USBC-2026-0628; Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions

Federal Comment

These changes threaten the constitutional integrity, accuracy, and utility of the nation’s most comprehensive population count, with far reaching consequences for representative government, public education, civil rights enforcement, and the allocation of public resources.

Submitted on: November 2, 2026

November 2, 2026

Submitted via regulations.gov

Howard Lutnick
Secretary of Commerce
U.S. Department of Commerce
1401 Constitution Avenue NW
Washington, DC 20230

Re: USBC-2026-0628; Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Question

Dear Secretary Lutnick:

On behalf of the approximately three million members of the National Education Association (NEA), we respectfully submit these comments in response to the U.S. Department of Commerce’s proposed rule, “Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions,” published in the Federal Register on September 10, 2026. NEA strongly opposes this proposal because it would fundamentally alter the longstanding principles governing who is counted in the decennial census, where individuals are counted, and what demographic information is collected. These changes threaten the constitutional integrity, accuracy, and utility of the nation’s most comprehensive population count, with far reaching consequences for representative government, public education, civil rights enforcement, and the allocation of public resources.

The proposal raises two fundamental and interrelated concerns. First, the proposed residence criteria would depart from longstanding census practices by making lawful presence a condition of residence, replacing Census Day with an extended enumeration period, and expanding the role of administrative records in determining where individuals reside. These changes risk excluding millions of people from the population count, misassigning others to locations that do not reflect their usual residence, and disproportionately undercounting populations whose living arrangements may not be accurately captured by administrative data. NEA believes these changes raise serious constitutional and statutory concerns and would undermine the requirement to count the population accurately for congressional apportionment and other essential purposes.

Second, the proposed restrictions on demographic questions would prohibit the collection of race and ethnicity information in the decennial census and foreclose the ability to consider sexual orientation questions in future censuses. These restrictions would represent a significant departure from longstanding census practice and would deprive policymakers, researchers, educators, and communities of critical information about the nation’s population. Without reliable demographic data at the local level, governments and school districts would be less able to plan for student enrollment, allocate education funding, address disparities in educational opportunity, monitor segregation, and enforce federal civil rights laws. The loss of these data would also weaken the reliability of the American Community Survey and other federal statistical programs that depend on decennial census counts as a foundational population benchmark.

These substantive concerns are compounded by serious shortcomings in the proposed rulemaking process. The Department has not provided a sufficiently transparent and comprehensive assessment of the proposed changes’ methodological, legal, operational, and distributional consequences, not has it demonstrated that the changes are supported by the research, expert consultation, and rigorous testing necessary to justify such a consequential departure from established practice. The limited opportunity for public review further constrains the ability of affected communities, statistical experts, and other interest holders to evaluate the proposal and its potential effects.

For NEA and its members, an accurate and comprehensive census is essential to ensuring that public resources reflect the needs of the communities they serve. Census data inform decisions about school funding, student services, staffing, school facilities, transportation, and the distribution of federal resources, while also supporting the broader statistical infrastructure on which public policy and research rely. People omitted from the census will continue to live in their communities and depend on public services; inaccurate data will simply make it more difficult for governments and institutions to plan for and meet those needs.

For these reasons, NEA urges the Department to withdraw the proposed rule in its entirety. The Department should preserve the longstanding objective of counting everyone once, only once, and in the right place; retain the collection of essential demographic information; and ensure that any future changes to the census methods are guided by transparent evidence, rigorous testing, meaningful consultation, and a demonstrated commitment to improving data quality. The following comments explain the legal, methodological, and practical grounds for NEA’s opposition and the consequences these proposed changes would have for students, educators, communities, and the nation as a whole.

Comment Period and Process

The Commerce Department and Census Bureau propose significant changes that alter how the census has been conducted since 1790. The impact of these changes is substantial, and providing only 53 days for the public to comment on this proposed rule is insufficient. The Department initially provided a comment period that NEA considers inadequate given the scope and consequences of the proposed changes. The Census Bureau subsequently extended the deadline from October 13 to November 2, 2026. Although NEA appreciates this extension, the additional time does not resolve the underlying concern that the proposal warrants a more thorough period of public review, particularly given the need to assess its methodological, legal, operational, and distributional consequences.

In addition, the proposal does not appear to be supported by a sufficiently transparent and comprehensive assessment of its methodological consequences, not does the accompanying record demonstrate meaningful consultation with the statistical experts, affected communities, and other interest-holders needed to evaluate its likely effects. Past proposals to change the census have been introduced in conjunction with methodological research that outlines benefits and risks, and have involved extensive public input and consultation with technical experts. In June 2025, the Trump Administration terminated the three census advisory committees – the National Advisory Committee on Racial, Ethnic and Other Populations (NAC), the Census Scientific Advisory Committee (CSAC), and the 2030 Census Advisory Committee (2030 CAC) – in accordance with Executive Order 14217, “Commencing the Reduction of the Federal Bureaucracy.” NEA is concerned that the termination of these advisory committees, together with the absence of a clearly documented, independent assessment in the publicly available rulemaking record, leaves critical questions about the proposal’s accuracy, feasibility, and distributional effects insufficiently addressed. This lack of research and engagement is a serious shortcoming and is yet another reason that NEA opposes the proposed changes.

