[SciPy-user] help with scipy.stats.mannwhitneyu

Wavy Davy bloodearnest@gmail....
Thu Feb 5 10:24:45 CST 2009

2009/2/5 Bruce Southey <bsouthey@gmail.com>:
> I think the docstring is referring to the distribution of the actual
> U-test. So for small samples typically the pvalue is directly computed
> from the sampling distribution. However, Scipy is using the normal
> approximation is which is not meant to be that great.
> http://faculty.vassar.edu/lowry/utest.html
> http://faculty.vassar.edu/lowry/ch11a.html
> http://www.alglib.net/statistics/hypothesistesting/mannwhitneyu.php

OK - that makes more sense. Thanks.

I've ended up using the Kruskal-Wallis extension to Mann-Whitney
anyway, as I have multiple data samples. Which of course scipy
provides with the kurskal function. Confusing docstrings aside, its
been a pleasure to us :)


More information about the SciPy-user mailing list