Skip To Content

Introduction to causal diagrams (a.k.a. Directed Acyclic Graphs or DAGs)



Full course description


Causal diagrams are a powerful tool for visually encoding and communicating our assumptions about how risk factors, exposures, treatments, and health outcomes are related and identifying which variables to adjust for a given analysis.

This mini-course provides a practical foundation in the basics of causal diagrams and explains the rigorous rules for using these diagrams to determine how to minimize bias in statistical analyses. The ideas will be applied to concerns about confounding, selection bias, and mediation.

Those slightly familiar with causal diagrams may find the section about the rules to be a helpful refresher.

The course begins May 1, 2024

Time commitment

The course consists of 1-2 hours of recorded lectures and a set of practice exercises designed for learners wanting to build their confidence in using these diagrams.


Upon completion of all videos, an an online assessment, participants will be awarded a certificate of completion.  

Intended audience

This course will be most valuable to those new to causal diagrams and to those needing a refresher on the rules for creating and using DAGs.

Course structure

·       Interactive lectures: 3-4 short lectures totalling 1-2 hours (including working through the exercises).

·       Live session to go over the exercises: 12:00 noon Pacific time May 7, 2024

·       Practice exercises: one in-depth “word problem” exercise and 14 little diagrams to interpret.

·       Question and Answer forum hosted by the Course Facilitator and optional office hours by appointment.

About the course facilitator

Sally Picciotto has a PhD in mathematics but later trained with causal inference methodologists Miguel Hernán and Jamie Robins. She has been guest-lecturing on causal diagrams in courses at UC Berkeley’s School of Public Health for over a decade. Her research focuses primarily on the health effects of occupational and environmental exposures in longitudinal settings.

Sign up for this course today!