The Center for Social Data Analytics Colloquium speaker: Kaitlyn Webb

Google map

Title:  "Anonymization Is Dead (Long Live Formal Data Privacy)!"

Abstract: Traditional disclosure avoidance methods like de-identification, cell suppression, and row swapping are increasingly vulnerable in today's data-rich and advanced computing world. These methods often provide far weaker privacy protection than researchers realize and provide limited transparency about how released data are altered. This talk introduces formal privacy and differential privacy as practical alternatives for social science research with mathematically provable privacy guarantees. Two examples show how these tools can be applied in practice with social science data: a privacy framework for Quarterly Census of Employment and Wages (QCEW) data that preserves analytical usefulness in highly skewed datasets, and an R package that generates privacy-preserving synthetic data for replication packages. Together, these examples show how formal privacy tools can support data accessibility, reproducibility, and confidentiality in contemporary social science research.

Bio: Kaitlyn Webb is a Ph.D. candidate in the Department of Statistics at Penn State University who is pursuing a dual title in Statistics and Social Data Analytics and anticipates graduating in summer 2026. Her research focuses on statistical data privacy, such as formal privacy and differential privacy, with applications to government statistics and economic data. Webb is a Janet L. Norwood Science Achievement Graduate Fellow in Penn State’s Eberly College of Science. Webb’s work has earned national and international recognition, including the 2023 Wray-Jackson Smith Scholarship and a student paper award at the 2024 International Conference for Establishment Statistics. She is actively engaged in the Penn State community, serving as president of the Statistics Graduate Student Association in 20and mentored new statistics PhD students and an undergraduate.  Her career goals are to pursue a postdoc or faculty role in academia, where she can serve her community, collaborate on research, and teach future statisticians.