Confirmation of correct test for repeated measures of ordinal data: Friedman's Test

#1
Dear Stats Community,

I performed an experiment in which I measured the behavioral responses on an ordinal scale (1-4) of a group of non-human animals to a series of auditory stimuli (A,B,C,D). The order of the presentation of the stimuli was randomized, but each animal experienced all four stimuli. I need to measure whether the responses to the stimuli were different.

I believe that because of the dependent nature of the data, the non-parametric Mann-Whitney-U/ Kruskal-Wallace tests are unviable. The Friedman test seems to be able to handle repeated measures with non-parametric data, but I do not fully understand it or its post-hoc tests. I would greatly appreciate advice about the best (and most robust) test to use in this situation, and any advice about problems I may face when running these data in R.

Thank you greatly for your advice and time!
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
Re: Confirmation of correct test for repeated measures of ordinal data: Friedman's Te

From your description, Friedman's tests seems appropriate.