Assessment of Methods of Education Imputation in the LEHD

Laura D. Quinby, Oliver K. Shih and Gal Wettstein

CES Technical Notes Series from Center for Economic Studies, U.S. Census Bureau

Abstract: This technical memorandum describes our attempt to improve upon the LEHD imputation of education. The existing imputation relies on links to the decennial Census of 2000, which leads to a degradation of the quality of imputation over time. This memo relies on the American Community Survey, which is annual, to provide more timely reference points for education. We test the relative performance of four alternative imputation methodologies: Coarsened Exact Matching, Ordinary Least Squares, Random Forest, and Boosted Trees. We split the LEHD into training and testing samples and evaluate the five approaches (our four and the original LEHD method) on the testing sample on the basis of mean squared errors. We do this for the full population and for subsamples based on year, race, ethnicity, age, gender, and quartile of income.

Keywords: LEHD ICF; LEHD ECF; LEHD EHF; ACS (search for similar items in EconPapers)
Date: 2026-07
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www2.census.gov/ces/tn/CES-TN-2026-16.pdf Abstract (application/pdf)
https://www.census.gov/about/adrm/ced/apply-for-access.html?CES-TN-2026-16 Confidential main document (application/pdf)
Researchers need to have obtained appropriate permissions.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cen:tnotes:26-16

Access Statistics for this paper

More papers in CES Technical Notes Series from Center for Economic Studies, U.S. Census Bureau Contact information at EDIRC.
Bibliographic data for series maintained by Danielle H. Sandler ().