Non-existent outcomes in research on inequality: A causal approach

• In Event: 20412 - New Methods & Data in Inequality Research

Abstract

Scholars of social stratification often study exposures that shape life outcomes. Job training may increase hourly wages, for example. But those same exposures may also determine whether outcomes exist at all: job training may help someone find employment so that they have an hourly wage when they would not otherwise. We show how a common research practice---dropping cases with non-existent outcomes---can obscure causal effects when the existence of the outcome is selective, in the sense that it is caused by the exposure of interest. We show how the effects of both beneficial and harmful treatments can be underestimated. Drawing on existing approaches for principal stratification, we show how to study (1) the average effect on whether an outcome exists and (2) the average effect on the outcome among the latent subgroup whose outcome would exist in either treatment condition. To extend our approach to the selection-on-observables settings common in applied research, we develop a framework involving regression and simulation to enable principal stratification estimates that adjust for measured confounders. We illustrate through an empirical example about the effect of motherhood on women's labor market outcomes.

Authors

• Ian Lundberg, University of California, Los Angeles

• Soonhong Cho, UCLA