Date:         Tue, 7 Jun 2005 11:07:02 -0700
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: Multiple comparison
In-Reply-To:  <4A6EB3AE96290A4CBC9337ABDBC84707F0901A@PA100-EX-01.Endo.com>
Content-type: text/plain; charset=US-ASCII

"Li,Tom" <Li.Tom@ENDO.COM> wrote:
> Suppose I have 3 treatment groups with placebo, low dose and high
dose.
> I use ANOVA's overall F-test to test u0 = u1 = u2 and I just present
the
> means of the 3 groups and overall p-value from the overall F-test.  Do
I
> still need to do the multiple comparison adjustment?  TIA.

If you are *only* showing the overall test p-value, then you are
not doing multiple comparisons.  So you don't need that adjustment.
'Multiple comparisons' is used to describe the process when you
test a number of the (usually pairwise) comparisons of means, not
when you only perform the overall F-test and then stop.

YWATF (You're Welcome After The Fact),
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician