The COVID-19 pandemic has taken a heavy toll on the world: lives lost, livelihoods lost, and solutions to be found. But other issues have not disappeared, such as persistent social and racial inequalities. Finding solutions in an evidence-based world requires usable and precise data. Questions as to the right balance between usability and privacy do not have easy, or easily accessible answers.
The ASA’s Committee on Privacy and Confidentiality’s mission is to keep track of evolving technical, policy, and legal developments, and to communicate with the statistical community on these issues.
For the 2021 Joint Statistical Meetings - coming up in August - the session on “Collecting, Protecting and Sharing COVID-19 Data and the Stories They Tell,” organized by Committee chair Lars Vilhuber, considers some of the issues and concerns around collecting and publishing sensitive data during the pandemic, as well as the perceptions in the overall population on these issues. Damien Desfontaines (Google) and Amaç Herdağdelen (Facebook) will discuss privacy protections in mobility data publication, Sarah LaRocca (Facebook) will discuss issues surrounding data collection on symptoms during the pandemic, Jennifer Childs (U.S. Census Bureau) will present views on privacy and confidentiality towards surveys, and Frederic Gerdorn will expand that to privacy attitudes to contact tracing. ASA Fellows and former Committee members Alan Karr and Aleksandra Slavković (Penn State) will serve as discussants.
For the bi-annual 2021 World Statistics Congress in July, the session on “Formal privacy methods in NSO: Challenges and Solutions,” also organized by Committee chair Vilhuber, brings an international perspective to the recent advances in privacy protection, in particular from the point of view of implementations. Challenges faced, and the reasons why national statistical offices (NSO) have tackled these challenges will be discussed in a virtual conference accessible to North American participants as well. John Abowd (Chief Scientist at the US Census Bureau) will discuss differential privacy (DP) in the 2020 US Census of Population and Housing, and his Census Bureau colleague Andrew Foote will present the use of DP for the publication of education statistics. Gerome Miklau (Tumult Labs) will discuss privacy-accuracy trade-off in a DP release of earnings statistic, Joseph Chien (Australian Bureau of Statistics) and Jörg Drechsler (IAB Germany) will discuss privacy implementations for agriculture statistics and geocoded data, respectively.
Early registration for the World Statistics Congress ends June 11, and JSM early registration ends June 15.