Research project: Disclosure risk assessment for survey microdata

Since social surveys are often based on area samples or address lists, the characteristics of the population are unknown and need to be estimated for the purpose of disclosure risk assessment. Chris Skinner, Natalie Shlomo and Robin Mitra have worked in the development of probabilistic statistical models for estimating a disclosure risk measure for social survey microdata. Chris Skinner and Natalie Shlomo have expanded these models to take into account misclassification. Jon Forster has used Bayesian Methods for assessing the disclosure risk in survey microdata.

Outputs from this research include

• Rinott, Y. and Shlomo, N. (2007) Variances and confidence intervals for sample disclosure risk measures. Invited paper, 2007 ISI Conference, Lisbon.
• Skinner, C. J. and Shlomo, N. (2008) Assessing identification risk in survey micro-data using log linear models. Journal of American Statistical Association, 483, 989-1001.
• Skinner, C.J. (2008). Assessing disclosure risk for record linkage in Privacy in Statistical Databases, (Domingo-Ferrer, J. And Saygin, Y., eds.), Lecture notes in Computer Science 5262, Berlin: Springer, 166-176.
• Shlomo, N. and Skinner, C.J. (2009) Assessing the protection provided by misclassification for survey microdata. s3ri-workingpaper-M09-14.pdf
• Shlomo, N. (2009) Releasing microdata: misclosure risk estimation, data masking and assessing utility. s3ri-workingpaper-M09-02.pdf