Research Design Meets Market Design: Using Centralized Assignment for Impact Evaluation
A growing number of school districts use centralized assignment mechanisms to allocate school seats in a manner that reflects student preferences and school priorities. Many of these assignment schemes use lotteries to ration seats when schools are oversubscribed. The resulting random assignment opens the door to credible quasi-experimental research designs for the evaluation of school effectiveness. Yet the question of how best to separate the lottery-generated variation integral to such designs from non-random preferences and priorities remains open. This paper develops easily-implemented empirical strategies that fully exploit the random assignment embedded in the widely-used deferred acceptance mechanism and its variants. We use these methods to evaluate charter schools in Denver, one of a growing number of districts that integrate charter and traditional public schools in a unified assignment system. The resulting estimates show large achievement gains from charter school attendance. Our approach expands the scope for impact evaluation by maximizing the number of students and schools that can be studied using random assignment. We also show how to use DA to identify causal effects in models with multiple school sectors.
We thank Alisha Chiarelli, Brian Eschbacher, Van Schoales, and the staff at Denver Public Schools for answering our questions and facilitating access to data. Isaiah Andrews, Eduardo Azevedo, Gary Chamberlain, Victor Chernozhukov, Jerry Hausman, Peter Hull, Hide Ichimura, Guido Imbens, Rafael Lalive, Edward Lazear, Jacob Leshno, Anna Mikuscheva, Chris Walters and seminar participants at Harvard, SOLE, Stanford University, the Fall 2015 NBER market design meeting, Kyoto, Osaka, and Tokyo provided helpful feedback. We're especially indebted to Mayara Felix and Ye Ji Kee for expert research assistance and to MIT SEII program manager Annice Correia for invaluable administrative support. We gratefully acknowledge funding from the Laura and John Arnold Foundation and the National Science Foundation (under awards SES-1056325 and SES-1426541). Data from Denver Public Schools were made available to us through the Institute for Innovation in Public School Choice. Abdulkadiroglu and Pathak are Scientific Advisory Board members of the Institute for Innovation in Public School Choice. Angrist's daughter teaches at a Boston charter school. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
Parag A. Pathak
Pathak is a member of the Scientific Advisory Board of the Institute for Innovation in Public School Choice.
Atila Abdulkadiroğlu & Joshua D. Angrist & Yusuke Narita & Parag A. Pathak, 2017. "Research Design Meets Market Design: Using Centralized Assignment for Impact Evaluation," Econometrica, Econometric Society, vol. 85, pages 1373-1432, September. citation courtesy of