Is there any reference that I might use when I mention the median split method in a paper? i.e. turning a continuous variable into a categorical one by splitting a sample into 2 (above and below the median).

  • 2
    $\begingroup$ Are you after a reference explaining how it's usually a really bad idea? I think I can dig up several of those. (If you're after a reference that says it's a good idea, I don't actually know of any.) $\endgroup$ – Glen_b Nov 14 '15 at 16:42
  • $\begingroup$ No actually I am using it...but to stratify my data for a first check... $\endgroup$ – gabboshow Nov 14 '15 at 16:50
  • $\begingroup$ For example I stratify according to the age (2 groups: above and below the median). $\endgroup$ – gabboshow Nov 14 '15 at 16:52
  • 1
    $\begingroup$ So you want a reference that advocates using it as some kind of first check? Or just at all? $\endgroup$ – Glen_b Nov 14 '15 at 21:17
  • 2
    $\begingroup$ This answer by Jeromy Anglim contains links, some of which have references, a few of which might be of some use to you, since they discuss dichotomizing. There's also links in the answers here that you may be able to find something in. [There are very few situations where it makes sense to do this, even for a first check -- it's no harder to run a regression than a t-test] $\endgroup$ – Glen_b Nov 15 '15 at 2:13

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.