Federal Data Users

A new proposed rule, scheduled for publication tomorrow (Sept. 10), would make two independent but related changes to the decennial census. First, it would establish new residence criteria under which foreign citizens who are neither U.S. citizens nor lawful permanent residents would not be counted for congressional apportionment. Second, it would prohibit questions about race, ethnicity, and sexual orientation on questionnaires used for census enumeration, while still allowing questions on biological sex, date of birth, and household relationships for data verification purposes. 

These changes would mark a major break with more than two centuries of census practice: Race data have been collected in every census since 1790, and both citizens and noncitizens who usually reside in the United States have historically been included in congressional apportionment.

The proposal doesn't come as a surprise. The rulemaking first became public in July, when it was submitted to the White House for review, and an earlier draft obtained by TPM revealed many of the details. APDU and several partner organizations held a rapid-response briefing on the changes experts expected might be under consideration.

If finalized as proposed, the changes would alter congressional apportionment and Electoral College allocations, reshape the data available for redistricting and civil rights enforcement, and change the comparability of census data over time. They would also increase the role of administrative records in the census, including data obtained from other federal agencies. 

The Census Bureau is requesting public comments, due 30 days after publication. This is an important opportunity for data users to explain how the proposed changes would affect the accuracy, usefulness, and availability of census data.

Read the proposed rule

Read NPR's explainer

------------------------------
Mark Mather
Associate VP
PRB
------------------------------

As a first step, I would encourage groups/indviducals to write a letter asking for an extension. A 30 day comment period is meaningless for such a drastic change. 

Best, 

Meeta

------------------------------
Meeta Anand
The Leadership Conference on Civil and Human Right
------------------------------

Does anyone know if someone's working on a simulation of how the change in definition of usual residence would affect total state apportionments? If so, please contact me privately.  Thanks.

------------------------------
Michele Hayslett
michele_hayslett@unc.edu
UNC Library
------------------------------

All,

We’re going to have lots of questions that need answers, and one of mine is….”why now…?” Was it became the draft reg leaked so the media chatter has pushed this forward for a posting before the midterms? Meeta is right…proposed comment period is too short, and will stress the data community’s capacity to respond. Perhaps that’s the point.

And from my own “historical expertise” bailiwick….watch for the historical arguments. It’s pretty easy to document that “race” has been a census data collection point since 1790. And I’ve already seen how comments about how losing that variable is going to damage sampling frames, including those used in the private sector.

One small point I’ll add. In 2018 when Commerce Secretary Wilbur Ross tried unsuccessfully to put a citizenship question on the decennial, he claimed that he was simply “restoring” a query that had been asked before. It turned out that was not historically true. A “citizenship question” had never been asked of the entire US population. Before 1940, before sampling, and the “short form-long form" distinction, It was asked off and on as a subquestion of the “place of birth” question for foreign born residents only. And post 1940, it was asked on the sample, the long form, but never on the complete count questionnaire. So this proposal raises that issue again. The United States has never asked for information on the citizenship of the entire population.


Margo Anderson
margo@uwm.edu


Original Message:
Sent: 9/9/2026 3:19:00 PM
From: Michele Hayslett
Subject: RE: Census Bureau proposes major changes to the 2030 Census

Does anyone know if someone's working on a simulation of how the change in definition of usual residence would affect total state apportionments? If so, please contact me privately.  Thanks.

------------------------------
Michele Hayslett
michele_hayslett@unc.edu
UNC Library
------------------------------

My first question is, as is often the case with this administration, can they do something like this without congressional approval?  Doesn't Congress fund the Census, and thus responsible for approving changes of this magnitude?

------------------------------
Rob Schiffer
FuturePlan Consulting, LLC
------------------------------

Original Message:
Sent: 09-09-2026 04:02 PM
From: Margo Anderson
Subject: Census Bureau proposes major changes to the 2030 Census

All,

We're going to have lots of questions that need answers, and one of mine is…."why now…?" Was it became the draft reg leaked so the media chatter has pushed this forward for a posting before the midterms? Meeta is right…proposed comment period is too short, and will stress the data community's capacity to respond. Perhaps that's the point.

And from my own "historical expertise" bailiwick….watch for the historical arguments. It's pretty easy to document that "race" has been a census data collection point since 1790. And I've already seen how comments about how losing that variable is going to damage sampling frames, including those used in the private sector.

