Date: Wed, 2 May 2007 09:58:36 -0500
Reply-To: "Swank, Paul R" <Paul.R.Swank@uth.tmc.edu>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: "Swank, Paul R" <Paul.R.Swank@uth.tmc.edu>
Subject: Re: How to graph an interaction with covariates
Content-Type: text/plain; charset="us-ascii"
Yes. We typically use the mean of the covariate when graphing
interactions of other terms. But whne the covariate is dichotomous, it
really doesn't make sense to use the mean. Using one level or the other
is more fitting. The zero term of course is the easiest to use.
Paul R. Swank, Ph.D. Professor
Director of Reseach
Children's Learning Institute
University of Texas Health Science Center-Houston
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
Sent: Tuesday, May 01, 2007 6:25 PM
Subject: How to graph an interaction with covariates
How do I deal with covariates when I want to graph a
significant interaction (regression analysis). I want to graph a 2-way
interaction where the outcome is continuous and the two predictors are
continuous and dichotomous, respectively. Also include in the analysis
were ten covariates: eight dichotomous (0,1) and two continuous.
What I have done is set the dichotomous variables equal
to 0 and use the mean for the two continuous variables; that is, I add
the weighted means of the two continuous variables to the constant. I
then proceeded follow the steps outline in Jaccard et al (1990) and
Aiken and West (1991) to generate plot points.
My question is: Did I handle the covariates correctly?
Any advice will be greatly appreciated.
Stephen Salbod, Pace University, NYC