Date: Thu, 1 Sep 2005 07:18:15 -0400
Reply-To: "David M. Fresco" <firstname.lastname@example.org>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: "David M. Fresco" <email@example.com>
Subject: Decomposing interaction term in Cox Regression
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
I am hoping for some help in undertaking a decomposition of an
interaction effect in Cox regression.
I am tracking when relapse occurs in a sample of individuals as a
function of what treatment they received as well as two additional
factors A, B and A*B.
The output indicates that the interaction of A*B is significant and a
get a B and SE of B in the output. I am certainly familiar with
solving interactions for OLS, but would welcome some suggestions on
how to interpret this interaction.
Below is essentially the syntax I used.
/STRATA = txcond
/METHOD=ENTER a /METHOD=ENTER b /method = ENTER a*b
/CRITERIA=PIN(.05) POUT(.10) ITERATE(20) .
Thanks in advance and I would appreciate my email being cc'ed to
responses as I take the digest.
David M. Fresco, Ph.D. firstname.lastname@example.org
Department of Psychology Voice: (330) 672-4049
Kent State University Fax: (330) 672-3786
226 Kent Hall Annex
P.O. Box 5190 http://dept.kent.edu/
Kent, OH 44242-0001 http://www.personal.kent.edu/
ListOwner of Helplessness http://psych.upenn.edu/~fresco/
Listowner of Mindfulness http://listserv.kent.edu/archives/