Key points at a glance
- Verify independent groups.
- Use ranks for a suitable question.
- Separate overall and pairwise tests.
Verify independent groups
Kruskal-Wallis compares rank distributions across three or more independent groups. Repeated observations of the same people call for another method. Check measurement scale, outliers, distributions and research question; it is not an automatic substitute for every ANOVA.
Think through a suitable example
Example: Three separate seminar groups rate a tutorial’s clarity. Check that each person belongs to only one group and report valid response counts. Show distributions or medians with spread before presenting a p-value.
Interpret ranks carefully
The test evaluates a global difference among groups. Without similarly shaped distributions, do not reduce the result to a simple statement about medians. Check whether tied ranks and small groups affect the software procedure used.
Plan follow-up comparisons
A significant overall result does not identify the differing groups. Pairwise questions need suitable follow-up tests and justified multiple-testing control. Report H statistic, degrees of freedom, p-value, group sizes and a suitable effect measure.
Final submission context
Formal details can feel like a separate writing task, but they matter just as much for a printed submission. Anything missing, misplaced or inconsistently formatted in the document will also appear in the bound copy.
Use this guide together with your cover page, table of contents, page numbers, source notes and appendices. Prepare the final PDF for printing and binding only after the complete file has been checked.
Practical check before PDF export
- Do headings, chapter structure and page numbers match?
- Are sources, figures and tables included completely?
- Are required elements such as declarations, appendices or the cover page included where required?
- Did you open and check the final PDF after exporting it?
Checklist
- Verified independence.
- Examined distributions.
- Reported global result.
- Controlled follow-up tests.
Common mistakes
- Using paired data.
- Treating global test as pairwise evidence.
- Calling every rank difference a median difference.
Prepare the finished thesis for submission
Once the content and university requirements are checked, review your PDF and the available printing and binding options.
Review printing optionsRelated content
Explore related guidance on thesis structure, sources, formatting and final submission.
Frequently asked questions
Is it a nonparametric ANOVA?
It is a rank-based independent-groups test, but its interpretation is not identical.
Does it need normality?
Not in the usual ANOVA sense; other conditions still matter.
Which groups differ?
An appropriate follow-up comparison is needed.