Causal Inference for Dummies: A Tutorial on Directed Acyclic Graphs and Balancing Weights

Traditionally, causal claims in social cognition research have been reserved for experimental designs. However, restricting causal claims to experimental research limits the type of questions that can be answered satisfactorily – including questions about geographical differences or changes over time recently popularized in the field of social cognition. In this tutorial, we outline a principled approach to causal inference for non-experimental designs. We describe how researchers can use Directed Acyclic Graphs to make their causal model explicit and discuss one strategy to estimate causal effects: Balancing weights. We show how researchers can use balancing weights to obtain unbiased causal effects from non-experimental designs. We provide detailed R Code to implement balancing weights analyses and provide readers with resources to delve deeper into the field of causal inference.