How Well Do Automated Linking Methods Perform? Lessons from US Historical Data
Bibliographic Data
| ID | 23345366 |
|---|---|
| Authors | Martha J Bailey (University of California, Los Angeles and NBER.), Martha Bailey (0000-0002-7692-6040, University of California, Los Angeles and NBER.), Connor Cole (University of Michigan), Morgan Henderson (0000-0002-0869-5738, University of Michigan), Catherine Massey (0000-0002-3836-3857, University of Michigan) |
| Year | 2020 |
| Volume | 58 |
| Issue | 4 |
| Pages | 997-1044 |
| Publication date | 2020-12-01 |
| Peer Reviewed | Yes |
| Open Access | No |
| Type | ARTICLE |
| Venue | Journal of Economic Literature (JOURNAL) |
| Journal identifiers | ISSN: 0022-0515 • E-ISSN: 1547-1101 |
| Publisher | American Economic Association (PUBLISHER • US) |
| DOI | 10.1257/jel.20191526 |
| PMID | 34294947 |
| OpenAlex | W2944196176 |
| Language | EN |
| Citations received | 66 |
| References cited | 62 |
This paper reviews the literature in historical record linkage in the United States and examines the performance of widely used record-linking algorithms and common variations in their assumptions. We use two high-quality, hand-linked data sets and one synthetic ground truth to examine the direct effects of linking algorithms on data quality. We find that (i) no algorithm (including hand linking) consistently produces representative samples; (ii) 15 to 37 percent of links chosen by widely used algorithms are classified as errors by trained human reviewers; and (iii) false links are systematically related to baseline sample characteristics, showing that some algorithms may introduce systematic measurement error into analyses. A case study shows that the combined effects of (i)–(iii) attenuate estimates of the intergenerational income elasticity by up to 29 percent, and common variations in algorithm assumptions result in greater attenuation. As current practice moves to automate linking and increase link rates, these results highlight the important potential consequences of linking errors on inferences with linked data. We conclude with constructive suggestions for reducing linking errors and directions for future research. (JEL C45, C81, J62, N31, N32)
Algorithm · Baseline (sea) · Constructive · Data mining · Data quality · Data science · Econometrics · Economics · Ground truth · Linkage (software) · Machine learning · Operations management · Process (computing) · Quality (philosophy) · Sample (material) · Artificial Intelligence · Computer Science · Data Quality and Management · Health disparities and outcomes · Healthcare Policy and Management · Mathematics
The mortality gradient in town and country
"How Ya Gonna Keep 'Em Down on the Farm, After They've Seen Paree?" World War I Overseas Military Service and Rural Americans' Postwar Occupational Mobility
The Effects of the Great Depression on Children’s Intergenerational Mobility
Simple strategies for improving inference with linked data
Did military service during World War I affect the economic status of American veterans
Changes in the demographics of American inventors, 1870–1940
Population‐Scale Kinship Networks
Automated Linking of Historical Data
Intergenerational mobility across nine generations
Female labor force participation in historical census microdata
Environmental migration during the Great American Drought
Household-Anchored Record Linkage in Ottoman Population Registers. A Dynamic Tolerance Protocol for Low-Information Administrative Proto-Census Data
What about immigration? The closed-population assumption in research on intergenerational income mobility
Mobility and marriage
The Economics of Gender-Specific Minimum Wage Legislation
A Different Land of Opportunity
History and urban economics
A Comparison of Rule-based and Supervised Machine Learning Approaches for Record Linkage of Italian Historical Data
Durkheim and the roots of cliometric reasoning
Black residents of Windsor, Ontario during the U.S. Civil War
Were Small-town New Yorkers Life-cycle Savers
Black boxes in nominal record linkage with historical sources
A review of data linkages for policy-informing research in food and agricultural economics
The impact of kin proximity on net marital fertility and maternal survival in Sweden 1900–1910—Evidence for cooperative breeding in a societal context of nuclear families, or just contextual correlations
Links and Legibility
