The Promise of Process Tracing: Case selection in a World of Causal Complexity

Abstract

Case studies are widely used in political science, providing a richness of detail that standard quantitative studies often lack. Process-tracing case studies have gained significant prominence recently for their promise to use this strength to explicate causal mechanisms behind robust empirical correlations. Effective case selection is obviously central for process tracing to deliver on this promise, but there are no generally applicable methods for case selection when exploring complex relationships that are non-linear and feature multiple causal mechanisms. To fill this gap, we develop guidelines for selecting process-tracing cases that focus on the functional form of the relationship between relevant variables and the distribution of possible causal mechanisms. Through general examples and specific applications, we show how this approach can reveal opportunities for researchers, while failing to account for these factors can result in poor case selection, mistaken inferences, and an inability to generalize.