Directed acyclic graph (DAG)
Research Methods
Definition. A directed acyclic graph is a causal diagram in which arrows represent assumed causal effects and no path returns to its starting point.
Explanation
Greenland, Pearl and Robins introduced causal diagrams to epidemiology (Greenland et al., 1999). DAGs identify confounders, mediators and colliders, and help analyse selection bias (Arah, 2019).
Related concepts
Causal inference Confounding Mediation analysis
Read on The Health Topics
References
Arah, O. A. (2019). Analyzing selection bias for credible causal inference when in doubt, DAG it out. EPIDEMIOLOGY, 30.0 (4), 517–520. https://doi.org/10.1097/EDE.0000000000001033
Greenland, S., Pearl, J., & Robins, J. M. (1999). Causal diagrams for epidemiologic research. Epidemiology, 10(1), 37–48.
Cite this entry: Kurutkan, M. N. (2026). Directed acyclic graph (DAG). In The Health Topics glossary. https://healthtopic.org/glossary/directed-acyclic-graph/
Articles that discuss this concept
5 articles on The Health Topics mention this concept. Most relevant first.
- How to Use Directed Acyclic Graphs (DAGs): A Guide for Health Management Researchers 57
- The Marburg Declaration: Advancing Psychological Treatment 2
- Machine Learning for Public Health Policy Evaluation 1
- Confounder Selection in High-Impact Medical Journals 1
- The Sense and Nonsense of Effect Size: What Does Funder and Ozer’s (2019) Call Mean for Health Research? 1