Disclosure Avoidance and Differential Privacy

Purpose

Before the Census Bureau publishes any statistic, they apply safeguards that help prevent someone from being able to trace that statistic back to a specific respondent.

They call these safeguards “disclosure avoidance,” these methods are also known as “statistical disclosure controls” or “statistical disclosure limitations.”

Although it might appear that a published table shows information about a specific individual, the Census Bureau has taken steps to disguise the original data in such a way that the results are still useful. These steps include using statistical methods such as “data swapping” and “noise injection.”

Differential Privacy

The 2020 Census will use a new privacy protection system known in scientific circles as “differential privacy,” designed specifically for the digital age. Differential Privacy has generated some confusion and concern among census data users and the links below will help users understand what it is and the impact it will have on data products. The U.S. Census Bureau has also created a handbook on differential privacy and how it will be used in the 2020 Census. That handbook can be found here: Handbook on 2020 Census Disclosure Avoidance.

Additional a fact sheet has been released to help data users: Understanding Disclosure Avoidance-Related Variability in the 2020 Census Redistricting Data