I did chi square test on data of 60 patients having COVID 19. The variable in the rows is the severity of case (mild, moderate, severe) while the variable in the columns is lymphocytes count (normal, lymphopenia).

I found a significant difference and I liked to do a post hoc test using Bonferroni correction and it is known that we should divide the alpha level 0.05 on the number of pairwise comparisons and here I got confused about how to calculate the number of pairwise comparisons because I searched online and found two different ways to calculate the number of pairwise comparisons:

First way, I have three pairwise comparisons coming from comparison of severity categories (mild to moderate), (mild to severe) and (moderate to severe). Second way, I have six pairwise comparisons that come from multiplying number of rows by number of columns i.e. 3 by 2 equaling 6.
So, which way should I use and why?

  • $\begingroup$ You seem to be telling us you are making three comparisons and then you state you are making six comparisons. So let's begin there: how many comparisons do you plan to make? $\endgroup$
    – whuber
    Mar 20, 2022 at 14:06
  • $\begingroup$ Actually I think I must make 6 comparisons, I don't know if it is fine to make 3 comparisons. My question was about this point i.e. how to determine number of comparisons? $\endgroup$ Mar 20, 2022 at 16:26
  • $\begingroup$ The number of comparisons is ... the number you want to make. It's not determined statistically (or by anything else) other than your analytical objectives. $\endgroup$
    – whuber
    Mar 20, 2022 at 18:02


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.