[R] median test

Stefan Grosse singularitaet at gmx.net
Fri May 28 19:35:15 CEST 2010


 > **Disclaimer: I have no idea what your data represents or how
> (in)appropriate any of these tests may be**
>
> R can do the tests you mentioned (and many more).
>
> Wilcoxon test:
> wilcox.test(x=group1, y=group2, paired=FALSE)
> see ?wilcox.test
>    

I am not sure whether it is still valid but in case of ties (equal 
values in the same group) like in your example data you should use the 
wilcox_test from the coin package. That one has a correction for ties.

But as Joshua told correctly as far as we do not know your data, Linda, 
it is hard to tell what might be approproate...

Stefan



More information about the R-help mailing list