Causal Inference
Single World Intervention Graphs (SWIGs): A Practical Guide Dana Bezuidenhout* Sarah Forthal Dana Bezuidenhout
Recent decades have witnessed substantial advancements in methodologies for estimating causal effects. In particular, the potential outcomes approach now dominates most scholarship on causality. However, despite the growing popularity of this approach, the uptake of compatible graphical methods such as Single World Intervention Graphs (SWIGs) remains limited. SWIGs are causal graphs that explicitly depict the potential outcomes of interest, thus allowing users to clearly identify the independencies required to identify the causal effect of interest. We aim to increase understanding of SWIGs and demonstrate how they can be a useful resource for epidemiologists and researchers engaged in causal research. We reviewed existing literature to create a comprehensive and user-friendly guide to using SWIGs. First, we discuss the limitations of Directed Acyclic Graphs (DAGs) under the potential outcomes framework. Then, we introduce SWIGs as a simple but powerful tool for integrating potential outcomes explicitly into causal diagrams. We provide a step-by-step guide on transforming DAGs into SWIGs that includes practical insights into constructing SWIGs under various scenarios such as confounding, mediation, and sequential randomization. Highlighting the utility of SWIGs in practice, we illustrate their application in identifying the g-formula, showcasing their capacity to make causal estimands visually explicit. This project serves as a resource for epidemiologists and researchers interested in expanding their causal inference toolkit.