Generalizing the Results from Social Experiments: Theory and Evidence from India

How informative are treatment effects estimated in one region or time period for another region or time? In this article, I derive bounds on the average treatment effect in a context of interest using experimental evidence from another context. The bounds are based on (a) the information identified about treatment effect heterogeneity due to unobservables in the experiment and (b) using differences in outcome distributions across contexts to learn about differences in distributions of unobservables. Empirically, using data from a pair of remedial education experiments carried out in India, I show the bounds are able to recover average treatment effects in one location using results from the other while the benchmark method cannot.

Automated summary

In this article, the author examines the informative value of treatment effects estimated in one context for another context. By using experimental evidence from a different context, the author derives bounds on the average treatment effect in the context of interest. Empirical analysis using data from remedial education experiments in India demonstrates that these bounds can recover average treatment effects in one location using results from the other, while the benchmark method fails to do so.