Non-Random Assignment of Individual Identifiers and Selection into Linked Data: Implications for Research
January 2026
Working Paper Number:
CES-26-06
Abstract
The U.S. Census Bureau's Person Identification Validation System facilitates anonymous linkages between survey and administrative records by assigning Protected Identification Keys (PIKs) to person records. While PIK assignment is generally accurate, some person records are not successfully assigned a PIK, which can lead to sample selection bias in analyses of linked data. Using the American Community Survey (ACS) and the Current Population Survey Annual Social and Economic Supplement (CPS ASEC) between 2005 and 2022, we corroborate and extend existing findings on the drivers of PIK assignment, showing that the rate of PIK assignment varies widely across socio-demographic subgroups. Using earnings as a test case, we then show that limiting a survey sample of wage earners to person records with PIKs or successful linkages to W-2 wage records tends to overestimate self-reported wage earnings, on average, indicative of linkage-induced selection bias. In a validation exercise, we demonstrate that reweighting methods, such as inverse probability weighting or entropy balancing, can mitigate this bias.
Document Tags and Keywords
Keywords
Keywords are automatically generated using KeyBERT, a powerful and innovative
keyword extraction tool that utilizes BERT embeddings to ensure high-quality and contextually relevant
keywords.
By analyzing the content of working papers, KeyBERT identifies terms and phrases that capture the essence of the
text, highlighting the most significant topics and trends. This approach not only enhances searchability but
provides connections that go beyond potentially domain-specific author-defined keywords.
:
estimating,
survey,
respondent,
earnings,
bias,
matching,
race,
socioeconomic,
census bureau,
irs,
sampling,
household surveys,
earner,
linkage
Tags
Tags are automatically generated using a pretrained language model from spaCy, which excels at
several tasks, including entity tagging.
The model is able to label words and phrases by part-of-speech,
including "organizations." By filtering for frequent words and phrases labeled as "organizations", papers are
identified to contain references to specific institutions, datasets, and other organizations.
:
Internal Revenue Service,
Bureau of Labor Statistics,
Social Security Administration,
Current Population Survey,
American Community Survey,
Social Security Number,
Protected Identification Key,
W-2,
National Academy of Sciences,
Social and Economic Supplement,
Census Bureau Disclosure Review Board,
PIKed,
Person Validation System,
Person Identification Validation System,
Individual Taxpayer Identification Numbers,
SSA Numident,
COVID-19,
CPS ASEC
Similar Working Papers
Similarity between working papers are determined by an unsupervised neural
network model
know as Doc2Vec.
Doc2Vec is a model that represents entire documents as fixed-length vectors, allowing for the
capture of semantic meaning in a way that relates to the context of words within the document. The model learns to
associate a unique vector with each document while simultaneously learning word vectors, enabling tasks such as
document classification, clustering, and similarity detection by preserving the order and structure of words. The
document vectors are compared using cosine similarity/distance to determine the most similar working papers.
Papers identified with 🔥 are in the top 20% of similarity.
The 10 most similar working papers to the working paper 'Non-Random Assignment of Individual Identifiers and Selection into Linked Data: Implications for Research' are listed below in order of similarity.
