Changes to Census Bureau Data Products
The Census Bureau has a long history of protecting the confidentiality of respondents to the census and the American Community Surveys. As part of current confidentiality protection, the Census Bureau “adds noise” to data products in ways they cannot release. The proposed new disclosure avoidance system using differential privacy is purported to be more open and transparent to users. But the new system may come with a significant trade-off in data accuracy, making the public data useless for many applications. We are following this conversation, and we will share relevant information from the Census Bureau and others as it is released.
- NEW: Task Force on Differential Privacy for Census Data, Implications of Differential Privacy for Census Bureau Data and Research
- John M. Abowd and Ian M. Schmutte, An Economic Analysis of Privacy Protection and Statistical Accuracy as Social Choices
- John Abowd presentation to Census Scientific Advisory Committee: Disclosure Avoidance for Block Level Data and Protection of Confidentiality in Public Tabulations
We will continue to gather relevant information for the IPUMS user community and post here and share via IPUMS Twitter.
1940 Census Data
The Census Bureau is testing their methods on 1940 census data and they plan to release algorithms allowing registered IPUMS users to download and test them. Data are available and we will link to Census Bureau algorithms when available.
Federal Register Notice
The official notice deadline has passed for Federal Register Notice 83 FR 34111. We issued the following guidance to IPUMS users on responding to the notice.
We strongly encourage IPUMS users to respond to Federal Register Notice 83 FR 34111 regarding data products for the 2020 Census by November 8, 2018. Email responses to POP.2020.DataProducts@census.gov, referencing Federal Register Notice 83 FR 34111. Please copy firstname.lastname@example.org on your reply.
If you use ACS or census microdata:
- See Steve Ruggles’s email to the IPUMS user community.
- Two suggestions for your reply:
- Describe your research. What is your unit of analysis -- individuals, households, PUMAs etc? Do you use information across the household or at the community level (as independent or dependent variables)?
- Request opportunities for researchers to compare data under current methods and proposed methods in order to provide useful feedback.
If you use ACS or census summary files: