SPSS Propensity Matching Macro

[Julia Alexander]

Hello, does anyone have information about propensity score calculations and
matching w/ calpier using SPSS?  I have seen this macro mentioned on the
internet but SPSS has never heard of it. Let me know.  Thank you!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[la volta statistics]

Hi Julia

look at Reynald's SPSS Tools
http://www.spsstools.net/SampleSyntax.htm#RandomSampling

for example file 14

Hope this helps
Christian

-----Ursprüngliche Nachricht-----
Von: SPSSX(r) Discussion [mailto:[hidden email]]Im Auftrag von
Julia Alexander
Gesendet: Mittwoch, 11. Juni 2008 20:51
An: [hidden email]
Betreff: SPSS Propensity Matching Macro


Hello, does anyone have information about propensity score calculations and
matching w/ calpier using SPSS?  I have seen this macro mentioned on the
internet but SPSS has never heard of it. Let me know.  Thank you!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Julia Alexander]

Thank you!
Julie

Julie Alexander, MS
Licensed Psychologist
Doctoral Candidate
DHS Behavioral Health & Wellness Center
1515 Arch Street, Mezzanine Level
Philadelphia, PA 19102
Office phone  [phone]
Fax                   215-683-5769





la volta statistics <[hidden email]>
Sent by: "SPSSX(r) Discussion" <[hidden email]>
06/12/2008 04:06 AM
Please respond to
la volta statistics <[hidden email]>


To
[hidden email]
cc

Subject
AW: SPSS Propensity Matching Macro






Hi Julia

look at Reynald's SPSS Tools
http://www.spsstools.net/SampleSyntax.htm#RandomSampling

for example file 14

Hope this helps
Christian

-----Ursprüngliche Nachricht-----
Von: SPSSX(r) Discussion [mailto:[hidden email]]Im Auftrag von
Julia Alexander
Gesendet: Mittwoch, 11. Juni 2008 20:51
An: [hidden email]
Betreff: SPSS Propensity Matching Macro


Hello, does anyone have information about propensity score calculations
and
matching w/ calpier using SPSS?  I have seen this macro mentioned on the
internet but SPSS has never heard of it. Let me know.  Thank you!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Dennis Deck]

The core of applying the approach is running Logistic Regression.  I'm
not aware of a macro (or why you would need one in this case).

There are many online sources about propensity score analysis.  A couple
starter resources are:
  http://en.wikipedia.org/wiki/Propensity_score_matching
  http://www.chrp.org/love/ASACleveland2003Propensity.pdf
  http://www.urban.org/toolkit/data-methods/propensity.cfm

A couple formal articles follow.  First is a classic, one of the
introductory articles.
Other two address more practical issues about actually conducting one:

 - Rosenbaum PR, Rubin DB. The central role of the propensity score in
observational studies for causal effects. Biometrika. 1983;70:41-55.

 - D'Agostino RB. Tutorial in Biostatistics: Propensity score methods
for bias reduction in the comparison of a treatment to a non-randomized
control group. Statistics in Medicine. 1998;17(19):2265-2281.

 - Shadish, WR, Clark MH. An introduction to propensity scores.
Metodologia de las Ciencias del Comportamiento Journal.
2002;4(2):291-300.

There are various ways to apply the propensity score in the substantive
analysis once you have assessed overlap of the propoensity score between
the groups besides matching. One is Inverse Probability of Treatment
Weighted (IPTW) which I believe was developed by Raudenbush.

Dennis Deck, PhD
RMC Research Corporation
111 SW Columbia Street, Suite 1200
Portland, Oregon 97201-5843
voice: 503-223-8248 x715
voice: 800-788-1887 x715
fax:  [phone]
[hidden email]



-----Original Message-----
From: Julia Alexander [mailto:[hidden email]]
Sent: Wednesday, June 11, 2008 11:51 AM
Subject: SPSS Propensity Matching Macro

Hello, does anyone have information about propensity score calculations
and matching w/ calpier using SPSS?  I have seen this macro mentioned on
the internet but SPSS has never heard of it. Let me know.  Thank you!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Anton-24]

Hi Dennis-

Back in August 2006 you posted code to conduct an analysis using inverse
probability of treatment weights (IPTW), e.g.:

DO IF (CD=1) .
+ COMPUTE iCDp = 1 / CDp .
ELSE . + COMPUTE iCDp = 1 / (1-CDp) .
END IF .

You then turned the weights on and conducted a logistic regression to
compare treatment vs. control.

Am I correct to assume that once the weights are applied, one is free to
conduct inferential techniques comparing treatment vs. control groups?  I
have been reading about issues surrounding inaccurate standard errors with
some propensity score techniques and subsequent analyses.

