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

To perform a Kruskal-Wallis test:

- In the Summary view, use the confidence slider to choose a desired confidence level
- Command-click to select three or more numeric columns of interest on the left
- 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 columns, 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 columns.

See also:

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