I have a data set (n=330) that is composed of two treatments. Within each treatment there are 12 replicates. Within each replicate 5 individuals were chosen. Each individual had 5 measurements taken. The data violates the normality assumption and I am not able to transform it to meet this assumption. I think that the best option is to run a Kruskal Wallis Test as it does not require normality of the data but I am unable to get SAS or JMP to execute the analysis.
Any advise would be greatly appreciated.
Any advise would be greatly appreciated.