Protecting Data Confidentiality in Publicly Released Datasets: Approaches Based on Multiple Imputation

Statistical organizations that release data to the public typically are required to protect the confidentiality of survey respondents' identities and attribute values. Removing direct identifiers such as names and addresses generally is not sufficient to eliminate disclosure risks, so that statistical disclosure limitation strategies must be applied to the data before release. This chapter presents an overview of how multiple imputation, originally devised to handle missing data, can be adapted for disclosure limitation. It reviews the literature on inferential methods for analyzing such datasets. It concludes with discussion of implementation challenges and topics for future research. © 2012 Elsevier B.V.

Duke Scholars

Author Jerome P. Reiter Statistical Science