How to do Fisher's exact test in univariate survival comparison on 2 groups (10 vs 15 patients), small size events ( 5 vs 3) and multiple times to events?


1 Answer 1


Fisher’s “exact” test is only for categorical uncensored dependent variables.

  • $\begingroup$ Thank you sir! I wanted to work on by Log-rank test, but according to some books, my results will be invalid because there are several cases with expected values below 5. What should I do? $\endgroup$
    – Soffee
    Commented Dec 22, 2022 at 0:10
  • 1
    $\begingroup$ You're still mixing regular categorical Y analysis with survival analysis. I've never seen "5" used in the context of the logrank test. But there is no reason for the logrank test to exist. It is just a special case of the Cox proportional hazards regression model. Use the likelihood ratio $\chi^2$ test within a Cox model framework. $\endgroup$ Commented Dec 23, 2022 at 10:58

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.