Dear all
I like to use the Mann-Whitney U-test with one of the less conservative Bonferroni-type corrections when carrying out post-hoc multiple comparisons further to an application of the Kruskal-Wallis test. I am not a user of GraphPad Prism. However, I am aware that this package encourages the user to select Dunn's test as the appropriate post-hoc test for the Kruskal-Wallis test, presumably with the Bonferroni adjustments already applied.
I would be most grateful for thoughts on the relative merits of either of the above two approaches when deciding to do pairwise comparisons as post-hoc tests to the Kruskal-Wallis test.
Thank you very much in advance for your kind contributions.
Best wishes
Margaret
---------------------------------
Switch an email account to Yahoo! Mail, you could win FIFA World Cup tickets.
|