Assessing the risk of disclosure of confidential categorical data

Disclosure limitation involves the application of statistical tools to limit the identification of information on individuals (and enterprises) included as part of statistical databases such as censuses and sample surveys. We outline the major issues involved in assessing disclosure risk and assuring the protection of confidentiality for databases, especially those in the form of multi-way contingency tables, and we present a Bayesian framework for thinking about such problems both from the perspective of an intruder and the agency trying to protect its data.