« first day (5001 days earlier)      last day (10 days later) » 

User1865345
User1865345
8:19 AM
@Glen_b, I am thinking about whether it could be worthwhile to have an faq on wilcoxon vs t test and non-normality and small samples and their implications. Questions routinely come and revolve around this and ask for which test to perform.
Since there are already some good posts available, I don't think we need to create any new one. When I get time, I would propose in this vein at Meta CV.
Wait. I have to check first whether there are already canonical posts marked 🙂
 
 
3 hours later…
gung - Reinstate Monica
gung - Reinstate Monica
11:48 AM
@User1865345, I would say we already have a canonical Q on that:
139
Q: How to choose between t-test or non-parametric test e.g. Wilcoxon in small samples

SilverfishCertain hypotheses can be tested using Student's t-test (maybe using Welch's correction for unequal variances in the two-sample case), or by a non-parametric test like the Wilcoxon paired signed rank test, the Wilcoxon-Mann-Whitney U test, or the paired sign test. How can we make a principled dec...

hypothesis-testing t-test nonparametric small-sample wilcoxon-mann-whitney-test
 
Stephan Kolassa
Stephan Kolassa
Do we want to ask mods to put a faq tag on that? Maybe to replace the hypothesis-testing or the small-samples one? (Per my answer there, I would argue that the sample size is the least important consideration.)
 
User1865345
User1865345
12:03 PM
@gung-ReinstateMonica yes, gung. That was in my mind when I was saying there are already comprehensive posts. But I thought we need the faq tag for that.
@StephanKolassa exactly, Stephan.
There have been few link rots in these threads. Hopefully I will fix this (have already done two).
 
 
4 hours later…
gung - Reinstate Monica
gung - Reinstate Monica
4:31 PM
@Scortchi-ReinstateMonica, I would replace either [nonparametric] or [wilcoxon-mann-whitney-test]. There is too much overlap between those two.
 

« first day (5001 days earlier)      last day (10 days later) »