Measurement error and the returns to excess schooling

Abstract

This paper considers the degree of measurement error in estimates of required schooling for jobs and the bias created by errors when estimating the returns to overeducation. Using the Panel Study of Income Dynamics, it finds substantial differences between estimates of required schooling. Thus errors in measurement may seriously bias previous results examining the wage effects of overeducation. This paper uses instrumental variable techniques to correct for measurement error and finds no significant returns for surplus schooling.