BREAD Working Paper No. 275, August 2010

Selective Trials: A Principal-Agent Approach to Randomized Controlled Experiments

Sylvain Chassang, Gerard Padro i Miquel, Erik Snowberg

Abstract

We study the design of randomized controlled experiments in environments where outcomes are signi cantly aff ected by unobserved e ffort decisions taken by the subjects (agents). While standard randomized controlled trials (RCTs) are internally consistent, the unobservability of eff ort provision compromises external validity. We approach trial design as a principal-agent problem and show that natural extensions of RCTs -which we call selective trials- can help improve the external validity of experiments. In particular, selective trials can disentangle the eff ects of treatment, e ffort, and the interaction of treatment and e ffort. Moreover, they can help experimenters identify when measured treatment eff ects are a ffected by erroneous beliefs and inappropriate e ffort provision.

Keywords: randomized controlled trials, selective trials, blind trials, incentivized trials, marginal treatment e ffects, mechanism design, selection, heterogeneous beliefs, compliance

JEL classification codes: C81, C93, D82, O12

Download paper in PDF format