Project Grant 2217456

This National Science Foundation project grant of $400,000 awarded on August 15, 2022 through July 31, 2025 under the Social, Behavioral, and Economic Sciences program (CFDA 47.075) will fund research at Duke University to advance statistical and computational methods for releasing high-quality synthetic data as public use files. The research aims to develop novel techniques to improve disclosure risk assessment, quality verification for data analysts, and population generalizability when incorporating complex survey designs into synthetic data generation. A $100,000 sub-award to the University of Michigan will support extending methodology development on synthetic population generation and data integration to enhance confidentiality protection through collaborations with the Panel Study of Income Dynamics and Inter-university Consortium for Political and Social Research. The research outcomes will provide federal agencies and other data producers with improved methods for creating safer and more analytically useful synthetic data products.