[SciPy-user] help with scipy.stats.mannwhitneyu
Thu Feb 5 10:24:45 CST 2009
2009/2/5 Bruce Southey <firstname.lastname@example.org>:
> 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.
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