ISR to Collaborate with Census Bureau to Improve Record Linkage

October 27, 2021

The U.S. Census Bureau has announced the award of cooperative agreements that will help ensure it can take advantage of advances in entity resolution and record linkage methodology and technology, including an agreement with the Inter-university Consortium for Political and Social Research (ICPSR). 

The Census Bureau aims to quantify and incorporate record linkage error and uncertainty into its data products. A cooperative agreement awarded to the University of Michigan will work toward this objective. The team will develop a multiple imputation approach to quantifying the error and uncertainty and help the Census Bureau extend its record linkage coverage for people without Social Security numbers (SSNs) by introducing a new reference dataset containing records for them. It will also incorporate household relationships into record linkage by matching couples in tax data.

Principal investigator Margaret Levenstein is the director of ICPSR and a research professor at the Survey Research Center and School of Information at the University of Michigan. Her recent research develops a probabilistic record linkage methodology that propagates linkage uncertainty when conducting inferences. J. Trent Alexander, co-principal investigator, is ICPSR’s associate director and a research professor at the Institute for Social Research. He has two decades of experience building data infrastructure projects in academia and the federal government and currently leads the Census Bureau’s Decennial Census Digitization and Linkage Project. Researchers from the Survey Research Center and the Population Studies Center will also contribute to the project.

“ICPSR is thrilled to be part of this groundbreaking work bringing new data linkage technology to U.S. Census data products,” Levenstein said.

This project will generate computer code, data products, and research findings. All computer code will be shared on the ICPSR git repository. Links to the repository, as well as to all public-use data products will be shared through LinkageLibrary, an NSF-funded repository at ICPSR to support the record linkage community. Research findings will be documented in publicly available working papers available via the University of Michigan’s Deep Blue repository, via Census Bureau Working Paper Series when applicable, and ultimately via published research papers.

For more information, please see Census.gov.