One small point I'll add. In 2018 when Commerce Secretary Wilbur Ross tried unsuccessfully to put a citizenship question on the decennial, he claimed that he was simply "restoring" a query that had been asked before. It turned out that was not historically true. A "citizenship question" had never been asked of the entire US population. Before 1940, before sampling, and the "short form-long form" distinction, It was asked off and on as a subquestion of the "place of birth" question for foreign born residents only. And post 1940, it was asked on the sample, the long form, but never on the complete count questionnaire. So this proposal raises that issue again. The United States has never asked for information on the citizenship of the entire population.


Margo Anderson
margo@uwm.edu



Original Message:
Sent: 9/9/2026 3:19:00 PM
From: Michele Hayslett
Subject: RE: Census Bureau proposes major changes to the 2030 Census

Does anyone know if someone's working on a simulation of how the change in definition of usual residence would affect total state apportionments? If so, please contact me privately.  Thanks.

------------------------------
Michele Hayslett
michele_hayslett@unc.edu
UNC Library
------------------------------

------------------------------
Mark Mather
Associate VP
PRB
------------------------------

Original Message:
Sent: 09-10-2026 02:30 PM
From: Rob Schiffer
Subject: Census Bureau proposes major changes to the 2030 Census

My first question is, as is often the case with this administration, can they do something like this without congressional approval?  Doesn't Congress fund the Census, and thus responsible for approving changes of this magnitude?

------------------------------
Rob Schiffer
FuturePlan Consulting, LLC
------------------------------


Original Message:
Sent: 09-09-2026 04:02 PM
From: Margo Anderson
Subject: Census Bureau proposes major changes to the 2030 Census

All,

We're going to have lots of questions that need answers, and one of mine is…."why now…?" Was it became the draft reg leaked so the media chatter has pushed this forward for a posting before the midterms? Meeta is right…proposed comment period is too short, and will stress the data community's capacity to respond. Perhaps that's the point.

And from my own "historical expertise" bailiwick….watch for the historical arguments. It's pretty easy to document that "race" has been a census data collection point since 1790. And I've already seen how comments about how losing that variable is going to damage sampling frames, including those used in the private sector.

One small point I'll add. In 2018 when Commerce Secretary Wilbur Ross tried unsuccessfully to put a citizenship question on the decennial, he claimed that he was simply "restoring" a query that had been asked before. It turned out that was not historically true. A "citizenship question" had never been asked of the entire US population. Before 1940, before sampling, and the "short form-long form" distinction, It was asked off and on as a subquestion of the "place of birth" question for foreign born residents only. And post 1940, it was asked on the sample, the long form, but never on the complete count questionnaire. So this proposal raises that issue again. The United States has never asked for information on the citizenship of the entire population.


Margo Anderson
margo@uwm.edu




Original Message:
Sent: 9/9/2026 3:19:00 PM
From: Michele Hayslett
Subject: RE: Census Bureau proposes major changes to the 2030 Census

Does anyone know if someone's working on a simulation of how the change in definition of usual residence would affect total state apportionments? If so, please contact me privately.  Thanks.

------------------------------
Michele Hayslett
michele_hayslett@unc.edu
UNC Library
------------------------------

Thanks, Mark.  For some reason, this hasn't received as much press as it deserves. I did submit comments, per the earlier link, though this administration doesn't seem to care about other opinions.  I hope that enough folks are paying attention that a lawsuit will be forthcoming.

------------------------------
Rob Schiffer
FuturePlan Consulting, LLC
------------------------------

Original Message:
Sent: 09-11-2026 10:08 AM
From: Mark Mather
Subject: Census Bureau proposes major changes to the 2030 Census

Good questrion--my understanding is that while Congress has constitutional authority over the census, the Commerce Secretary has a lot of discretion in how it's implemented. Under Title 13, the Secretary can determine the "form and content" of the census, so changes like these don't automatically require Congressinal approval.

Congress can still intervene though, and the courts can review whether the administration has exceeded its authority. Excluding noncitizens from apportionment raises a direct constitutional issue, while eliminating race and ethnicity questions raises questions about statutory obligations, redistricting and Voting Rights Act reauirements, and whether the CB has adequately justified a break with more than 200 years of census practice.

I'd be interested in how others view the legal authority here.

