Date:         Wed, 14 Sep 2005 14:37:15 -0700
Reply-To:     Dale McLerran <stringplayer_2@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dale McLerran <stringplayer_2@YAHOO.COM>
Subject:      Re: GLM CONTRAST vs TTEST
In-Reply-To:  <BAY103-F2933CC3E02EF3AE731FD01B09F0@phx.gbl>
Content-Type: text/plain; charset=iso-8859-1

--- David L Cassell <davidlcassell@MSN.COM> wrote:

> In addition to Dale's excellent advice on heteroskedasticity in ANOVa
> models,
> you can also use randomization tests to address the problem.  There's
> SAS code for a randomization test wrapper in the SUGI 27 proceedings.
>

Indeed, it is randomization tests which I had in mind when suggesting
that nonparametric methods would have utility for small sample
problems.  In small sample problems, inspecting assumptions about
normality and equality of variance can be difficult at best.  A
randomization test can be constructed to evaluate any contrast of
interest without having to rely on distributional assumptions
and/or asymptotics.  I would heartily endorse use of David's
wrapper to make randomization tests easier to conduct.

Dale

---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@NO_SPAMfhcrc.org
Ph:  (206) 667-2926
Fax: (206) 667-5977
---------------------------------------

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com