Production Settings Demonstration Data Released with Today’s Redistricting Data
The Census Bureau today released a new set of demonstration data that reflect the Disclosure Avoidance System settings used for production of the official 2020 Census Redistricting Data (Public Law 94-171) Summary File, which was also released today.
This is the sixth and final set of Privacy-Protected Microdata Files (PPMFs) for the redistricting data that allow data users to compare the effect of the Disclosure Avoidance System settings on previously published 2010 Census data. These and previous PPMFs are only intended to demonstrate the redistricting data, not the Demographic and Housing Characteristics File (DHC) or other 2020 Census data products.
The Census Bureau released metrics and other materials associated with today’s PPMFs on July 1 (see: Detailed Summary Metrics, Data Metrics Overview, Privacy-loss Budget Allocation). As with previous PPMFs, IPUMS National Historical GIS has converted the PPMFs into more-user-friendly tabular-format files. Access all materials via our Demonstration Data and Progress Metrics page.
Updated Reliability and Variability Study Findings
We also recently released an update to the Research Paper: Assessing the Reliability and Variability of the TopDown Algorithm for Redistricting Data using the production settings on our 2010 Census data. Our research found that the production settings produced more reliable data for smaller block groups, Minor Civil Divisions (e.g. towns, townships), and places than had the previous (April 28) settings. For any block group with a total count of between 450 and 499 people (down from 550 to 599 people), and for Minor Civil Divisions and places containing between 200 and 249 people (down from 350 to 399 people), the difference between the TopDown Algorithm ratio of the largest demographic group and the corresponding swapping algorithm’s ratio for the largest demographic group is less than or equal to 5 percentage points at least 95% of the time.
In other words, using block groups as an example, at least 95% of block groups across the country that have between 450 and 499 people have a count for the largest demographic group that’s less than 5% different from the count using the swapping algorithm when expressed as a percentage of the total population. (See slides and more from our August 10 webinar on the updated paper.)
Next Steps
For almost two years, independent experts, stakeholders, and data users have provided extensive feedback at every stage of development of the new Disclosure Avoidance System. From the earliest “beta” version in October 2019 to the production version in today’s redistricting release, your feedback proved invaluable to development of the finely tuned “TopDown Algorithm” that protects data confidentiality in the digital age.
We have now shifted our full attention to building out the algorithms needed to produce the more detailed data products from the 2020 Census. We pledge our continued transparency and engagement as we work with you to develop those products.
Useful Links:
2021 Key Dates, Redistricting (P.L. 94-171) Data Product
August 12 (Today):
- Release 2020 Census P.L. 94-171 Redistricting data in the format we’ve used in previous decades, sometimes referred to as the Legacy Format Summary File.*
- Concurrently, release Privacy-Protected Microdata Files (PPMFs) that demonstrate the impact of the production version of the Disclosure Avoidance System using 2010 Census data. These PPMFs only apply to redistricting data, not the Demographic and Housing Characteristics File (DHC) or other 2020 Census data products.
By September 30:
- Release the TopDown Algorithm production code base for P.L. 94-171 redistricting summary data file and related technical papers.
- Release 2020 Census P.L. 94-171 data in a more user-friendly format** and “Introduction to Differential Privacy” Handbook.
* Released via Census Bureau FTP site.
** Released via data.census.gov.
Was this forwarded to you?
Sign up to receive your own copy!
|