-
Working PaperThe Nature of the Bias When Studying Only Linkable Person Records: Evidence from the American Community Survey
April 2014
Working Paper Number:
carra-2014-08
Record linkage across survey and administrative records sources can greatly enrich data and improve their quality. The linkage can reduce respondent burden and nonresponse follow-up costs. This is particularly important in an era of declining survey response rates and tight budgets. Record linkage also creates statistical bias, however. The U.S. Census Bureau links person records through its Person Identification Validation System (PVS), assigning each record a Protected Identification Key (PIK). It is not possible to reliably assign a PIK to every record, either due to insufficient identifying information or because the information does not uniquely match any of the administrative records used in the person validation process. Non-random ability to assign a PIK can potentially inject bias into statistics using linked data. This paper studies the nature of this bias using the 2009 and 2010 American Community Survey (ACS). The ACS is well-suited for this analysis, as it contains a rich set of person characteristics that can describe the bias. We estimate probit models for whether a record is assigned a PIK. The results suggest that young children, minorities, residents of group quarters, immigrants, recent movers, low-income individuals, and non-employed individuals are less likely to receive a PIK using 2009 ACS. Changes to the PVS process in 2010 significantly addressed the young children deficit, attenuated the other biases, and increased the validated records share from 88.1 to 92.6 percent (person-weighted).View Full Paper PDF
-
Working PaperDifferences in Disability Insurance Allowance Rates
August 2025
Working Paper Number:
CES-25-54
Allowance rates for disability insurance applications vary by race and ethnicity, but it is unclear to what extent these differences are artifacts of other differing socio-economic and health characteristics, or selection issues in SSA's race and ethnicity data. This paper uses the 2015 American Community Survey linked to 2015-2019 SSA administrative data to investigate DI application allowance rates among non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, non-Hispanic American Indian/Alaska Native, and Hispanic applicants aged 25-65. The analysis uses regression, propensity score matching, and inverse probability weighting to estimate differences in allowance rates among applicants who are similar on observable characteristics. Relative to raw comparisons, differences by race and ethnicity in multivariate analyses are substantially smaller in magnitude and are generally not statistically significant.View Full Paper PDF
-
Working PaperPotential Bias When Using Administrative Data to Measure the Family Income of School-Aged Children
January 2025
Working Paper Number:
CES-25-03
Researchers and practitioners increasingly rely on administrative data sources to measure family income. However, administrative data sources are often incomplete in their coverage of the population, giving rise to potential bias in family income measures, particularly if coverage deficiencies are not well understood. We focus on the school-aged child population, due to its particular import to research and policy, and because of the unique challenges of linking children to family income information. We find that two of the most significant administrative sources of family income information that permit linking of children and parents'IRS Form 1040 and SNAP participation records'usefully complement each other, potentially reducing coverage bias when used together. In a case study considering how best to measure economic disadvantage rates in the public school student population, we demonstrate the sensitivity of family income statistics to assumptions about individuals who do not appear in administrative data sources.View Full Paper PDF
-
Working PaperThe Design of Sampling Strata for the National Household Food Acquisition and Purchase Survey
February 2025
Working Paper Number:
CES-25-13
The National Household Food Acquisition and Purchase Survey (FoodAPS), sponsored by the United States Department of Agriculture's (USDA) Economic Research Service (ERS) and Food and Nutrition Service (FNS), examines the food purchasing behavior of various subgroups of the U.S. population. These subgroups include participants in the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), as well as households who are eligible for but don't participate in these programs. Participants in these social protection programs constitute small proportions of the U.S. population; obtaining an adequate number of such participants in a survey would be challenging absent stratified sampling to target SNAP and WIC participating households. This document describes how the U.S. Census Bureau (which is planning to conduct future versions of the FoodAPS survey on behalf of USDA) created sampling strata to flag the FoodAPS targeted subpopulations using machine learning applications in linked survey and administrative data. We describe the data, modeling techniques, and how well the sampling flags target low-income households and households receiving WIC and SNAP benefits. We additionally situate these efforts in the nascent literature on the use of big data and machine learning for the improvement of survey efficiency.View Full Paper PDF
-
Working PaperNational Experimental Wellbeing Statistics - Version 1
February 2023
Working Paper Number:
CES-23-04
This is the U.S. Census Bureau's first release of the National Experimental Wellbeing Statistics (NEWS) project. The NEWS project aims to produce the best possible estimates of income and poverty given all available survey and administrative data. We link survey, decennial census, administrative, and third-party data to address measurement error in income and poverty statistics. We estimate improved (pre-tax money) income and poverty statistics for 2018 by addressing several possible sources of bias documented in prior research. We address biases from 1) unit nonresponse through improved weights, 2) missing income information in both survey and administrative data through improved imputation, and 3) misreporting by combining or replacing survey responses with administrative information. Reducing survey error substantially affects key measures of well-being: We estimate median household income is 6.3 percent higher than in survey estimates, and poverty is 1.1 percentage points lower. These changes are driven by subpopulations for which survey error is particularly relevant. For house holders aged 65 and over, median household income is 27.3 percent higher and poverty is 3.3 percentage points lower than in survey estimates. We do not find a significant impact on median household income for householders under 65 or on child poverty. Finally, we discuss plans for future releases: addressing other potential sources of bias, releasing additional years of statistics, extending the income concepts measured, and including smaller geographies such as state and county.View Full Paper PDF
