Date: Tue, 16 Dec 2008 17:38:18 -0500
Reply-To: Peter Flom <firstname.lastname@example.org>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Peter Flom <peterflomconsulting@MINDSPRING.COM>
Subject: Re: Regression: do you always need main effects with interactions?
Content-Type: text/plain; charset=UTF-8
Kevin Viel <citam.sasl@GMAIL.COM> wrote
>I can also recall a conversation I had with a committee member, Prof. Drews-
>Botsch. State your model a priori and that is the one you use to test your
>hypothesis. After you run the regression, you do not drop covariates or
>add covariates-that is left for another study, you simply interpret your
>results. Choose your model with the best knowledge and science available
Prof Drews-Botsch is funny. If we did what he says, all the time, we would make almost no progress at all. Scientific advance requires exploration as well as confirmation.
In a data set I am working with now, there are about 1000 variables, and it took about 10 years to collect all the data. Shall we test each of 1000 models separately, getting new data each time?
Or, to prevent that, perhaps we should have several thousand hypotheses to test, a priori ....
Peter L. Flom, PhD
www DOT peterflom DOT com