I am writing code for Conover Squared rank test in Python but I am unable to find right formula to make corrections for ties in order to calculate the p value. I understand the average of the tied ranks must be used, but what I don't know is how is the formula for p-value calculate is modified.

I have found the most detailed discussion for handling ties in Conover's squared rank in following sources: Practical Nonparametric Statistics by Conover page 302, https://ccrg.rit.edu/~whelan/courses/2018_3fa_STAT_345/notes05.pdf\

Yet I unable to find clear approach to manage tied ranks, such as available for Mann-Whitney U test on its Wiki page.

I have inserted the screen shots of the key parts of the formula From Practical Nonparametric Statistics by Conover page 302-304

On the same note, how important it is to correct for ties in Conover test.

  • $\begingroup$ The details all seem to be on p26-27 $\endgroup$ – Glen_b Nov 26 '19 at 0:07
  • $\begingroup$ Hi Glen, I need to calculate the Z score in order to calculate the p-value. This is present in the image I shared from the book by Conover, but same is missing in p26-27. Perhaps T1 on page 27 is what I need. Is T1 the Z-score after adjusting for ties? $\endgroup$ – Vineet Nov 26 '19 at 14:56

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.