Causal Inference Grounded in Causal Diagrams: Benefits, Limitations, and Opportunities for the Information Systems Field

We join the important effort of embracing diverse views on causality reflected in a prior MIS Quarterly Editor’s Comment (Mithas et al., 2022b). Specifically, we aim to expand the discussion around a major causal framework and toolkit that, we believe, is largely missing and needed in empirical studies in the field of information systems: that of causal diagrams and structural causal modeling (SCM). Being relatively new, the SCM framework has faced resistance from economists that has only recently begun to soften and, for this reason, remains largely absent in econometrics textbooks and in the curriculum of most information systems PhD programs. In contrast to viewpoints emphasizing the dichotomy between SCM and econometrics or potential outcomes approaches, we explain how SCM can serve as a complementary layer of identification and communication that aligns with such proven design and analysis frameworks. We discuss the current limitations of the SCM framework and opportunities for new research.

• 저널MIS Quarterly · 49(2) · 409–428
• 토픽Data Quality and Management · Management Science and Operations Research
• DOI10.25300/misq/2024/18422
• 저자Ali Tafti, Galit Shmueli