Which test is used to compare more than two groups of non-parametric data?

The Kruskal-Wallis Test is specifically designed for comparing more than two independent groups when the data does not follow a normal distribution, making it suitable for non-parametric analysis. This test assesses whether there are statistically significant differences between the medians of the groups. It does this by ranking the data from all groups together, analyzing these ranks to determine if the groups come from the same distribution.

When dealing with non-parametric data, traditional parametric tests, such as the ANOVA, are not appropriate. The ANOVA requires normally distributed data and homogeneity of variances, which can be violated in the context of non-parametric data. The Mann-Whitney U-test is a non-parametric alternative but is used only when comparing two independent groups rather than more than two. Similarly, the Student-T test is another parametric test intended for comparing the means of two groups, and is also not suitable for more than two groups, nor for non-parametric scenarios.

Therefore, when the objective is to analyze differences across three or more groups with non-parametric data, the Kruskal-Wallis Test is the correct and only appropriate choice.