------------------------------
Mark Mather
Associate VP
PRB
------------------------------


Original Message:
Sent: 09-10-2026 02:30 PM
From: Rob Schiffer
Subject: Census Bureau proposes major changes to the 2030 Census

My first question is, as is often the case with this administration, can they do something like this without congressional approval?  Doesn't Congress fund the Census, and thus responsible for approving changes of this magnitude?

------------------------------
Rob Schiffer
FuturePlan Consulting, LLC
------------------------------



Original Message:
Sent: 09-09-2026 04:02 PM
From: Margo Anderson
Subject: Census Bureau proposes major changes to the 2030 Census

All,

We're going to have lots of questions that need answers, and one of mine is…."why now…?" Was it became the draft reg leaked so the media chatter has pushed this forward for a posting before the midterms? Meeta is right…proposed comment period is too short, and will stress the data community's capacity to respond. Perhaps that's the point.

And from my own "historical expertise" bailiwick….watch for the historical arguments. It's pretty easy to document that "race" has been a census data collection point since 1790. And I've already seen how comments about how losing that variable is going to damage sampling frames, including those used in the private sector.

One small point I'll add. In 2018 when Commerce Secretary Wilbur Ross tried unsuccessfully to put a citizenship question on the decennial, he claimed that he was simply "restoring" a query that had been asked before. It turned out that was not historically true. A "citizenship question" had never been asked of the entire US population. Before 1940, before sampling, and the "short form-long form" distinction, It was asked off and on as a subquestion of the "place of birth" question for foreign born residents only. And post 1940, it was asked on the sample, the long form, but never on the complete count questionnaire. So this proposal raises that issue again. The United States has never asked for information on the citizenship of the entire population.


Margo Anderson
margo@uwm.edu





Original Message:
Sent: 9/9/2026 3:19:00 PM
From: Michele Hayslett
Subject: RE: Census Bureau proposes major changes to the 2030 Census

Does anyone know if someone's working on a simulation of how the change in definition of usual residence would affect total state apportionments? If so, please contact me privately.  Thanks.

------------------------------
Michele Hayslett
michele_hayslett@unc.edu
UNC Library
------------------------------

In addition to the awful substantive implications of this proposed rule, one thing that jumped out for me was FRN's discussion of the Census Bureau's access to admin data and potentially other government data sets.  I was curious what others thought about this.

In particular, in the middle of the PDF (see attached highlighted excerpt), there are several paragraphs emphasizing that the Census Bureau has "enormously broad access to administrative data" and "to a vast and expanding universe of records and information.... The Census Bureau also has a highly-advanced data linkage infrastructure-one that has only improved in recent years as the Census Bureau has continued to develop and integrate new technologies."

To me, this is very worrisome because I think it signals that the administration has discovered how useful the Census Bureau can be to help further their nefarious purposes.  I think the stakeholder community has been operating under the assumption that because the population data collected by the Bureau is either anonymized or protected by Title 13, the Census Bureau wouldn't be as important to the current admin as other federal agencies such as IRS, etc. in terms of obtaining data they might want.

But I think the quotes above make it clear that the admin now views the Bureau as a way to access and manipulate so much more data.  It would vastly undermine the trust that the public has in the Census Bureau, which greatly compounds the worry about adding a question on citizenship or legal status and removing questions about race.

------------------------------
Steven Romalewski
CUNY Graduate Center
------------------------------

I agree with Steve, noticed the same thing, and have been reading and rereading the FRN to try to understand what is going on.
Sent from my iPhone


Original Message:
Sent: 9/11/2026 6:03:00 PM
From: Steven Romalewski
Subject: RE: Census Bureau proposes major changes to the 2030 Census

In addition to the awful substantive implications of this proposed rule, one thing that jumped out for me was FRN's discussion of the Census Bureau's access to admin data and potentially other government data sets.  I was curious what others thought about this.

In particular, in the middle of the PDF (see attached highlighted excerpt), there are several paragraphs emphasizing that the Census Bureau has "enormously broad access to administrative data" and "to a vast and expanding universe of records and information.... The Census Bureau also has a highly-advanced data linkage infrastructure-one that has only improved in recent years as the Census Bureau has continued to develop and integrate new technologies."