Human Capital and the Managerial Revolution in the United States
Finding John Smith
Mobility for All
Reevaluating the Long-Term Impact of In Utero Exposure to the 1918 Influenza Pandemic
Reconstructing History
Happily Ever After
Gender and Racial Differences in Migration Behavior
The problem of false positives in automated census linking
US migration from 1850 to 1920
A new strategy for linking U.S. historical censuses
Children and grandchildren of Union Army veterans
The Ipums multigenerational longitudinal panel
The creation of Life-M
Immigration, Poverty, and Infant and Child Mortality in the City of Madrid, 1916–1926
Business Entry and Exit
Selection Bias Encountered in the Systematic Linking of Historical Census Records
Vocational Rehabilitation and Labor Market Outcomes
The Longevity Benefits of Homeownership
Contextual Boundaries
US immigrants’ secondary migration and geographic assimilation during the Age of Mass Migration
Hana
Life after crossing the border
Perks and pitfalls of city directories as a micro-geographic data source
Early-life lead exposure and male longevity
The geography of opportunity after the Civil War
Examining the role of training data for supervised methods of automated record linkage
Linked samples and measurement error in historical US census data
Breakthroughs in historical record linking using genealogy data
Institutional discrimination and assimilation
The mortality risk of being overweight in the twentieth century
Name-Based Estimators of Intergenerational Mobility
Her Property Transactions
Social Mobility in Sweden before the Welfare State
Black Americans' Landholdings and Economic Mobility after Emancipation
La "Doña" è Mobile
Like Father Like Son? Intergenerational Immobility in England, 1851-1911
Tratamento de dados do Sistema Nacional de Gerenciamento de Produtos Controlados para estudos de utilização de medicamentos com antimicrobianos
Social Mobility in the Long Run
New Data Sources for Demographic Research
Migrations canadiennes-françaises au 19e siècle
The Long-Term Development of the Mortality Gradient
A Theory for Record Linkage
Characterizing Selection Bias Using Experimental Data
Labor Market Institutions and the Distribution of Wages, 1973-1992
The Scandinavian Fantasy
Moving to Higher Ground
Recent Developments in Intergenerational Mobility
The Long-Run Impact of Cash Transfers to Poor Families
Maximum Likelihood from Incomplete Data Via the EM Algorithm
A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
Social science history
Intergenerational Mobility in the Labor Market
Is the United States Still a Land of Opportunity? Recent Trends in Intergenerational Mobility
Europe's Tired, Poor, Huddled Masses
Life-Cycle Variation in the Association between Current and Lifetime Earnings
When the Levee Breaks
The Sensitivity of Economic Statistics to Coding Errors in Personal Identifiers
Season of Birth and Later Outcomes
Fortunate Sons
Estimating the Impacts of Program Benefits
Have the poor always been less likely to migrate? Evidence from inheritance practices during the age of mass migration
Matchmaker, Matchmaker, Make Me a Match”
Union Army veterans, all grown up
Simple strategies for improving inference with linked data
Notes from the National Panel Study
Decennial Life Tables for the White Population of the United States, 1790–1900
Playing with matches
Census Matching and Migration
Testing Methods of Record Linkage on Swedish Censuses
New Methods of Census Record Linking
A “Mulatto Escape Hatch” in the United States? Examining Evidence of Racial and Social Mobility During the Jim Crow Era
Patterns of occupational mobility among Negro men
A Nation of Immigrants
Multiple Measures of Historical Intergenerational Mobility
A Nineteenth-Century Resource for Agricultural History Research in the Twenty-First Century
Ideology and Migration after the American Civil War
Obama, Katrina, and the Persistence of Racial Inequality
The Great Migration in Black and White
Intergenerational Mobility in the United States
Quantifying Quantitative Literacy
Targeting Lynch Victims
Historical Census Record Linkage
Intergenerational Coresidence and Family Transitions in the United States, 1850-1880
| Unique citing works | 66 |
|---|---|
| Citations per year | 9,43 |
| Citation span | 2019 - 2026 (8) |
| Citation velocity | current |
| Highly cited | No |
| Citation types | Neutral: 60 |