Afternoon Plenary Lecture
Protecting Data Confidentiality in an Era Without Privacy
Mrs. Alexander Hehmeyer Professor of Statistical Science,
Department of Statistical Science,
Abstract: Many statistical agencies, survey organizations, research groups, and private companies seek to share data with others. Wide dissemination of data facilitates scientific advances, improves policy-making, allows students to train on data analysis, and helps citizens understand their societies. Typically, however, data releasers are ethically or even legally obligated to protect the confidentiality of data subjects' identities and sensitive attributes. Stripping direct identifiers like names and addresses may not suffice. When the data include variables that are readily avalaible on external files, such as demographic charateristics, ill-intentioned users may be able to match the released data to records on external files. In this talk, I describe how statistical methods can be used to assess and reduce the risks of such disclosures. I describe the methods in general terms, focusing on intuition rather than mathematical formality. I also describe some applications of confidentiality protection in large-scale, government statistical databases.
The talk could be downloaded here.
Biosketch: Jerry Reiter is the Mrs. Alexander Hehmeyer Professor of Statistical Science at Duke University. He received his PhD in statistics from Harvard University in 1999. His main research areas include methods for protecting data confidentiality, for handling missing values, and for analyzing complex surveys in the social sciences. He works extensively with the US Bureau of the Census and other federal statistical agencies. He received the Alumni Distinguished Undergraduate Teaching Award at Duke in 2007.