Propensity scores can be used as covariates or for matching but one
would not typically use them for both at the same time. Matching has the
usual disadvantage of reducing sample size by eliminating cases that do
not match whereas the covariate approach can be used with all subjects.
Paul R. Swank, Ph.D. Professor
Director of Reseach
Children's Learning Institute
University of Texas Health Science Center-Houston
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Jeff Mather
Sent: Tuesday, September 25, 2007 12:17 PM
To:
[hidden email]
Subject: Propensity matching question
Hi,
This is a stats question in general, not specific to SPSS.
I'm comparing 2 types of stents from an observational data set. To
match the stent groups, I used propensity matching, and successfully
matched the groups on variables they were significantly different in
prior to matching.
I now would like to run a cox-proportional hazards survival analysis
(death or mi as the event) on the matched data set. Almost all the
method sections in manuscripts I'm referring to ,use the propensity
score as a covariate in the regression model, with or without other
covariates. I'm confused as to why the propensity score is needed since
the groups are successfully matched already, and the propensity score
was developed to adjust for differences between the stent groups.
thanks
Jeff
Jeffrey F. Mather
Director of Data Management
Research Administration
Hartford Hospital
Hartford, Connecticut 06106
860-545-3560
[hidden email]
Locked
