I have 3 groups of 30 fish.

group 1: without anaesthesia
group 2: clove oil anaesthesia
group 3: MS-222 anaesthesia

Every group was investigated at six levels of time (0, 0.5, 1, 2, 6 and 24 hours after anaesthesia). The data aren't normal. Which test to use? Kruskal-Wallis or Friedman?

  • 2
    $\begingroup$ What is your exact research question? Do you want to check all 6 levels of time or only the final level with respect to initial level? What exactly are you testing? Is the outcome variable a numeric one or categorical / ordered? Add more details to your question. $\endgroup$ – rnso Apr 12 '15 at 7:32
  • $\begingroup$ I want to compare all 6 levels of time in within and between groups. In fact, I want to compare the effects of different methods of anesthesia on the number of the specific cells. $\endgroup$ – Sara Apr 13 '15 at 13:20
  • 1
    $\begingroup$ Graphical methods may be the best for this. If you post some sample data here, it will be helpful. $\endgroup$ – rnso Apr 13 '15 at 14:44

Briefly spoken you use the Kruskal-Wallis test if the data has unmatched pairs and the Friedman-test is the data has matched pairs.

In my opinion the data of the three tests is unmatched. Therefore the Kruskal-Wallis test is the better test.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.