Abstract

There are generally two fundamental approaches to conducting counterfactual decomposition in social scientific research, but their distinction is still elusive. This study makes contributions to the literature by (1) clarifying the rationale for the two counterfactual decomposition approaches within Rubin’s counterfactual causal inference framework; (2) reviewing prior sociological and econometric literature that underscores the distinction between these two approaches; (3) distinguishing between them in terms of the equivalence of empirical results, the property of additive decomposability, and the strategies for constructing treatment and control conditions; (4) illustrating these methodological differences with an empirical case; and (5) addressing the different types of substantive questions these two approaches are, respectively, able to answer. This article concludes with a discussion about some common limitations of these two counterfactual decomposition approaches.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call