To me, this is very worrisome because I think it signals that the administration has discovered how useful the Census Bureau can be to help further their nefarious purposes.  I think the stakeholder community has been operating under the assumption that because the population data collected by the Bureau is either anonymized or protected by Title 13, the Census Bureau wouldn't be as important to the current admin as other federal agencies such as IRS, etc. in terms of obtaining data they might want.

But I think the quotes above make it clear that the admin now views the Bureau as a way to access and manipulate so much more data.  It would vastly undermine the trust that the public has in the Census Bureau, which greatly compounds the worry about adding a question on citizenship or legal status and removing questions about race.

------------------------------
Steven Romalewski
CUNY Graduate Center
------------------------------

Having spent several days parsing the very dense text and related regulatory impacts, I second Meeta's call for people to submit requests to extend the comment period.

Relatedly, on Friday 9/18 we're doing a briefing on the proposed changes. Join us

The Census Is in Trouble: What You Need to Know About Proposed Changes

https://us06web.zoom.us/webinar/register/WN_6HIA7xWGT9mdBFCQ_TfODg#/registration 

------------------------------
Beth Jarosz
------------------------------

Here is another question. Footnote 11 starts off: "Questions about tribal membership are not necessarily questions about race and/or ethnicity." Is this a hint that the questionnaire might include American Indian or Alaska Native? Or that Census might use tribal records to impute that category? 

------------------------------
Connie Citro
CNSTAT/NAS
------------------------------

Original Message:
Sent: 09-14-2026 03:53 PM
From: Beth Jarosz
Subject: Census Bureau proposes major changes to the 2030 Census

Having spent several days parsing the very dense text and related regulatory impacts, I second Meeta's call for people to submit requests to extend the comment period.

Relatedly, on Friday 9/18 we're doing a briefing on the proposed changes. Join us

The Census Is in Trouble: What You Need to Know About Proposed Changes

https://us06web.zoom.us/webinar/register/WN_6HIA7xWGT9mdBFCQ_TfODg#/registration 

------------------------------
Beth Jarosz
------------------------------

I noticed that same footnote, Connie. I was wondering whether that meant they may ask about tribal membership on the questionnaire?

------------------------------
David Van Riper
IPUMS
------------------------------

Original Message:
Sent: 09-15-2026 02:22 PM
From: Connie Citro
Subject: Census Bureau proposes major changes to the 2030 Census

Here is another question. Footnote 11 starts off: "Questions about tribal membership are not necessarily questions about race and/or ethnicity." Is this a hint that the questionnaire might include American Indian or Alaska Native? Or that Census might use tribal records to impute that category? 

------------------------------
Connie Citro
CNSTAT/NAS
------------------------------


Original Message:
Sent: 09-14-2026 03:53 PM
From: Beth Jarosz
Subject: Census Bureau proposes major changes to the 2030 Census

Having spent several days parsing the very dense text and related regulatory impacts, I second Meeta's call for people to submit requests to extend the comment period.

Relatedly, on Friday 9/18 we're doing a briefing on the proposed changes. Join us

The Census Is in Trouble: What You Need to Know About Proposed Changes

https://us06web.zoom.us/webinar/register/WN_6HIA7xWGT9mdBFCQ_TfODg#/registration 

------------------------------
Beth Jarosz
------------------------------

The comment period was just EXTENDED! 🎉

https://public-inspection.federalregister.gov/2026-20328.pdf

------------------------------
Beth Jarosz
------------------------------

The Independent Census Scientific Advisory Committee met on September 17th, to review these issues. The ICSAC raised concerns about the proposed changes and passed recommendations.

Presentations, Recommendations, and Public Comments, as well as a video, are available on the web site of the Independent Census Scientific Advisory Committee.

------------------------------
Kristin Moore
Child Trends
------------------------------

Original Message:
Sent: 10-02-2026 09:17 AM
From: Beth Jarosz
Subject: Census Bureau proposes major changes to the 2030 Census

The comment period was just EXTENDED! 🎉

https://public-inspection.federalregister.gov/2026-20328.pdf

------------------------------
Beth Jarosz
------------------------------

Barry Johnson has a helpful piece about the NPRM mention of using tax records to determine usual residence on the Tax Policy Center website: Using Tax Records To Verify Residence Could Put 2030 Census Accuracy At Risk.

Steve 

------------------------------
Steve Pierson
American Statistical Association
------------------------------