Microdata Protection through Noise Addition

  • Ruth Brand
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2316)


Microdata protection by adding noise is being discussed for more than 20 years now. Several algorithms were developed that have different characteristics. The simplest algorithm consists of adding white noise to the data. More sophisticated methods use more or less complex transformations of the data and more complex error-matrices to improve the results. This contribution gives an overview over the different algorithms and discusses their properties in terms of analytical validity and level of protection. Therefore some theoretical considerations are shown and an illustrating empirical example is given.


Statistical disclosure control microdata protection noise addition 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Brand, R. (2000): Anonymität von Betriebsdaten, Beiträge zur Arbeitsmarktund Berufsforschung Beitrag 237, Institut für Arbeitsmarkt-und Berufsforschung, Nürnberg.Google Scholar
  2. 2.
    Brand, R., Bender, S. and S. Kohaut (1999): Possibilities for the Creation of a Scientific Use File for the IAB-Establishment-Panel, Statistical Data Confidentiality, Proceedings of the Joint Eurostat/UN-ECE Work session on Statistical Data Confidentiality in March 1999.Google Scholar
  3. 3.
    Chauduri, A. and R. Mukerjee (1988): Randomized Response: Theory and Techniques, New York, Base: Marcel Dekker.Google Scholar
  4. 4.
    de Waal, A.G. and L.C.R.J. Willenborg (1997): Statistical Disclosure Control and Sampling Weights, Journal of Official Statistics 13, 417–434.Google Scholar
  5. 5.
    Domingo-Ferrer, J. and J.M. Mateo-Sanz (2001): Comparing SDC Methods for Microdata on the Basis of Information Loss and Disclosure Risk, paper presented at the Joint ECE/Eurostat Worksession on Statistical Confidentiality in Skopie (The former Yugoslav Republic of Macedonia), 14–16 March 2001Google Scholar
  6. 6.
    Fienberg, S.E. (1997): Confidentiality and Disclosure Limitation Methodology: Challenges for National Statistics and Sstatistical Research, Technical Report No. 611, Carnegie Mellon university, Pittsburgh.Google Scholar
  7. 7.
    Fuller, W.A. (1993): Masking Procedures for Microdata Disclosure Limitation, Journal of Official Statistics 9, 383–406.Google Scholar
  8. 8.
    Gouweleeuw, J.M., Kooiman, P., Willenborg, L.C.R.J. and P.-P. de Wolf (1998): Post Randomisation for Statistical Disclosure Control: Theory and Implementation, Journal of Official Statistics 14, 463–478.Google Scholar
  9. 9.
    Harho., D. and G. Licht (1993): Anonymisation of Innovation Survey Data, the Disguise Factor Method, Zentrum für Europäische Wirtschaftsforschung, Mannheim.Google Scholar
  10. 10.
    Kim, J.J. (1986): A Method for Limiting Disclosure in Microdata based on Random Noise and Transformation, Proceedings of the Section on Survey Research Methods 1986, American Statistical Association, 303–308.Google Scholar
  11. 11.
    Kim, J.J. (1990): Subpopulation Estimation for the Masked Data, Proceedings of the Section on Survey Research Methods 1990, American Statistical Association, 456–461.Google Scholar
  12. 12.
    Kim, J.J. and W.E. Winkler (1995): Masking Microdata Files, Proceedings of the Section on Survey Research Methods 1995, American Statistical Association, 114–119.Google Scholar
  13. 13.
    Kim, J.J. and W.E. Winkler (1997): Masking Microdata Files, Statistical Research Division RR97/03, US Bureau of the Census, Washington, DC.Google Scholar
  14. 14.
    Kim, J.J. and W.E. Winkler (2001): Multiplicative Noise for Masking Continouus Data, unpublished manuscript.Google Scholar
  15. 15.
    McGuckin, R. H. and S.V. Nguyen (1990), Public Use Microdata: Disclosure and Usefulness, Journal of Economic and Social Development 16, 19–39.Google Scholar
  16. 16.
    McGuckin, R. H. (1993): Analytic Use of Economic Microdata: A Model for Researcher Access with Confidentiality Protection, Proceedings of the International Seminar on Statistical Confidentiality, 08–10. Sept. 1992, Dublin, Ireland; Eurostat, 83–97.Google Scholar
  17. 17.
    Moore, R.A. (1996): Analysis of the Kim-Winkler Algorithm for Masking Microdata files—How Much Masking is Necessary and Sufficient? Conjectures for the Development of a Controllable Algorithm, Statistical Research Division RR96/05, US Bureau of the Census, Washington D.C.Google Scholar
  18. 18.
    Paass, G. (1988): Disclosure Risk and Disclosure Avoidance for Microdata, Journal of Business & Economic Statistics 6, 487–500.CrossRefGoogle Scholar
  19. 19.
    Roque, G. M. (2000): Masking Microdata Files with Mixtures of Multivariate Normal Distributions, unpublished PhD-Thesis, University of California, Riverside.Google Scholar
  20. 20.
    Spruill, N.L (1983) The Confidentiality and Analytic Usefulness of Masked Business Microdata, Proceedings of the Section on Survey Research Methods 1983, American Statistical Association, 602–610.Google Scholar
  21. 21.
    Sullivan, G.R. (1989): The Use of Added Error to Avoid Disclosure in Microdata Releases, unpublished PhD-Thesis, Iowa State University.Google Scholar
  22. 22.
    Sullivan, G.R. and W.A. Fuller (1989): The Use of Measurement Error to Avoid Disclosure, Proceedings of the Section on Survey Research Methods 1989, American Statistical Association, 802–807.Google Scholar
  23. 23.
    Sullivan, G.R. and W.A. Fuller (1990): Construction of Masking Error for Categorial Variables, Proceedings of the Section on Survey Research Methods 1990, American Statistical Association, 453–439.Google Scholar
  24. 24.
    Tendick, P. (1988): Bias Avoidance and Measures of Confidentiality for the Noise Addition Method of Database Disclosure Control, unpublished PhD-Thesis, University of California, Davis.Google Scholar
  25. 25.
    Tendick, P. (1991): Optimal Noise Addition for Preserving Confidentiality in Multivariate Data, Journal of Statistical Planning and Inference 27, 341–353.zbMATHCrossRefMathSciNetGoogle Scholar
  26. 26.
    Tendick, P. and N. Matlo. (1994): A Modified Random Perturbation Method for Databse Security, ACM Transactions on Database Systems 19, 47–63.CrossRefGoogle Scholar
  27. 27.
    Willenborg, L. and T. de Waal (1996): Statistical Disclosure Control in Practice, New York, Berlin, Heidelberg: Springer.zbMATHGoogle Scholar
  28. 28.
    Winkler, W.E. (1995): Matching and Record Linkage, Cox, B.G., Binder, D.A., Chinnappa, B.N., Christianson, A., Colledge, M.J. und P.S. Kott (ed.): Business Survey Methods, Wiley Series in Probability and Mathematical Statistics, New York: John Wiley & Sons, 335–384.Google Scholar
  29. 29.
    Winkler, W.E. (1998): Re-identification Methods for Evaluating the Confidentiality of Analytically Valid Microdata, Statistical Data Protection 1998, Lisbon, Portugal.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Ruth Brand
    • 1
  1. 1.Federal Statistical Office of GermanyWiesbaden

Personalised recommendations