[R] mann whitney u
Stefan Grosse
singularitaet at gmx.net
Wed Aug 26 17:22:14 CEST 2009
On Wed, 26 Aug 2009 12:18:53 +0100 "Mcdonald, Grant"
<grant.mcdonald08 at imperial.ac.uk> wrote:
MG> is the wilcox.test function equivalent to mann-whitney u? Is there
Yes, the test is the same. It is also called wilcoxon mann whitney test
because the authors created the test independently.
MG> a way to gain the U-value as apposed to the W-value in R?
What would be the purpose of this when you already have the p-value
given? As far as I know W and U are related. Look at
http://en.wikipedia.org/wiki/Mann%E2%80%93Whitney_U
and calculate the U1 U2 with the values of the example(wilcox.test).
You have not described how your data look like. In case of tie's
(equal values) use wilcox_test of the coin package.
hth
Stefan
More information about the R-help
mailing list