Here's a new paper on matching for adoption:
The Pennsylvania Adoption Exchange Improves Its Matching Process
Vincent W. Slaugh, Mustafa Akan, Onur Kesten
Tepper School of Business, Carnegie Mellon University, Pittsburgh, PA 15213
vslaugh@cmu.edu, akan@cmu.edu, okesten@andrew.cmu.edu
M. Utku Unver
Department of Economics, Boston College, Chestnut Hill, MA, 02467 unver@bc.edu
Abstract: We utilize operations research and market design frameworks to improve the e ectiveness of a recommender system for children in state custody and prospective adoptive families, and describe the implementation of changes to the Pennsylvania Adoption Exchange (PAE). The purpose of PAE is to nd permanent families for hard-to-place children in state custody, and it provides match recommendations to over 200 children each year. After describing PAE's operation, we use data provided by PAE to analyze the relationship between child attributes and adoption outcomes. With the primary objective of improving adoption outcomes, our recommendations cover the information that PAE collects to make recommendations, the decision rules that produce child-family match suggestions, and the interactions between PAE and system participants. We demonstrate the value of improved information about families and children using a discrete-event simulation of PAE. As a result of our collaboration, PAE has begun to collect new information from families and children to better predict match success, to use a spreadsheet tool that provides a more nuanced comparison of family preferences and child attributes, and to change the match recommendation process to reduce strategic manipulation by families. A survey of child case workers in Pennsylvania and interviews with PAE managers guided these changes. Insights from this project are relevant to other adoption exchanges, as well as the placement of children in foster care.
*************************************
A related paper, which I blogged about earlier, is here:
Child-Adoption Matching: Preferences for Gender and Race, 2013, by Mariagiovanna Baccara, Allan Collard-Wexler, and Leonardo Felli, and Leeat Yariv, American Economic Journal: Applied Economics, forthcoming.
The Pennsylvania Adoption Exchange Improves Its Matching Process
Vincent W. Slaugh, Mustafa Akan, Onur Kesten
Tepper School of Business, Carnegie Mellon University, Pittsburgh, PA 15213
vslaugh@cmu.edu, akan@cmu.edu, okesten@andrew.cmu.edu
M. Utku Unver
Department of Economics, Boston College, Chestnut Hill, MA, 02467 unver@bc.edu
Abstract: We utilize operations research and market design frameworks to improve the e ectiveness of a recommender system for children in state custody and prospective adoptive families, and describe the implementation of changes to the Pennsylvania Adoption Exchange (PAE). The purpose of PAE is to nd permanent families for hard-to-place children in state custody, and it provides match recommendations to over 200 children each year. After describing PAE's operation, we use data provided by PAE to analyze the relationship between child attributes and adoption outcomes. With the primary objective of improving adoption outcomes, our recommendations cover the information that PAE collects to make recommendations, the decision rules that produce child-family match suggestions, and the interactions between PAE and system participants. We demonstrate the value of improved information about families and children using a discrete-event simulation of PAE. As a result of our collaboration, PAE has begun to collect new information from families and children to better predict match success, to use a spreadsheet tool that provides a more nuanced comparison of family preferences and child attributes, and to change the match recommendation process to reduce strategic manipulation by families. A survey of child case workers in Pennsylvania and interviews with PAE managers guided these changes. Insights from this project are relevant to other adoption exchanges, as well as the placement of children in foster care.
*************************************
A related paper, which I blogged about earlier, is here:
Child-Adoption Matching: Preferences for Gender and Race, 2013, by Mariagiovanna Baccara, Allan Collard-Wexler, and Leonardo Felli, and Leeat Yariv, American Economic Journal: Applied Economics, forthcoming.
Posts
No comments:
Post a Comment