Counterfactual Slopes and Their Applications in Social Stratification

Ang Yu,Jiwei Zhao

DOI: https://doi.org/10.48550/arXiv.2401.07000

2024-09-22

Abstract:This paper addresses two prominent theses in social stratification research, the great equalizer thesis and Mare's (1980) school transition thesis. Both theses describe the role of an intermediate educational transition in the association between socioeconomic status and an outcome variable. However, the descriptive regularities of the two theses may be driven by differential selection into the intermediate transition, which prevents the two theses from having substantive interpretations. We propose a set of novel counterfactual slope estimands, which capture these theses under hypothetical interventions that eliminate the differential selection. We thereby construct selection-free tests for these theses. Compared with the existing literature, we are the first to explicitly provide nonparametric causal estimands, which enable us to conduct more principled analysis. We are also the first to develop flexible, efficient, and robust estimators for the two theses based on efficient influence functions. We apply our framework to a nationally representative dataset in the United States and re-evaluate the two theses. Findings from our selection-free tests suggest that the descriptive regularities are misleading for the substantive interpretation of the great equalizer thesis, but not for that of the school transition thesis. Additionally, the counterfactual slopes also provide a new framework for evaluating the inequality impacts of policy interventions.

Methodology,Applications