A Kruskal-Wallis test indicates whether a numeric column’s values tend to differ in size across three or more categories. If the values are drawn from distributions having the same shape but different locations across categories, the Kruskal-Wallis test can be interpreted as a test of whether the group medians are equal. It is an extension of the Mann-Whitney test (which applies to only two categories).

To perform a Kruskal-Wallis test:

- In the Summary view, use the confidence slider to choose a desired confidence level
- Choose the numeric column of interest on the left
- Choose a category column that takes three or more values in the Covariate table
- Click the button labeled
*Boxes*below the Covariate view - The result of the Kruskal-Wallis test will appear in the Bottom Line

Note that even if the observed medians are equal across categories, the Kruskal-Wallis test can indicate that the population medians are not equal. This is most likely due to the distribution of values differing across categories.

See also:

- Performing a Kruskal-Wallis test of three or more numeric columns’ medians
- Performing a Mann-Whitney test of a numeric column’s median across two categories
- Adjusting a p-value to account for multiple comparisons
- Performing an ANOVA test of a numeric column’s mean across values of another column
- Performing a Kolmogorov-Smirnov test of a numeric column’s distribution across values of another column
- Viewing the correlation coefficient between two numeric columns
- Formatting descriptive and inferential statistics
- About the Summary view