I would like to test if there is an association between:

-A binary factor: 0, 1 (recovered vs not recovered) -A ordered factor: Iron level 1, 2, 3 (low, medium, high)

I was wondering if the rank of the iron level variables is considered, when testing for an association between those two variables using a Chi-Square test?



No, I don't think so. 1, 2, 3 are treated pretty much like unordered labels like "red", "green", "blue".

  • $\begingroup$ Thank you! Argh, I was worried it wouldn't, as the p-value is exactly the same when I tested the iron level as nominal factor. Do you have advice about which test I could use to test the association between an ordered factor and a binary categorical variable as in the example above? $\endgroup$ – evouk Aug 14 '19 at 11:30
  • 1
    $\begingroup$ A chi-square test surely is an option (would work even if the variable is ordered). You could treat the ordinal values as continuous and do the t-test to see if the mean is different between recovered vs not-recovered. Of course, this isn't really treating this variable as ordinal though. And this is a deep area. Have a look like this researchgate.net/post/… $\endgroup$ – Kota Mori Aug 14 '19 at 11:44
  • $\begingroup$ Thank you, that's really helpful. I think I'll take your advice and do a t-test to see if there is a significant difference in the means. $\endgroup$ – evouk Aug 14 '19 at 13:28

Your Answer

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