In the APA guidelines of reporting non-parametric tests they suggest reporting the median results when using non-parametric tests (http://evc-cit.info/psych018/Reporting_Statistics.pdf - search for 'non-parametric').

I have a significant result for a large non-parametric data set where the medians are the same, but the means are different. (I know that there is a debate on the usefulness of means for non-parametric data, but my professors have argued that it is useful in this case.)

Would it be 'OK' to report the results using the mean, or both the mean or the medium, or just stick to the median?

Thank you for any help.

  • $\begingroup$ Your data set cannot be "non parametric" - presumably you mean that your data are ordinal or something like that? $\endgroup$
    – Bryan Krause
    Jul 19 '19 at 16:15
  • $\begingroup$ @BryanKrause yes this is what I mean $\endgroup$ Jul 21 '19 at 13:41

Your Answer

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

Browse other questions tagged or ask your own question.