Propensity Score Matching Macro - Forced Matches

[ach7210]

Dear all,

This is my first time posting. I have used the Propensity Score Matching syntax & macro (by R.Levesque adapted by J.Painter). The macro seems to works out pretty well, but I had one question:

Does anyone know how to modify the syntax in order to force a match on a specific level of a variable (for example, force gender to be exact)? Thanks so much.

Kind regards,
Aaron

[David Marso]

So we are supposed to know which specific syntax you are referring to?
Link please!!!
---

ach7210 wrote

Dear all,

This is my first time posting. I have used the Propensity Score Matching syntax & macro (by R.Levesque adapted by J.Painter). The macro seems to works out pretty well, but I had one question:

Does anyone know how to modify the syntax in order to force a match on a specific level of a variable (for example, force gender to be exact)? Thanks so much.

Kind regards,
Aaron

[ach7210]

Yes, sorry, the syntax is here: http://www.unc.edu/~painter/SPSSsyntax/propen.txt

[David Marso]

Well, taking a quick peek I would suggest that that is really not the way to write a macro.
Hard coded variable names?  Requiring manual input for n of cases?
I suspect one way to go would be to segregate the sex variable into 2 files and run separately.
Of course remove it from the LOG REG.

ach7210 wrote

Yes, sorry, the syntax is here: http://www.unc.edu/~painter/SPSSsyntax/propen.txt

[Maguin, Eugene]

To improve my understanding of your question, should I assume that the logistic regression for participation prediction includes gender?

I have never used the Levesque/Painter syntax but ...

If you say that gender is not included, doesn't that reduce to two separate propensity match operations, one for males, one for females?

If you say that gender is included, doesn't that cause less than optimal matches?

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ach7210
Sent: Thursday, August 09, 2012 9:57 AM
To: [hidden email]
Subject: Re: Propensity Score Matching Macro - Forced Matches

Yes, sorry, the syntax is here:
http://www.unc.edu/~painter/SPSSsyntax/propen.txt



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Propensity-Score-Matching-Macro-Forced-Matches-tp5714641p5714643.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
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

[Jon K Peck]

Another approach would be to use the FUZZY extension command.  It is set up for a different structure, but it includes the ability to specify a match tolerance for each variable or to construct a custom difference measure.  This requires the Python Essentials available from the SPSS Community website (www.ibm.com/developerworks/spssdevcentral).

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]




From:        ach7210 <[hidden email]>
To:        [hidden email]
Date:        08/09/2012 07:49 AM
Subject:        [SPSSX-L] Propensity Score Matching Macro - Forced Matches
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Dear all,

This is my first time posting. I have used the Propensity Score Matching
syntax & macro (by R.Levesque adapted by J.Painter). The macro seems to
works out pretty well, but I had one question:

Does anyone know how to modify the syntax in order to force a match on a
specific level of a variable (for example, force gender to be exact)? Thanks
so much.

Kind regards,
Aaron



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Propensity-Score-Matching-Macro-Forced-Matches-tp5714641.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
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