Census Bureau Seeks Feedback on Privacy System

Categories: Big Data, Opportunities

Tags: , ,

The digital age demands constant vigilance to combat threats that come along with technological advancements, and in that vein the U.S. Census Bureau is implementing a new disclosure avoidance system for data stewardship. Statistical safeguards — referred by the Census Bureau as “disclosure avoidance” — aim to prevent tracing personal information to a specific individual. Research on the 2010 Census suggested that existing safeguards would be insufficient to protect privacy in 2020.

The system to safeguard the 2020 Census data is referred as “differential privacy” in scientific circles. Safeguarding the data involves a mathematical algorithm that precisely controls the use of noise —  also known as uncertainty — to be added.

Census Bureau scientists John M. Abowd and Victoria A. Velkoff explained the concept in a blog post at the Census.gov website:

With the new method, the uncertainty (noise) is added to the statistics in the tables themselves.  The traditional method added uncertainty only to some records and variables deemed at higher risk, and prior to tabulation. The new method allows us to precisely control the amount of uncertainty that we add according to privacy requirements. 

The Census Bureau produced demonstration data products to familiarize the public with the disclosure avoidance system. Data products produced by the Census demonstrate the current computational capabilities via tests on data from the 2010 Census.

The Census Bureau is encouraging data analyst to review the data and provide input as the bureau continues to finalize the disclosure avoidance system. The National Academy of Sciences Committee on National Statistics is hosting a workshop about the files which will feature data experts presenting their analyses about the demonstration files. The workshop will take place on December. 11-12. which requires prior registration through the U.S. Census website.

Feedback about the disclosure avoidance system or the data products may be emailed to dcmd.2010.demonstration.data.products@census.gov. While comments which are submitted prior do December 4 will help in planning of the workshop, there is no present deadline for comments to be submitted to the National Academy website.

Leave a Reply