Date: Thu, 25 Feb 2010 07:55:08 -0800
Reply-To: J P <jp7837@yahoo.com>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: J P <jp7837@yahoo.com>
Subject: Re: Practical question regarding the Propensity Score Matching
macro (probably an easy question).
In-Reply-To: <6C7978C7CEE1CE42B887DD3909B991B301D03A8B@PHSXMB18.partners.org>
Content-Type: multipart/alternative;
You might find materials located here useful:
http://ssw.unc.edu/VRC/Lectures/index.htm
________________________________
From: "Ergul, Emel A." <EERGUL@PARTNERS.ORG>
To: SPSSX-L@LISTSERV.UGA.EDU
Sent: Thu, February 25, 2010 9:18:50 AM
Subject: Re: Practical question regarding the Propensity Score Matching macro (probably an easy question).
Hi Jane:
1. There is a great article on propensity (easy to read) written by Ralph
D'agostino published in statistics in medicine (1998).
2. I have to tell you that myself is not an expert either. I had used
stratification model once in the past. If you would go with stratification,
technically, once you estimate propensity score by logistic regression (or
discriminant analysis), you have to decide about stratum boundaries (combined
group propensity score or treated or control group alone).
I think the article above is well articulated and very helpful to understand the
mechanics of propensity.
Good luck!
emel
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
[text/html]