The Kruskal–Wallis one-way analysis of variance by ranks is a non-parametric equivalent of the one-way ANOVA. It assumes independent samples and has the null hypothesis, that the distribution is identical. If the calculated p-value is <0.05, it is concluded that the distributions are nonidentical at the 0.05 significance level (R example).
The Kruskal-Wallis configuration window is illustrated below.
Up to four groups can be compared, according to the Type selection
If Merge columns is enabled, the data of all columns will be pooled, otherwise the test is performed between the respective columns of the aggregates.
With the One aggregate - columns as groups setting it is assumed that the groups to be compared are arranged as columns in a single aggregate.
The script produces a result table (kruskal_wallis_result) listing the p-value for each aggregate column and indicating its significance.