Privacy at What Cost? The Differential Effects of Disclosure Avoidance Methods on Data About Small Groups

• Activity Number: 496
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Government Statistics Section
• Author(s): Michael Brewster Hawes*+ and Mark Asiala*+ and Jacob Bournazian*+ and Chris Chapman*+
• Companies: U.S. Department of Education and U.S. Census Bureau and U.S. Energy Information Administration and Bureau of Labor Statistics
• Keywords: Privacy ; Disclosure ; SDL ; American Community Survey ; Education ; Health

Abstract

Public release of survey, administrative, or program data typically requires the application of one or more statistical disclosure limitation (SDL) techniques to protect the privacy of the individuals (or other entities) from whom the information was collected. These SDL methods protect privacy by typically suppressing, altering, and/or blurring the data to reduce the ability of a third party to re-identify specific individuals with their reported data. Data about small groups (those with uncommon or unique characteristics, or those within a small geographic area) pose the greatest risk for re-identification, and thus are typically the most affected by SDL methods. This panel will examine the effects of traditional SDL methodologies on the utility of data concerning small groups in various sectors, and will discuss possible alternative methodologies that could mitigate some of these effects while still protecting privacy.