Thanks in advance for your advice.

Anton




On Thu, 12 Jun 2008 12:00:05 -0700, Dennis Deck <[hidden email]> wrote:

>The core of applying the approach is running Logistic Regression.  I'm
>not aware of a macro (or why you would need one in this case).
>
>There are many online sources about propensity score analysis.  A couple
>starter resources are:
>  http://en.wikipedia.org/wiki/Propensity_score_matching
>  http://www.chrp.org/love/ASACleveland2003Propensity.pdf
>  http://www.urban.org/toolkit/data-methods/propensity.cfm
>
>A couple formal articles follow.  First is a classic, one of the
>introductory articles.
>Other two address more practical issues about actually conducting one:
>
> - Rosenbaum PR, Rubin DB. The central role of the propensity score in
>observational studies for causal effects. Biometrika. 1983;70:41-55.
>
> - D'Agostino RB. Tutorial in Biostatistics: Propensity score methods
>for bias reduction in the comparison of a treatment to a non-randomized
>control group. Statistics in Medicine. 1998;17(19):2265-2281.
>
> - Shadish, WR, Clark MH. An introduction to propensity scores.
>Metodologia de las Ciencias del Comportamiento Journal.
>2002;4(2):291-300.
>
>There are various ways to apply the propensity score in the substantive
>analysis once you have assessed overlap of the propoensity score between
>the groups besides matching. One is Inverse Probability of Treatment
>Weighted (IPTW) which I believe was developed by Raudenbush.
>
>Dennis Deck, PhD
>RMC Research Corporation
>111 SW Columbia Street, Suite 1200
>Portland, Oregon 97201-5843
>voice: 503-223-8248 x715
>voice: 800-788-1887 x715
>fax:  [phone]
>[hidden email]
>
>
>
>-----Original Message-----
>From: Julia Alexander [mailto:[hidden email]]
>Sent: Wednesday, June 11, 2008 11:51 AM
>Subject: SPSS Propensity Matching Macro
>
>Hello, does anyone have information about propensity score calculations
>and matching w/ calpier using SPSS?  I have seen this macro mentioned on
>the internet but SPSS has never heard of it. Let me know.  Thank you!
>
>=====================
>To manage your subscription to SPSSX-L, send a message to
>[hidden email] (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

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Dennis Deck]

UPDATE:
Out of curiosity I did a quick search on propensity and macro and came
up with one written by John Painter.
Find link in: http://ssw.unc.edu/VRC/Lectures/index.htm
I have not looked at it yet but consider the source reliable.
I would recommend gaining understanding of the technique first before
applying any macro.

As I indicated you do not need a macro to apply propensity scores using
SPSS but reading his code may help you plan your own analysis.

------

The core of applying the approach is running Logistic Regression.  I'm
not aware of a macro (or why you would need one in this case).

There are many online sources about propensity score analysis.  A couple
starter resources are:
  http://en.wikipedia.org/wiki/Propensity_score_matching
  http://www.chrp.org/love/ASACleveland2003Propensity.pdf
  http://www.urban.org/toolkit/data-methods/propensity.cfm

A couple formal articles follow.  First is a classic, one of the
introductory articles.
Other two address more practical issues about actually conducting one:

 - Rosenbaum PR, Rubin DB. The central role of the propensity score in
observational studies for causal effects. Biometrika. 1983;70:41-55.

 - D'Agostino RB. Tutorial in Biostatistics: Propensity score methods
for bias reduction in the comparison of a treatment to a non-randomized
control group. Statistics in Medicine. 1998;17(19):2265-2281.

 - Shadish, WR, Clark MH. An introduction to propensity scores.
Metodologia de las Ciencias del Comportamiento Journal.
2002;4(2):291-300.

There are various ways to apply the propensity score in the substantive
analysis once you have assessed overlap of the propoensity score between
the groups besides matching. One is Inverse Probability of Treatment
Weighted (IPTW) which I believe was developed by Raudenbush.

Dennis Deck, PhD
RMC Research Corporation
111 SW Columbia Street, Suite 1200
Portland, Oregon 97201-5843
voice: 503-223-8248 x715
voice: 800-788-1887 x715
fax:  [phone]
[hidden email]



-----Original Message-----
From: Julia Alexander [mailto:[hidden email]]
Sent: Wednesday, June 11, 2008 11:51 AM
Subject: SPSS Propensity Matching Macro

Hello, does anyone have information about propensity score calculations
and matching w/ calpier using SPSS?  I have seen this macro mentioned on
the internet but SPSS has never heard of it. Let me know.  Thank you!

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Maguin, Eugene]

The solution I offered to this question was just plain wrong. Richard
Oliver's works.

Gene Maguin

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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