A colleague said "no one does ANOVAs anymore" (which I found odd and hard to believe), and "do a permutation test." I tried this, and it returns equivalent results as the Willcoxon Rank-Sum, both non-significant. From what I understand, if the ANOVA does meet all the criteria, then it would seem the appropriate test because it would be more sensitive.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |