November 06, 2026 2:00 pm - 3:00 pm ET
Statistics and Data Science Seminar Series
LGRT 1685
Abstract: Broadly across the sciences, a disparity is taken to represent a difference in outcomes that is in some sense unfair, unjust, or associated with disadvantage. Accordingly, many frameworks for measuring disparity begin by observing some difference in outcomes and attempting to quantify the portion that is morally concerning (a decomposition) in a causal sense. In this talk, I will outline why this intuitive proposal faces serious challenges that limit its valid application in practical settings. I will argue the merits of an alternate model that is descriptive and avoids decomposition by merely comparing groups who are similarly situated on relevant characteristics. If time permits, I will also discuss how this model can be extended to evaluate the causal effects of hypothetical interventions on disparity.
 
Please see this link for the related paper:
 
John W. Jackson. 2026. Causal Inference in Health Disparities Research. Annual Review of Public Health 47:1-18.https://doi.org/10.1146/annurev-publhealth-071123-032637
 

Bio: Dr. John Jackson is an Associate Professor of Epidemiology at the Johns Hopkins Bloomberg School of Public Health with joint appointments in the Departments of Mental Health and Biostatistics. His research primarily focuses on the development, application, and dissemination of epidemiologic methods with a particular focus on translational health disparities research. His research earned him the NIH Matilda White Rely Honors Award for Early-Stage Investigators in 2022. His work has been funded by the Johns Hopkins University Provost's Office, the Robert E. Meyerhoff Foundation, and the National Heart, Lung, and Blood Institute. He serves as an Editor-in-Residence at the Americal Journal of Epidemiology and also serves on the Editorial Boards of Epidemiology and also of Sociological Methods & Research.