Epi Soul logo image Epi Soul logo image
  1. UW Home |
  2. SPH  |
  3. Other Departments |
online giving

Epi Special Seminar

backBack to Epi Seminars

Tuesday, January 10, 2012

Journal Club
12:30 pm to 1:30 pm
Health Sciences F-257
Lunch provided

Causal Inference: Implications of Rubin’s Causal Model and Other Counterfactual Approaches
3:30 pm to 4:50 pm
Health Sciences K-069
Reception to follow

Dr. Sharon Schwartz
Professor of Clinical Epidemiology
Columbia University
Mailman School of Public Health

  Dr. Sharon Schwartz

Sharon Schwartz, PhD, currently focuses her work on the relationship between potential outcomes approaches to causality and systems dynamics. Dr. Schwartz is particularly interested in how methodological tools, the assumptions on which they are based and the language in which they are discussed, frame the interpretation of data. In the course of studying the effect of social factors on psychiatric disorders, Dr. Schwartz became intrigued by the diametrically opposed conclusions that scientists from different disciplines often draw from the same data. This current interest resulted in a research program that encompasses a wide range of specific methodological issues, such as heritability estimates, well controls, diagnostic validity, the definition of "disorder," as well as more general problems of causal inference, including the implications of Rubin's causal model and other counterfactual approaches, and the relationship between the Cook and Campbell tradition in social psychology and epidemiologic traditions. She also does research on the effects of prejudice and discrimination on the mental health of disadvantaged groups. Dr. Schwartz is the training coordinator for the Psychiatric Epidemiology Training Program and teaches epidemiologic methods. She participates in the Teagle Colleguim on Psychological Science and Student Learning to integrate evidenced based teaching and mentoring practices into epidemiology.

The overall theme of the UW visit will be to discuss the implications of a fairly radical change that is occurring in epidemiology with the popularization of potential outcomes (aka counterfactual) frameworks.  As these frameworks become normative, they help clarify many central epidemiologic concepts and provide new investigative tools.  At the same time, they have implications for the types of research questions that have legitimacy and for how research results are interpreted.  I hope to provide an accessible introduction to this conceptual framework and begin a conversation about its potential utility and impact on epidemiologic research.

Suggested Reading:

"When reading the article, people who plan to attend Journal Club should focus on the implications of following the advice of Talati, Fyer and Weissman to use well controls."


Updated on November 30, 2011