Author(s)
Ganesh Karapakula
James J. Heckman
This paper presents a simple decision-theoretic economic approach for analyzing social experiments with compromised random assignment protocols that are only partially documented. We model administratively constrained experimenters who satisfice in seeking covariate balance. We develop design-based small-sample hypothesis tests that use worst-case (least favorable) randomization null distributions. Our approach accommodates a variety of compromised experiments, including imperfectly documented re-randomization designs. To make our analysis concrete, we focus much of our discussion on the influential Perry Preschool Project. We reexamine previous estimates of program effectiveness using our methods. The choice of how to model reassignment vitally affects inference.
Publication Type
Working Paper
File Description
First version, August 10, 2020
JEL Codes
C10: Econometric and Statistical Methods and Methodology: General
C40: Econometric and Statistical Methods: Special Topics: General
I21: Analysis of Education
Keywords
randomized controlled trial
randomization tests
worst-case inference
least favorable null distributions
partial identification
small-sample hypothesis testing