Re: [R] multiple comparison test

From: Peter Dalgaard <p.dalgaard_at_biostat.ku.dk>
Date: Mon 06 Jun 2005 - 20:16:37 EST

Julie BERTRAND <juliebertrand2001@yahoo.fr> writes:

> hello,
>
> after an anova I use pairwise.t.test(), it gives only
> p.value and I want the t.stat.
> I try to get these by computing the Welch
> approximation of the degree of freedom and using the
> qt(p.value,df) function but when I test this method
> with t.test results (the function gives p.value and
> t.test), I doesn't find the same t.stat.
>
> I also use the simtest(x~y,type="Tukey") function and
> it gives me only negative t.stat, resulting in a weard
> distribution...
>
> thank you in advance for all the attention you 'll
> give to my problems.
> Julie BERTRAND

Notice that the p-values in pairwise.t.test() are adjusted for multiple comparisons and use a pooled SD. You're not going to get anything similar to the t.test output unless you set pool.sd=FALSE and p.adjust.method="none". You probably also want to make the tests one-sided, or halve the p-value.

It could well be easier just to modify the function to give you the result that you desire...

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Received on Mon Jun 06 20:53:39 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:32:23 EST