TY - JOUR
T1 - Comparing Inference Approaches for RD Designs
T2 - A Reexamination of the Effect of Head Start on Child Mortality
AU - Cattaneo, Matias D.
AU - Titiunik, Rocío
AU - Vazquez-Bare, Gonzalo
N1 - Funding Information:
We thank Martha Bailey, Jake Bowers, Sebastian Calonico, Max Farrell, Xinwei Ma, Doug Miller, Ariel Pihl, Tomás Rau, and Elizabeth Stuart for comments and suggestions. We also thank Jens Ludwig and Doug Miller for sharing their original dataset. Finally, we thank the Editor in Chief, the Methods Section Editor, and four anonymous reviewers for their insightful and detailed comments, which greatly improved our manuscript. Financial support from the National Science Foundation (SES 1357561) is gratefully acknowledged.
Publisher Copyright:
© 2017 by the Association for Public Policy Analysis and Management
PY - 2017/6/1
Y1 - 2017/6/1
N2 - The regression discontinuity (RD) design is a popular quasi-experimental design for causal inference and policy evaluation. The most common inference approaches in RD designs employ “flexible” parametric and nonparametric local polynomial methods, which rely on extrapolation and large-sample approximations of conditional expectations using observations somewhat near the cutoff that determines treatment assignment. An alternative inference approach employs the idea of local randomization, where the very few units closest to the cutoff are regarded as randomly assigned to treatment and finite-sample exact inference methods are used. In this paper, we contrast these approaches empirically by re-analyzing the influential findings of Ludwig and Miller (), who studied the effect of Head Start assistance on child mortality employing parametric RD methods. We first review methods based on approximations of conditional expectations, which are relatively well developed in the literature, and then present new methods based on randomization inference. In particular, we extend the local randomization framework to allow for parametric adjustments of the potential outcomes; our extended framework substantially relaxes strong assumptions in prior literature and better resembles other RD inference methods. We compare all these methods formally, focusing on both estimands and inference properties. In addition, we develop new approaches for randomization-based sensitivity analysis specifically tailored to RD designs. Applying all these methods to the Head Start data, we find that the original RD treatment effect reported in the literature is quite stable and robust, an empirical finding that enhances the credibility of the original result. All the empirical methods we discuss are readily available in general purpose software in R and Stata; we also provide the dataset and software code needed to replicate all our results.
AB - The regression discontinuity (RD) design is a popular quasi-experimental design for causal inference and policy evaluation. The most common inference approaches in RD designs employ “flexible” parametric and nonparametric local polynomial methods, which rely on extrapolation and large-sample approximations of conditional expectations using observations somewhat near the cutoff that determines treatment assignment. An alternative inference approach employs the idea of local randomization, where the very few units closest to the cutoff are regarded as randomly assigned to treatment and finite-sample exact inference methods are used. In this paper, we contrast these approaches empirically by re-analyzing the influential findings of Ludwig and Miller (), who studied the effect of Head Start assistance on child mortality employing parametric RD methods. We first review methods based on approximations of conditional expectations, which are relatively well developed in the literature, and then present new methods based on randomization inference. In particular, we extend the local randomization framework to allow for parametric adjustments of the potential outcomes; our extended framework substantially relaxes strong assumptions in prior literature and better resembles other RD inference methods. We compare all these methods formally, focusing on both estimands and inference properties. In addition, we develop new approaches for randomization-based sensitivity analysis specifically tailored to RD designs. Applying all these methods to the Head Start data, we find that the original RD treatment effect reported in the literature is quite stable and robust, an empirical finding that enhances the credibility of the original result. All the empirical methods we discuss are readily available in general purpose software in R and Stata; we also provide the dataset and software code needed to replicate all our results.
UR - http://www.scopus.com/inward/record.url?scp=85017166894&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85017166894&partnerID=8YFLogxK
U2 - 10.1002/pam.21985
DO - 10.1002/pam.21985
M3 - Article
C2 - 28654224
AN - SCOPUS:85017166894
SN - 0276-8739
VL - 36
SP - 643
EP - 681
JO - Journal of Policy Analysis and Management
JF - Journal of Policy Analysis and Management
IS - 3
ER -