The Kruskal-Wallis ANOVA is a nonparametric method for testing the equality of different samples' medians. The Kruskal-Wallis test is an extension of Mann-Whitney U test to three or more populations. This test requires that the populations are identically distributed. The null hypothesis is that all of the population medians are equal. The alternative hypothesis is that at least two of the population medians differ.
It is a powerful alternative to the one-way ANOVA. Unlike one-way ANOVA, Kruskal-Wallis ANOVA does not require normality of the populations. It places less restriction on the comparison, and therefore has wider applications.
To perform a Kruskal-Wallis ANOVA:
Topics covered in this section: