Tracing Causal Paths from Experimental and Observational Data

2022/04/14 by Xiang Zhou, Teppei Yamamoto · 51 citations
Mathematics · Social Sciences · #Advanced Causal Inference Techniques #Causal analysis #Causal inference #Causal model #Computer science #Data mining #Data science #Econometrics #Electoral Systems and Political Participation #Framing (construction) #Machine learning #Mathematics #Mediation #Observational study #Path analysis (statistics) #Qualitative Comparative Analysis Research #Social science #Sociology #Statistics #Tracing

paper · open access · doi:10.1086/720310

published in The Journal of Politics 85(1), 250-265 (University of Chicago Press)

openalex publication_date 2022/04/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/26

Abstract

The study of causal mechanisms abounds in political science, and causal mediation analysis has grown rapidly across different subfields. Yet, conventional methods for analyzing causal mechanisms are difficult to use when the causal effect of interest involves multiple mediators that are potentially causally dependent—a common scenario in political science applications. This article introduces a general framework for tracing causal paths with multiple mediators. In this framework, the total effect of a treatment on an outcome is decomposed into a set of path-specific effects (PSEs). We propose an imputation approach for estimating these PSEs from experimental and observational data, along with a set of bias formulas for conducting sensitivity analysis. We illustrate this approach using an experimental study on issue-framing effects and an observational study on the legacy of political violence. An open-source R package, paths, is available for implementing the proposed methods.

Citations

• Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies
• Constructing Inverse Probability Weights for Marginal Structural Models
• Public Opinion and the Democratic Peace

Cited by

• It is not what you weigh, it is how you present it: body size, attractiveness, physical functioning, and access to partnership and sexuality for older men and women
• Yours, mine, and ours: Childhood disadvantage and late-life social connectedness in marital dyads
• Recent Developments in Causal Inference and Machine Learning
• Estimating controlled direct effects with panel data: an application to reducing support for discriminatory policies
• Causal Mediation Analysis with Multiple Mediators: A Simulation Approach
• Decomposition of Probabilities of Causation with Two Mediators
• Assessing Racial Disparities in Healthcare Expenditures via Mediator Distribution Shifts
• GoT-CD: Graph-of-Thoughts Causal Discovery and the Fragility of Post-hoc Path-Specific Fairness Audits
• Childhood maltreatment and outcomes following electroconvulsive therapy in adults with depression. [europepmc]
• Invisible disabilities and college academic success: New evidence from a mediation analysis. [europepmc]
• How, and For Whom, Does Higher Education Increase Voting? [europepmc]
• Crossing Gender Boundaries: Exploring the Chain-Mediated Causal Role of Social Media Sharing in Shaping Interpersonal Networks and Enhancing Job Satisfaction. [europepmc]
• Women and Ethnic Minority Candidates Face Dynamic Party Divergent Glass Cliff Conditions in French Elections. [europepmc]
• It is not what you weigh, it is how you present it: body size, attractiveness, physical functioning, and access to partnership and sexuality for older men and women. [europepmc]

Related

• Explaining Causal Findings Without Bias: Detecting and Assessing Direct Effects
• Causal inference with latent outcomes
• Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies
• Path-specific Effects Based on Information Accounts of Causality
• Identification, Inference and Sensitivity Analysis for Causal Mediation Effects