Bringing it all back home: Incentives in the age of general population sampling

Abstract

Monetary incentives have long been a cornerstone of economic experiments. However, unincentivized measures of economic preferences and skills are becoming more common as experimenters move beyond the lab to study general population samples. This paper examines how monetary incentives influence inferences about truth-telling, competitiveness, and cognitive skills using a large, nationally representative U.S. sample. We find that incentives substantially alter the levels and patterns of truth-telling, competitiveness, and cognitive skills and increase the time participants spend reading instructions and making decisions. Crucially, in numerous instances, monetary incentives affect the conclusions derived from the data concerning group differences (e.g., age groups, gender, income groups, and educational attainment) as well as the estimated associations between income and the measured preferences/skills.