← Math & statistics

StatisticsIndependent 3+ groups

Kruskal-Wallis Test Calculator

Other languages 日本語 | English

Use Kruskal-Wallis when 3 or more independent groups should be compared with ranks instead of mean-based ANOVA assumptions. This page focuses on the H statistic, tie correction, and group rank summaries, not post-hoc multiple comparisons.

How to use

  1. Enter one independent group per line in Label: values form.
  2. Use at least 3 groups and keep each group on its own line.
  3. Read the p-value together with H, tie correction, medians, and rank sums.

Wave 38 nonparametric multi-group entry

Compare 3 or more independent groups by rank

The share URL stores only decimal places.

Enter 3 or more independent groups to review the Kruskal-Wallis H statistic and p-value.

How to choose between nearby pages

Use ANOVA when the question is mean differences under parametric assumptions, and Mann-Whitney U when there are only 2 independent groups. This page covers the rank-based 3-or-more-group case without post-hoc comparisons.

FAQ

What does this page calculate?

It runs a Kruskal-Wallis test across 3 or more independent groups and returns the H statistic, p-value, tie correction, and group rank summaries.

When should I use Kruskal-Wallis instead of ANOVA?

Use Kruskal-Wallis when 3 or more independent groups should be compared with rank-based, nonparametric logic instead of mean-based ANOVA assumptions.

When should I use Mann-Whitney U instead?

Use Mann-Whitney U when there are only 2 independent groups. Kruskal-Wallis is the 3-or-more-group extension in the same nonparametric lane.

Does the share URL include my grouped values?

No. The share URL stores only light settings such as decimal places. Your grouped values stay in your browser.