Balancing Utility Versus Privacy in the 2020 Census: Sentiments from Data Users

Abstract

When designing disclosure avoidance systems, federal statistical agencies must understand how data users balance data utility and privacy in their assessments of privacy protection frameworks, such as differential privacy. Through one-on-one semi-structured interviews with 33 participants across government, research institutions, and community organizations, we collected data users’ opinions on the new privacy safeguards implemented by the U.S. Census Bureau for the 2020 Census. We identified several messaging themes, which were organized into seven groups. We utilized this feedback to develop a series of recommendations, including creating a separate data file without post-processing, developing disclosure avoidance system criteria transparently, and opening more avenues for accessing unprotected microdata.

Keywords: differential privacy, census, US Census Bureau, data privacy