-
Working PaperThe Census Historical Environmental Impacts Frame
October 2024
Working Paper Number:
CES-24-66
The Census Bureau's Environmental Impacts Frame (EIF) is a microdata infrastructure that combines individual-level information on residence, demographics, and economic characteristics with environmental amenities and hazards from 1999 through the present day. To better understand the long-run consequences and intergenerational effects of exposure to a changing environment, we expand the EIF by extending it backward to 1940. The Historical Environmental Impacts Frame (HEIF) combines the Census Bureau's historical administrative data, publicly available 1940 address information from the 1940 Decennial Census, and historical environmental data. This paper discusses the creation of the HEIF as well as the unique challenges that arise with using the Census Bureau's historical administrative data.View Full Paper PDF
-
Working PaperThe Measurement of Medicaid Coverage in the SIPP: Evidence from California, 1990-1996
September 2002
Working Paper Number:
CES-02-21
This paper studies the accuracy of reported Medicaid coverage in the Survey of Income and Program Participation (SIPP) using a unique data set formed by matching SIPP survey responses to administrative records from the State of California. Overall, we estimate that the SIPP underestimates Medicaid coverage in the California populaton by about 10 percent. Among SIPP respondents who can be matched to administrative records, we estimate that the probability someone reports Medicaid coverage in a month when they are actually covered is around 85 percent. The corresponding probability for low-income children is even higher ' at least 90 percent. These estimates suggest that the SIPP provides reasonably accurate coverage reports for those who are actually in the Medicaid system. On the other hand, our estimate of the false positive rate (the rate of reported coverage for those who are not covered in the administrative records) is relatively high: 2.5 percent for the sample as a whole, and up to 20 percent for poor children. Some of this is due to errors in the recording of Social Security numbers in the administrative system, rather than to problems in the SIPP.View Full Paper PDF
-
Working PaperForeign-Born and Native-Born Migration in the U.S.: Evidence from IRS Administrative and Census Survey Records
July 2018
Working Paper Number:
carra-2018-07
This paper details efforts to link administrative records from the Internal Revenue Service (IRS) to American Community Survey (ACS) and 2010 Census microdata for the study of migration among foreign-born and native-born populations in the United States. Specifically, we (1) document our linkage strategy and methodology for inferring migration in IRS records; (2) model selection into and survival across IRS records to determine suitability for research applications; and (3) gauge the efficacy of the IRS records by demonstrating how they can be used to validate and potentially improve migration responses for native-born and foreign-born respondents in ACS microdata. Our results show little evidence of selection or survival bias in the IRS records, suggesting broad generalizability to the nation as a whole. Moreover, we find that the combined IRS 1040, 1099, and W2 records may provide important information on populations, such as the foreign-born, that may be difficult to reach with traditional Census Bureau surveys. Finally, while preliminary, the results of our comparison of IRS and ACS migration responses shows that IRS records may be useful in improving ACS migration measurement for respondents whose migration response is proxy, allocated, or imputed. Taking these results together, we discuss the potential application of our longitudinal IRS dataset to innovations in migration research on both the native-born and foreign-born populations of the United States.View Full Paper PDF
-
Working PaperUnderstanding the Quality of Alternative Citizenship Data Sources for the 2020 Census
August 2018
Working Paper Number:
CES-18-38R
This paper examines the quality of citizenship data in self-reported survey responses compared to administrative records and evaluates options for constructing an accurate count of resident U.S. citizens. Person-level discrepancies between survey-collected citizenship data and administrative records are more pervasive than previously reported in studies comparing survey and administrative data aggregates. Our results imply that survey-sourced citizenship data produce significantly lower estimates of the noncitizen share of the population than would be produced from currently available administrative records; both the survey-sourced and administrative data have shortcomings that could contribute to this difference. Our evidence is consistent with noncitizen respondents misreporting their own citizenship status and failing to report that of other household members. At the same time, currently available administrative records may miss some naturalizations and capture others with a delay. The evidence in this paper also suggests that adding a citizenship question to the 2020 Census would lead to lower self-response rates in households potentially containing noncitizens, resulting in higher fieldwork costs and a lower-quality population count.View Full Paper PDF
-
Working PaperEarnings Measurement Error, Nonresponse and Administrative Mismatch in the CPS
July 2025
Working Paper Number:
CES-25-48
Using the Current Population Survey Annual Social and Economic Supplement matched to Social Security Administration Detailed Earnings Records, we link observations across consecutive years to investigate a relationship between item nonresponse and measurement error in the earnings questions. Linking individuals across consecutive years allows us to observe switching from response to nonresponse and vice versa. We estimate OLS, IV, and finite mixture models that allow for various assumptions separately for men and women. We find that those who respond in both years of the survey exhibit less measurement error than those who respond in one year. Our findings suggest a trade-off between survey response and data quality that should be considered by survey designers, data collectors, and data users.View Full Paper PDF
