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?
You can't make much of a distributional assessment at n=5 -- you can (sometimes) see non-normality at n=5 -- but at such small sample sizes the non-normality has to be quite strong to have a good chance to pick it up. The level sensitivity of the paired t-test to heavy tails can be substantial for example, but that can be tricky to spot. If you have no good basis on which to assume normality, you may have difficulties there.
With sample size of 5, your smallest attainable significance level for a two tailed signed rank test is 1/16 (that is, if you want to set a significance level of 0.05, you can never reject the null). This issue is covered in a number of questions on site. Which is to say you may have little choice other than some parametric assumption (possibly some suitably chosen form of robustification may help, allowing you to bound the potential influence of one observation without so severely limiting possible significance levels)
