|Date: ||Thu, 24 Apr 1997 10:47:46 -0400|
|Reply-To: ||Rodney Sparapani <spara002@MC.DUKE.EDU>|
|Sender: ||"SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>|
|From: ||Rodney Sparapani <spara002@MC.DUKE.EDU>|
|Organization: ||Duke Clinical Research Institute|
|Subject: ||Re: Bonferroni correction?|
|Content-Type: ||text/plain; charset=us-ascii|
Margaret Ann Goetz wrote:
> I'm currently working on a linear regression model with four independent
> variables and sample size of 77. The set of regression models can be
> described as exploratory.
> When I perform my final linear regression models, is it necessary to include
> correction for possible Type I error such as the Bonferroni correction? I'd
> particularly interested in arguments, both in favor of and against.
> And, if YES, how would the Bonferroni correction be correctly applied for
> model conducted? (e.g., 4 regression models at alpha =.05 becomes .05/4 or a
> significance level of .0125 for rejecting each indep var as significant?)?
> Any help is very much appreciated!! Thanks in advance!
I think what you are looking for is something like Holm's
Sequentially Rejective Procedure, but I don't have a reference
that you are likely to find in your library.
You can't pick your ex.
Rodney Sparapani Go Blue Devils.
Duke Clinical Research Institute Go Packers.