NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH
loading...

How Well Do Automated Linking Methods Perform? Lessons from U.S. Historical Data

Martha Bailey, Connor Cole, Morgan Henderson, Catherine Massey

NBER Working Paper No. 24019
Issued in November 2017, Revised in May 2019
NBER Program(s):Program on the Economics of Aging, Program on the Development of the American Economy, Labor Studies Program

This paper reviews the literature in historical record linkage in the U.S. and examines the performance of widely-used automated record linking algorithms in two high-quality historical datasets and one synthetic ground truth. Focusing on algorithms in current practice, our findings highlight the important effects of linking methods on data quality. We find that (1) no method (including hand-linking) consistently produces representative samples; (2) 15 to 37 percent of links chosen by prominent machine linking algorithms are identified as false links by human reviewers; and (3) these false links are systematically related to baseline sample characteristics, suggesting that machine algorithms may introduce complicated forms of bias into analyses. We find that prominent linking algorithms attenuate estimates of the intergenerational income elasticity by up to 20 percent and common variations in algorithm choices result in greater attenuation. These results recommend that current practice could be improved by placing more emphasis on reducing false links and less emphasis on increasing match rates. We conclude with constructive suggestions for reducing linking errors and directions for future research.

download in pdf format
   (4465 K)

email paper

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w24019

Users who downloaded this paper also downloaded* these:
Aghion, Bergeaud, Boppart, Klenow, and Li w24023 Missing Growth from Creative Destruction
Mankiw and Reis w24043 Friedman's Presidential Address in the Evolution of Macroeconomic Thought
Aghion, Jones, and Jones w23928 Artificial Intelligence and Economic Growth
Davis, Guryan, Hallberg, and Ludwig w23925 The Economics of Scale-Up
Johnson w24027 Measuring Global Value Chains
 
Publications
Activities
Meetings
NBER Videos
Themes
Data
People
About

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us