--- title: "Post-hoc tests (Tukey or Dunn)" slug: "post-hoc-tests" updated: 2026-05-27T14:47:05Z published: 2026-05-27T14:47:05Z canonical: "userguide.dataclassroom.com/post-hoc-tests" --- > ## Documentation Index > Fetch the complete documentation index at: https://userguide.dataclassroom.com/llms.txt > Use this file to discover all available pages before exploring further. # Post-hoc tests (Tukey or Dunn) Having run an ANOVA or 2-way ANOVA test, you will have received a P-value which tells you about the likelihood of there being a significant difference between some of the groups. But it doesn't tell you **which groups differ from which** (In the T-test, there are only two groups, so the question does not arise). This is where a Tukey's post-hoc test comes in. A test with many namesThe test goes under a number of names. It is called Tukey's Range Test, Tukey's Honestly Significant Difference, or just Tukey's HSD. You should see the button below the table of results: ![](https://cdn.document360.io/62f8f41e-e041-42c4-b9bc-8ae1eb45d064/Images/Documentation/image-1623736789164.png) And running the test gives you something like this table, where you can see each of the four groups compared with each of the other three. ![](https://cdn.document360.io/62f8f41e-e041-42c4-b9bc-8ae1eb45d064/Images/Documentation/image-1623737158339.png) So what does this mean? If you look at the table carefully, you'll see that **all the P-values associated with the LowCal group are significant (under 0.05)**, and this should correlate with what your eyes tell you when you look at the graph (mean and 95% confidence interval shown), i.e. that the **LowCal group is the one that is different from all the others**: ![](https://cdn.document360.io/62f8f41e-e041-42c4-b9bc-8ae1eb45d064/Images/Documentation/image-1623736954781.png) ### Dunn's test (post-hoc for Kruskal-Wallis) If you have run a Kruskal-Wallis test for non-parametric data (rather than an ANOVA), the post-hoc test offered will be a Dunn's test. Otherwise, details of the interface are as per Tukey above.