A Mann-Whitney test indicates whether a numeric column’s values appear to be larger in one of two categories. If the values are drawn from distributions having the same shape but different locations across categories, the Mann-Whitney test can be interpreted as a test of whether the group medians are equal.
To perform a Mann-Whitney test:
Note that even if the observed medians are equal across categories, the Mann-Whitney test can indicate that the population medians are not equal. This is most likely due to the distribution of values differing across categories.
To analyze differences in a numeric column’s median across three or more categories, use a Kruskal-Wallis test.