I have 2 samples of data. The first sample is the precision of the first algorithm A for 25 test cases and the second sample is precision for 25 test cases of second algorithm named B. I test both samples and first is normal distributed but second is non-normal distributed. (I check it with QQ plot, lillietest and shapiro-wilk test in R language).

Can I test this two samples with different distribution for example do they have an equal mean? If yes, which test should I use in this case?

  • $\begingroup$ With samples as large as 25, it might be OK to use a two-sample t test, provided the second sample isn't very far from normal. It's difficult to say without seeing the data. // There are 'nonparametric' tests that do not assume normal data, but they do involve other assumptions. Without seeing the data it would not be responsible to recommend a nonparametric test--It might actually be worse than just doing a t test. $\endgroup$ – BruceET Dec 29 '18 at 22:10

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.