JPSM Distinguished Lecture 2026

"Life trajectories and life chances: New approaches from population registries and AI”

Thursday, April 09, 2026

3:00 PM

McKeldin Library, Special Events Room 6137 (6th Floor)

University of Maryland, College Park

In-Person Event

RSVP

Matthew J. Salganik

Alexander Stewart 1886 Professor of Sociology

Princeton University

Abstract:
How do life trajectories evolve over time? How much do birth circumstances constrain where people end up? I describe two works in progress that address these questions with new approaches using population-scale registry data from the Netherlands and methods inspired by AI. First, we develop the "book of life" approach, which represents event-based life course data as narrative text rather than as a conventional numeric feature vector, and then fine-tune an open-weight large language model on millions of these narratives to predict life outcomes such as having a child. Second, using 200,000 twin pairs who share birth circumstances — everything measured and unmeasured about family, social position, and historical timing — we estimate how much of life is determined at birth versus shaped by what comes after. This approach allows us to measure fundamental limits to prediction, no matter how sophisticated the model or complete the data. Together, these projects illustrate an emerging direction for social data science in which AI and prediction become tools for developing new kinds of understanding.

Sponsored by: Joint Program in Survey Methodology, University of Maryland, College Park, MD.

Co-Sponsor by: Washington Statistical Society (WSS).