Process tracing method in the social sciences seek to assess hypothesized causal mechanisms in individual cases, but practitioners face a problem: How to use within-case evidence to evaluate singular causal relationships within hypothesized mechanism. This paper presents a partial solution to this problem by treating detailed features of the outcomes as observable consequences of the hypothesized causal relationships, given suitable auxiliary assumptions. This approach is then illustrated through analysis of a process-tracing study by Vesla Weaver. The paper concludes by examining how the feature-based approach relates to other frameworks for assessing singular causal claims.
Yafeng Wang (Mon,) studied this question.