topic Re: Paired t-test or wilcoxon signed rank for N=5 in Statistical Procedures

Paired t-test or wilcoxon signed rank for N=5

LucyB — Fri, 24 Mar 2017 03:09:57 GMT

I have some pre post data for an extremely small sample (n=5). Outcome is continuous and no control. What statistics would be more appropriate to report? Accordingly, is a paired t-test or a wilcoxon signed rank test more appropriate?

I feel 5 data points are insufficient to assess normality. Should i report both?

Re: Paired t-test or wilcoxon signed rank for N=5

Reeza — Fri, 24 Mar 2017 03:21:42 GMT

Report mean and maybe direction of change? 5 is too small to make any generalizations - may even want to just report all the numbers. We summarize in first place because too much data to understand otherwise. 5 points don't have that issue.

Re: Paired t-test or wilcoxon signed rank for N=5

LucyB — Fri, 24 Mar 2017 03:27:51 GMT

So maybe a t-test and wilcoxon sign test?

Re: Paired t-test or wilcoxon signed rank for N=5

PGStats — Fri, 24 Mar 2017 03:28:51 GMT

Report the Wilcoxon signed rank test result. It is non parametric and the value reported by proc univariate is exact for samples of size < 20.