Bounding the Effects of Social Experiments: Accounting for Attrition in Administrative Data
32 Pages Posted: 1 Mar 2013 Last revised: 22 Dec 2024
Date Written: February 2013
Abstract
Social experiments frequently exploit data from administrative records. However, most administrative data systems are state-specific, designed to track earnings or benefit payments among residents within a single state. Once an experimental participant moves out of state, his earnings and benefits in his state of origin consist entirely of zeros, giving rise to a form of attrition. In the presence of such attrition, the average treatment effect of the experiment is no longer point-identified, despite random assignment. I propose a method to estimate such attrition and, for binary outcomes such as employment, to construct bounds on the average treatment effect. Results from a welfare-reform experiment considered to have sizeable effects appear quite ambiguous after accounting for attrition. The results have important implications for planning social experiments.
Suggested Citation: Suggested Citation
Do you have a job opening that you would like to promote on SSRN?
Recommended Papers
-
By Alberto Abadie, Joshua D. Angrist, ...
-
An MCMC Approach to Classical Estimation
By Victor Chernozhukov and Han Hong
-
What Mean Impacts Miss: Distributional Effects of Welfare Reform Experiments
By Marianne P. Bitler, Jonah B. Gelbach, ...
-
Cash Welfare as a Consumption Smoothing Mechanism for Single Mothers
-
Why Do the Poor and the Less-Educated Pay More for Long-Distance Calls?
By Jerry A. Hausman and J. Gregory Sidak
-
Heterogeneous Impacts in PROGRESA
By Habiba Djebbari and Jeffrey A. Smith