Kruskal-Wallis Test Calculator

Want to calculate a Kruskal-Wallis test? You can do that with a few clicks right here in your browser. Just select a metric or ordinal variable and a nominal variable with more than two characteristics. Then click on Nonparametric Test.

The Kruskal-Wallis test can be used to determine whether the groups studied differ significantly from one another.

Calculate Kruskal-Wallis Test

When calculating the Kruskal-Wallis test, it is not the mean values of the individual groups that are compared, but the rank sums. This has the advantage that the data do not have to be normally distributed. Just try it out with the sample data, it is very easy with the Kruskal-Wallis test calculator. DATAtab then gives you the results including interpretation in the following form:

Of course, the Dunn-Bonferroni post-toc test is also given out to check which groups are different.

When is the Kruskal-Wallis test calculated?

The Kruskal-Wallis test compares more than two samples and it is used whenever the conditions for calculating a one-way analysis of variance are not met. For example, if the test for normal distribution shows that your data are not normally distributed, you must use the Kruskal-Wallis test.

If you don't have an independent sample, then evaluate your data with the Friedman test calculator. For more information, see our tutorial on the Kruskal-Wallis test.

The hypotheses for the Kruskal-Wallis test are:

• Null Hypothesis (H0): The distributions of all groups (or samples) are equal. This implies that there's no significant difference in the median ranks between the groups.
• Alternative Hypothesis (H1): At least one sample stochastically dominates one other sample. This suggests that at least one group tends to have values that are larger or smaller than another group.

If the Kruskal-Wallis test is significant, it indicates that at least one of the groups differs from the others. However, the test itself doesn't tell you which groups are different from each other. For that, post-hoc pairwise comparisons are needed.

No matter what hypothesis test you want to calculate, with the statistical software DATAtab you can do it directly online, unlike SPSS and Excel.