Exactly how many sample to be minimum size for non parametric test?

Right now i have 3 data with really2 small sample size 1. 5 pairs for wilcoxon signed rank test (pretest and postest in 1 place) 2. 4 pairs for Mann Whitney U Test (2 independent group) 3. 1 pairs for Mann Whitney U Test (2 independent group), the sample so small because the sample is the head office so there is only one in each place

The group represent the place,i only can do test in 2 place because lack of time, resources and funding.

The question is can i do statistical test for every sample? please give me reference book also. If i cant do statitical test, than what i can do to explain that the results is significantly different or not?

Please answer me ASAP (sorry for the bad english)

  • $\begingroup$ This is not possible to answer without you giving us more details! Tell us what is your real problem, what do you want to proove with the data, and what data do you have, in practical language. We can translate to statistics! $\endgroup$ – kjetil b halvorsen Mar 20 '17 at 14:34

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.