I have conducted a mixed-design two-way ANOVA (between subjects variable : 2 levels; within subjects variable : 8 levels). I have a significant main effect of between and within subject variables and a significant interaction. I expanded on my interaction by running multiple t-tests ( 8 paired to compare treatments at each within-subject variables level and 28 independent t-tests to compare within subject variables at each level of between subjects variable; ie: 28 comparisons per treatment group). When reporting my adjusted p value, do I adjust it to 0.05/28 for independent and 0.05/8 for paired t-tests or do I put them all together 0.05/36? Graphpad calculates them separately. Is that the correct way?

  • $\begingroup$ You may be better off doing all this in one complete mixed model. $\endgroup$ – user2974951 Jan 30 '19 at 12:01
  • $\begingroup$ How can I do that in SPSS? $\endgroup$ – N.Doostdar Jan 30 '19 at 13:13
  • $\begingroup$ If you know nothing about mixed models you may be better off asking for help (this site will help you only to a certain degree). Anyway, I do not know SPSS so I cannot help you there, have a look at the documentation stats.idre.ucla.edu/spss/seminars/spss-mixed-command $\endgroup$ – user2974951 Jan 30 '19 at 13:17

Your Answer

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

Browse other questions tagged or ask your own question.