Good day!
I ran a non-parametric permutation test for Lagged coherence connectivity analysis between 2 independent groups, then I applied a p treshold with FDR correction,
I would like to ask what is the best approach for getting the effect size, I know the stat is in the file, but I mean a stadardized effect size (e.g. Cohen's D, biserial rank correlation, etc)
Since the permutation test allows to select the student's T test (despite non parametric data) I don't know what type of effect size coeficient is more suitable for the data,(Squared Eta?)
thanks in advance!
Due permutation test uses Student's T test as basis, I guess I should compute effect size in the same way we usually compute it for T test, right? 
Sorry, I am not sure I follow what you want to achieve. Effect size is something you derive from the data and is independent of the inference statistical test.
3 Likes