|Date: ||Wed, 14 Feb 2001 15:56:36 -0800|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||"David L. Cassell" <Cassell.David@EPAMAIL.EPA.GOV>|
|Subject: ||Re: Bartlett's test|
|Content-type: ||text/plain; charset=us-ascii|
Dale McLerran wrote:
> In some experiments, you may run into small numbers of observations
> for at least some combinations of factors. I think you could get
> into trouble performing a test for homogeneity of variance when
> some groups have few observations.
A good point.
> Would you have any qualms
> with fitting the full model, generating the residuals, and then
> performing an HOVTEST on the residuals for each factor separately?
> This may not be an optimal test, but it seems to me that it
> should suffice for most problems.
This is one reason why I like the Brown-Forsythe and Fligner-Killeen tests,
as well as the *adjusted* Bartlett's test if those are not available. You
get real robustness with these [as well as with Levene which has a
problem]. Note that SAS does not do the latter two. Your suggestion seems
reasonable to me, although we would have to watch the alpha for the
We might be better off just doing residual plots...
David Cassell, OAO Corp. Cassell.David@epa.gov
Senior computing specialist