Using Auxiliary Data to Guide the Recruitment of Sites for Randomized Controlled Trials

Abstract

Sampling methods such as Stratified Random Sampling can be used to select representative samples of schools for randomized controlled trials of educational interventions. However, these methods may still yield external validity bias when participation by schools is voluntary and participation decisions are associated with unobserved variables. This paper offers a new sampling method called Stratified Random Sampling with Quotas. Under this method, quotas are set to avoid including too many schools of a particular type, as defined by auxiliary variables that are unobserved in the sampling frame, but whose population distribution can be estimated from external data. Our simulations find that when the auxiliary variables affect whether or not a school participates in the study, quotas set based on those variables reduce external validity bias. These results suggest that when auxiliary data are available on strong impact moderators for the target population, these data can be used to address non-ignorable self-selection by schools into randomized controlled trials.