Is there any method to conduct a non-parametric repeated measures ANOVA in a replicated design? I'm using a non-parametric approach because the response variable does not pass the parametric assumptions, and a Box-Cox transformation didn´t work.

I have a design with 4 treatments (habitat types), each one 10 fold replicated (plots at least 300 meters away) and I took measures (predation probability) in three consecutive times (repeated measures) of each replicate.

I need to test whether the treatments have an effect over the predation.

  • $\begingroup$ Just stumbled across this question and saw that it's essentially a duplicate of an earlier question. Maybe some of those answers can help, if you're still interested in the problem. $\endgroup$ – ThomasH Feb 18 '13 at 15:21