[R] Re: Multiple Comparisons (was empty)

Frank E Harrell Jr fharrell at virginia.edu
Sat Feb 16 13:42:18 CET 2002


I am glad to see that adjustment for multiple comparisons is becoming more and more controversial.  Such adjustments are very arbitrary, non-unique, and help to propagate the use of P-values for binary decision making.  And the points made by Brian is very much forgotten by most practitioners: violation of assumptions and model uncertainty cause problems that are only compounded by multiple comparison adjustments.

Frank Harrell


On Sat, 16 Feb 2002 08:30:43 +0000 (GMT)
Prof Brian D Ripley <ripley at stats.ox.ac.uk> wrote:

> On Sat, 16 Feb 2002, Andrew Criswell wrote:
> 
> > Hello ALL:
> >
> > I am looking for multiple comparison tests for ANOVA models. Does R have any
> > of these?
> >
> >     Fisher least significant difference test.
> >
> >     Tukey henestly significant difference test.
> 
> is that `Honest' as in Tukey's HSD?
> 
> >     Newman - Keul test.
> >
> >     Dunnett's test
> >
> >     Scheffe test.
> 
> See Tukey() in package Devore5.
> 
> Beyond that, most of these are easy to program, and some are just a matter
> of usage of existing facilities (like F tests in summary.aov / drop1).
> 
> I've been looking into programming something for R (Miller's book is on my
> bedside table), but I am more interested in confidence intervals for
> contrasts than in tests.  I have also a feeling that providing software
> would encourage people to use these, and they are only appropriate in very
> special circumstances (including no data cleaning and no model selection,
> as well as close to normal iid errors).
> 
> -- 
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272860 (secr)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._


-- 
Frank E Harrell Jr              Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine  http://hesweb1.med.virginia.edu/biostat
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list