Working Paper: NBER ID: w22468
Authors: Sebastian Galiani; Patrick J. McEwan; Brian Quistorff
Abstract: This paper analyzes a geographic quasi-experiment embedded in a cluster-randomized experiment in Honduras. In the experiment, average treatment effects on school enrollment and child labor were large—especially in the poorest blocks—and could be generalized to a policy-relevant population given the original sample selection criteria. In contrast, the geographic quasi-experiment yielded point estimates that, for two of three dependent variables, were attenuated. A judicious policy analyst without access to the experimental results might have provided misleading advice based on the magnitude of point estimates. We assessed two main explanations for the difference in point estimates, related to external and internal validity.
Keywords: geographic quasiexperiment; conditional cash transfers; school enrollment; child labor; external validity; internal validity
JEL Codes: O22
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| imbalance in unobserved variables (C29) | attenuation of treatment effects (C22) |
| differences in distribution of observed variables (C46) | attenuation of treatment effects (C22) |
| Lenca indigenous group (O54) | unobserved socioeconomic outcomes (I24) |
| CCT program (C88) | school enrollment (I21) |
| CCT program (C88) | child labor (J82) |
| GQE (D58) | school enrollment (I21) |
| GQE (D58) | child labor (J82) |