Differential Privacy and Redistricting
The 2020 Decennial Census will be released with a new disclosure avoidance system in place, putting differential privacy in the spotlight for a wide range of data users. We consider several key applications of Census data in redistricting, developing tools and demonstrations for practitioners who are concerned about the impacts of this new noising algorithm called TopDown. Based on a close look at reconstructed Texas data, we find reassuring evidence that TopDown will not threaten the ability to produce districts with tolerable population balance or to detect signals of racial polarization for Voting Rights Act enforcement.
Our first publication on this topic has now appeared in Foundations of Responsible Computing, and can be accessed here.
A follow-up study that considered demographic impacts on small, diverse districts in Arizona was commissioned by the Arizona Independent Redistricting Commission. The presentation can be accessed here.
A further study, updated to consider the final production parameters used in the 2020 Decennial Census, is in revision at the Harvard Data Science Review.