I am looking to do a non-parametric test of a 2x2 mixed anova as my dependent variables are far from normally distributed?

  • $\begingroup$ Note: ANOVA assumption number 2: Normality – the distributions of the residuals are normal. $\endgroup$ – user2974951 Feb 12 at 13:13
  • $\begingroup$ The nonparametric equivalent of ANOVA is Kruskal–Wallis one-way analysis of variance. $\endgroup$ – Digio Feb 12 at 13:20
  • 2
    $\begingroup$ Define the experimental design, i.e., what 2x2 and mixed mean in your context. $\endgroup$ – Frank Harrell Feb 12 at 13:21
  • $\begingroup$ My 2 factors are participant gender (between) and then the gender portrayed in the vignettes I use - but all participants see both conditions so this is my within variable. $\endgroup$ – Charlotte Feb 12 at 13:39
  • 1
    $\begingroup$ So, ANOVA is irrelevant. See a text on categorical data analysis. $\endgroup$ – Nick Cox Feb 12 at 14:43

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.