On Sat, Dec 26, 2020 at 07:58:40PM -0600, Alan Mead wrote:
This paper, https://arxiv.org/pdf/2008.01200.pdf, suggests that the test
is flawed both in small samples and in samples with distinctly
non-normal underlying data. I don't know what it means to be "normally
distributed" for ranks... Ranks are always distributed uniformly unless
there are ties. Their method is implemented in the 'perk' library and is
also a sampling/resampling approach.
This is a criticism of fundamentals of the test itself, rather than
its implementation.
Alternatively, I wouldn't be upset if PSPP refuses to print any p-value
for N < 30. I think ideally we would add a keyword requesting a more
advanced algorithm.
I think this is probably the best course of action at least in the short
term.
Finally, I don't think any of this discussion bears on why the p-value
is missing from the Pearson r in CROSSTABS.
True. We need to look into that as well.
J'
