Causal Feature Learning in the Social Sciences
This addresses variable selection issues for researchers in social sciences, but it appears incremental as an extension of an existing theoretical framework.
The paper tackles the challenge of variable selection in causal modeling for social sciences by extending the Causal Feature Learning (CFL) framework to model attributes as macro-level abstractions, and it applies the CFL algorithm to diverse datasets to compare CFL-derived macrostates with traditional microstates in downstream tasks.
Variable selection poses a significant challenge in causal modeling, particularly within the social sciences, where constructs often rely on inter-related factors such as age, socioeconomic status, gender, and race. Indeed, it has been argued that such attributes must be modeled as macro-level abstractions of lower-level manipulable features, in order to preserve the modularity assumption essential to causal inference. This paper accordingly extends the theoretical framework of Causal Feature Learning (CFL). Empirically, we apply the CFL algorithm to diverse social science datasets, evaluating how CFL-derived macrostates compare with traditional microstates in downstream modeling tasks.