qualified... What does post hoc tells me exactly? $\endgroup$ – sara morsy May 2 '16 at 17:50

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.