Estimating Coherency between Survey Data and Incentivized Experimental Data

Abstract

Imagine the situation in which an econometrician can infer the distribution of welfare gains induced by the provision of higher education nancial aid using survey data obtained from a set of individuals, and can estimate the same distribution using a highly incentivized eld experiment in which the same set of individuals participated. In the experimental setting relying on incentivized choices, making the wrong decision can be costly. In the survey, the stakes are null and reporting false intentions and expectations is costless. In this paper, we evaluate the extent to which the decomposition of the two welfare gain distributions into latent factors are coherent. We nd that individuals often put a much dierent weight to a specic set of determinants in the experiment and in the survey and that the valuations of nancial aid are rank incoherent. About 66% of Biased Incoherency (dened as the tendency to have a higher valuation rank in the experiment than in the survey) is explained by individual heterogeneity in subjective benets, costs and other factors and about half of these factors aect the welfare gains of nancial aid in the survey and in the experiment in opposite directions. Ex-ante policy evaluation of a potential expansion of the Canadian higher education nancial aid system may therefore depend heavily on whether or not the data have been obtained in an \incentivized" context.

JEL codes

Reference

Christian Belzil, Julie Pernaudet, and François Poinas, “Estimating Coherency between Survey Data and Incentivized Experimental Data”, TSE Working Paper, n. 21-1234, July 2021.