5 Statistical Disclosure Limitation

Within most national statistical systems, the primary approach to protecting respondent privacy has been statistical disclosure limitation or SDL. Anderson & Seltzer (2007) describes the history of threats to confidentiality in the U.S. statistical system prior to 1965. Fellegi (1972) initiated the statistical analysis of data confidentiality. Dalenius (1977) recognized that statistical agencies would need to do more than just protect against direct disclosures, and thus warned against what he called inferential disclosure. His idea was formalized by Duncan & Lambert (1986), and provides the ultimate rationale for formal privacy in national statistics.

Abowd & Schmutte (2015) review the SDL methods currently in use and discuss their application to economic data. They argue that the analysis of SDL-laden data is inherently compromised because the details of the SDL protections cannot be disclosed. If they cannot be disclosed, their consequences for inference are unknowable, and, as they show, potentially large. Garfinkel (2015) discusses techniques for de-identifying data and the many ways in which modern computing tools and a data-rich environment may render effective de-identification impossible. Finally, Harris-Kojetin et al. (2005) provides the most comprehensive review of SDL methods currently in use across the U.S. statistical system.

References

Abowd, J. M., & Schmutte, I. M. (2015). Economic analysis and statistical disclosure limitation. Brookings Papers on Economic Activity, 221–267. https://doi.org/10.1353/eca.2016.0004

Anderson, M., & Seltzer, W. (2007). Challenges to the confidentiality of US federal statistics, 1910-1965. Journal of Official Statistics, 23(1), 1. Retrieved from https://www.scb.se/contentassets/ff271eeeca694f47ae99b942de61df83/challenges-to-the-confidentiality-of-u.s.-federal-statistics-1910-1965.pdf

Dalenius, T. (1977). Towards a methodology for statistical disclosure control. Statistik Tidskrift, 15, 429–444. https://doi.org/10.1145/320613.320616

Duncan, G., & Lambert, D. (1986). Disclosure-limited data dissemination. Journal of the American Statistical Association, 81(393), 10–18. https://doi.org/10.1080/01621459.1986.10478229

Fellegi, I. P. (1972). On the question of statistical confidentiality. Journal of the American Statistical Association, 67(337), 7–18. https://doi.org/10.2307/2284695

Garfinkel, S. (2015). De-Identification of personal information (Internal Report No. 8053). https://doi.org/10.6028/nist.ir.8053

Harris-Kojetin, B. A., Alvey, W. L., Carlson, L., Cohen, S. B., Cohen, S. H., Cox, L. H., … Groves, R. (2005). Statistical policy working paper 22: Report on statistical disclosure limitation methodology [Research Report]. Retrieved from U.S. Federal Committee on Statistical Methodology website: https://nces.ed.gov/FCSM/pdf/spwp22.pdf