Suanna Oh

Title

A Recall Method for Eliciting Choices and Beliefs

Abstract

Accurately eliciting people’s preferences and beliefs is a central focus of social science research. Without incentives for accuracy, respondents may answer inattentively and compromise on accuracy. But conventional incentive schemes cannot be used to incentivize hypothetical or subjective questions. We introduce the “recall method”, which incentivizes survey choices by telling respondents that they will be given a randomly selected subset of the questions they had previously answered at the end of the survey and paid to restate their answers correctly. If answers are easier to restate when they are carefully thought-out and truthful, then participants will have an incentive to respond more truthfully. We run a Prolific experiment to test this hypothesis. Our results suggest that the method can increase respondents’ internal consistency without compromising accuracy. We benchmark the effects against other payment schemes, including conventional incentives and a gift exchange treatment.