Research project: Disclosure risk assessment for survey microdata
Statistical agencies generally have controlled release of social survey microdata to meet the demands of researchers and policy makers.
Statistical agencies generally have controlled release of social survey microdata to meet the demands of researchers and policy makers.
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