Is it possible to use the Kruskal-Wallis test for two factors and their interaction (i.e., two way ANOVA: Y~a*b
)? And which post-hoc test is possible to use after the Kruskal-Wallis test?
Asked
Active
Viewed 338 times
1
-
1Regarding post-hoc tests, the [Mann-Whitney U-test](http://en.wikipedia.org/wiki/Mann%E2%80%93Whitney_U_test) is a natural choice following a Kruskal-Wallis test. In R, this can be conducted with [?wilcox.test](http://stat.ethz.ch/R-manual/R-patched/library/stats/html/wilcox.test.html). You may (or may not) want to use an alpha adjustment (eg, Bonferroni correction) to address the multiple comparisons issue. – gung - Reinstate Monica Jun 27 '14 at 17:50
-
Kruskal-Wallis is only one-way; several post hocs designed for Kruskal-Wallis include Dunn, Conover-Iman and Dwass-Steel-Critchlow-Fligner. – Glen_b Jul 05 '15 at 08:58