EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

The 2020 US Decennial Census is more private than you (might) think

Buxin Su, Weijie J Su iD, Chendi Wang

Proceedings of the National Academy of Sciences · 11.11.2025 · 122 (45)

DOI: 10.1073/pnas.2500337122 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 05.11.2025. Verfasst von Buxin Su und 2 weiteren Autor:innen. Beteiligte Einrichtung: Applied Mathematics and Computational Science, University of Pennsylvania, Philadelphia, PA 19104. Hauptthemen laut MEDLINE: Censuses, Privacy, Confidentiality. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The US Decennial Census serves as the foundation for many high-profile policy decision-making processes, including federal funding allocation and redistricting. In 2020, the Census Bureau adopted differential privacy to protect the confidentiality of individual responses through a disclosure avoidance system that injects noise into census data tabulations. The Bureau subsequently posed an open question: Could stronger privacy guarantees be obtained for the 2020 US Census compared to their published guarantees, or equivalently, had the privacy budgets been fully utilized? In this paper, we address this question affirmatively by demonstrating that the 2020 US Census provides significantly stronger privacy protections than the officially published guarantees suggest at each of the eight geographical levels, from the national level down to the block level. This finding is enabled by our precise tracking of privacy losses using f -differential privacy, applied to the composition of private queries across these geographical levels. Our analysis reveals that the Census Bureau introduced unnecessarily high levels of noise to meet the specified privacy guarantees for the 2020 Census. Consequently, we show that noise variances could be reduced by 15.08% to 24.82% while maintaining nearly the same level of privacy protection for each geographical level, thereby improving the accuracy of privatized census statistics. We empirically demonstrate that reducing noise injection into census statistics mitigates distortion caused by privacy constraints in downstream applications of private census data, illustrated through a study examining the relationship between earnings and education.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CensusesPrivacyConfidentialityUnited StatesHumans

Stichwörter der Autor:innen

US Censusdata utilitydifferential privacyprivacy accounting

Autor:innen und Institutionen

  • Applied Mathematics and Computational Science, University of Pennsylvania, Philadelphia, PA 19104.
  • Wharton Statistics and Data Science Department, University of Pennsylvania, Philadelphia, PA 19104.
  • Department of Statistics and Data Science, School of Economics, Xiamen University, Xiamen, Fujian 361000, China.
  • Ministry of Education lab of Key Laboratory of Econometrics and Wang Yanan Institute for Studies in Economics, Xiamen University, Xiamen, Fujian 361000, China.

ORCID-Profile: Weijie J Su

Zitieren

APA

Su, B., Su, W. J., & Wang, C. (2025). The 2020 US Decennial Census is more private than you (might) think. Proceedings of the National Academy of Sciences, 122(45). https://doi.org/10.1073/pnas.2500337122

Vancouver

Su B, Su WJ, Wang C. The 2020 US Decennial Census is more private than you (might) think. Proceedings of the National Academy of Sciences. 2025;122(45). doi:10.1073/pnas.2500337122

BibTeX
@article{Su2025,
  author = {Su, Buxin and Su, Weijie J and Wang, Chendi},
  title = {The 2020 US Decennial Census is more private than you (might) think},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2025},
  volume = {122},
  number = {45},
  doi = {10.1073/pnas.2500337122}
}
Kostenlos starten →