Department Seminar: Jeff Smith

"Quantifying Non-Sampling Variation: College Quality and the Garden of Forking Paths"

Abstract:  "Empirical economics papers report standard errors to take into account uncertainty associated with sampling variation but rarely consider non-sampling variation from researcher choices about measurement of key variables, functional form choice, identification strategy, and data set. In this paper, we review the literature on alternative methods for taking account of non-sampling variability, develop a typology of sources of non-sampling variation, and conduct an empirical exercise in which we estimate the relative and absolute importance of different types of non-sampling variation. The empirical exercise proceeds in the context of the literature that seeks to estimate the causal effect of college quality on educational and labor market outcomes."