Control Group "Start Date" recidivism study

[DKUKEC]

I am working on a project that is examining the impact of adult drug court on recidivism rates using historical administrative data (2004-2011).  In addition to examining the recidivism rates for drug court participants, I also wanted to include a control group.  

For drug court participants, the study will track criminal conviction after they have signed-up (program start date) for the drug court program; this will enable us to track convictions during program and after they have completed the program.  While the program group “start date” is straight forward, I am having difficulty identifying the control group “start date”, especially when they have multiple arrest dates during the study period.

While I have reviewed a number of articles that examine the effectiveness of drug courts; which employ propensity score matching using a “contemporaneous design”, there is no in-depth discussion concerning how to select a control - and identify a "start date" - on the ground so to speak.  

Any suggestions or references would be greatly appreciated.  I apologize in advance for my question, given that it is more about methods than SPSS.

Thank you
Damir

[Maguin, Eugene]

Damir,
How have you written the selection criteria for the comparison pool, by which I mean the people who will be 'matched' with drug court folks assuming the use of propensity scoring? Also, would you define 'contemporaneous design' and give a citation or two to articles using it. It's a term I've never heard before and I curious about its mechanics.

Gene Maguin


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of DKUKEC
Sent: Friday, February 01, 2013 2:50 PM
To: [hidden email]
Subject: Control Group "Start Date" recidivism study

I am working on a project that is examining the impact of adult drug court on recidivism rates using historical administrative data (2004-2011).  In addition to examining the recidivism rates for drug court participants, I also wanted to include a control group.

For drug court participants, the study will track criminal conviction after they have signed-up (program start date) for the drug court program; this will enable us to track convictions during program and after they have completed the program.  While the program group “start date” is straight forward, I am having difficulty identifying the control group “start date”, especially when they have multiple arrest dates during the study period.

While I have reviewed a number of articles that examine the effectiveness of drug courts; which employ propensity score matching using a “contemporaneous design”, there is no in-depth discussion concerning how to select a control
- and identify a "start date" - on the ground so to speak.

Any suggestions or references would be greatly appreciated.  I apologize in advance for my question, given that it is more about methods than SPSS.

Thank you
Damir




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Control-Group-Start-Date-recidivism-study-tp5717877.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

[DKUKEC]

Hi Gene,

Thank you for replying.  Yes, there are a number of "selection" criterion which will be based on FL statutue.  I am planning to select the control group in two steps: 1) from a pool of indiviudals that meet the statutuory requirements between 2004 - 2011; and 2) with PSM taking into account demographics, and other historical data.  Knowing which "start date" for the control group is important since previous studies have included "prior arrests" (convictions) by types (drug, violent, felony, misdemeanor, etc.) as part of the selection process using PSM - I hope that made sense?  Developing these variables is easy for the program group, however, not so mucj for the control group.  I was thinking of trying to first select individuals who have the closest arrest dates with the program participant start date and then constructing the variables noted above; however, I think I need to turn over a few more rocks first.

The following like two of the sources which address your questions.  Please let me know if you require further clarifiaction.

See Chapter 11. page 143 of the PDF file.  They have a pretty good description of the process.
http://www.courts.state.ny.us/whatsnew/pdf/NYSAdultDrugCourtEvaluation.pdf

http://www.hsrmethods.org/Glossary/Terms/N/Non%20randomized%20Trial%20with%20Contemporaneous%20Controls.aspx

Thank you
Damir
 

[Melissa Ives]

If you are selecting comparisons ' who have the closest arrest dates with the program participant start date' is there a reason you would not use that arrest date as the start date for that control?
Melissa

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of DKUKEC
Sent: Saturday, February 02, 2013 2:24 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Control Group "Start Date" recidivism study

Hi Gene,

Thank you for replying.  Yes, there are a number of "selection" criterion which will be based on FL statutue.  I am planning to select the control group in two steps: 1) from a pool of indiviudals that meet the statutuory requirements between 2004 - 2011; and 2) with PSM taking into account demographics, and other historical data.  Knowing which "start date" for the control group is important since previous studies have included "prior arrests" (convictions) by types (drug, violent, felony, misdemeanor, etc.) as part of the selection process using PSM - I hope that made sense?
Developing these variables is easy for the program group, however, not so mucj for the control group.  I was thinking of trying to first select individuals who have the closest arrest dates with the program participant start date and then constructing the variables noted above; however, I think I need to turn over a few more rocks first.

The following like two of the sources which address your questions.  Please let me know if you require further clarifiaction.

See Chapter 11. page 143 of the PDF file.  They have a pretty good description of the process.
http://www.courts.state.ny.us/whatsnew/pdf/NYSAdultDrugCourtEvaluation.pdf

http://www.hsrmethods.org/Glossary/Terms/N/Non%20randomized%20Trial%20with%20Contemporaneous%20Controls.aspx

Thank you
Damir




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Control-Group-Start-Date-recidivism-study-tp5717877p5717891.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

PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

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