Finally, NEA is concerned that codifying detailed residence criteria and restrictions on demographic questions in the Code of Federal Regulations (CFR) could unnecessarily constrain the Census Bureau’s ability to adapt its methods in response to changing population characteristics, methodological evidence, and operational needs. The CFR is intended for “general and permanent rules,” which runs counter to the ability to adapt the decennial census to changing societal norms and needs. Both the residence criteria and the approach to demographic questions are regularly reviewed and revised to achieve the most accurate data about the U.S. population. We request that any changes to the residence criteria or the demographic questions become Census Bureau policy rather than part of the Code of Federal Regulations, in line with past practice.

Proposed Residence Criteria (15 CFR Part 60)

The Proposed Residence Criteria Are Unlawful

The proposed residence criteria would exclude many people who live in the United States based on immigration status, including people who have longstanding ties to their communities. NEA believes that these exclusions conflict with the Constitution’s requirements to apportion representation based on the number of persons in each State, exceed the authority granted by the Census Act, and depart sharply from the longstanding practice of counting the resident population without making lawful immigration status a general prerequisite for enumeration.

The plain text of the U.S. Constitution clearly states that congressional apportionment requires “counting the whole number of persons in each State.” U.S. Const., 14th Amend. §2 (see also, U.S. Const. Art. 1 §2 requiring the count of the “whole number of free persons” and an “actual enumeration”). The Constitution does not use the word citizens when discussing the apportionment count but rather the word persons, even though the preceding sentence which lays out the qualifications for serving in Congress utilizes the term citizen; this was a deliberate choice signaling that all those residing in the U.S. must be counted. See U.S. Const. Art. 1 §2. The U.S. Supreme Court has consistently held that the word “persons” as used in the U.S. Constitution includes non-citizens and immigrants, regardless of their legal status. See Yick Wo v. Hopkins, 118 U.S. 356, 369 (1886) (The provisions of the 14th Amendment “are universal in their application to all persons . . . without regard to any differences of race, of color, or of nationality . . .”), Truax v. Raich, 239 U.S. 33, 34 (1915) (“The description in the Fourteenth Amendment of any person . . . includes aliens.”), Plyler v. Doe, 457 U.S. 202, 210 (1982) (“an alien is surely a ‘person’ in any ordinary sense of that term. Aliens, even aliens whose presence in this country is unlawful, have long been recognized as ‘persons’’’ under the U.S. Constitution), Zadvydas v. Davis, 533 U.S. 678, 693 (2001) (holding that all “persons” includes “aliens, whether their presence here is lawful, unlawful, temporary, or permanent.”). Excluding immigrants from the census count would be a direct violation of the Fourteenth Amendment, as well as the Constitution’s Enumeration clause, as refusing to count the entire population impacts “the population count itself.” Dept. of Commerce v. New York, 588 U.S. 752, 769 (2019).

Excluding large swaths of the population based on their immigration status, in addition to violating the Constitution, also violates the Census Act. The Census Act requires the Secretary of Commerce to “take a decennial census of population.” 13 U.S.C. § 141. Statutory text should be interpreted according to its “ordinary public meaning.” Bostock v. Clayton County, 590 U.S. 644, 655 (2020). Again, much like with the use of the term persons in the U.S. Constitution, the word “population” is distinct from the word citizen. The Cambridge English Dictionary defines population as “all the people living in a particular country, area, or place” – this definition does not rely on the concept of citizenship and is inclusive of immigrants, regardless of their legal status. The Department of Commerce may not adopt a regulation that exceeds its statutory authority or conflicts with controlling constitutional requirements. Indeed, Congress itself has repeatedly rejected attempts to exclude non-citizens from the census as it would be unconstitutional to do so via statute. As a 1929 Senate Legislative Counsel opinion stated, "there is no constitutional authority for the enactment of legislation excluding aliens from enumeration for purposes of apportionment of Representatives among the States." If Congress cannot make such a change to the census legislatively, an agency cannot make such a change via administrative regulations.

The understanding that the entire population, including immigrants with both lawful and unlawful statuses, must be counted is further buttressed by “the longstanding ‘practice of the government’” which “can inform [a court’s] determination of ‘what the law is.’” NLRB v. Noel Canning, 573 U.S. 513, 525 (2014) (first quoting McCulloch v. Maryland, 4 Wheat. 316, 401 (1819); then quoting Marbury v. Madison, 1 Cranch 137,177 (1803)). It has been an uninterrupted practice of the federal government to count all people residing in the U.S. since 1790, including non-U.S. citizens and many others who for large periods of our nation’s history were unable to vote or participate in our political system. The federal government has never excluded individuals from the census count on the basis of their immigration status as doing so would be a blatant violation of the Constitution and federal law.

In addition to the many legal problems with this proposal, there are many other factors that also counsel against excluding large categories of non-citizens from the apportionment count. First, undocumented immigrants, asylum applicants, asylees, refugees, DACA holders, people with pending legal permanent resident applications and other immigrants are long-settled residents in the United States. For instance, under federal law a refugee is a person who is unable or unwilling to return to his or her home country because of “persecution or a well-founded fear of persecution” due to race, membership in a particular social group, political opinion, religion, or national origin.

Federal law acknowledges that refugees cannot return to their home countries and that their residence is the U.S., so it is nonsensical for the Department to consider refugees as not residing in the U.S. Similarly, DACA holders have resided in the U.S. since childhood and know no other home than the U.S. which is why the government created this program to offer them protection from deportation.

Administrative Records and an Immigration Status Question Should Not Determine Census Inclusion

The Department states that it would utilize federal government data to determine respondents’ immigration status to exclude large numbers of immigrants from the census count. NEA opposes such a use of federal data because, as explained above, excluding immigrants from the census count is unconstitutional, it is likely that there will be errors resulting in excluding U.S. citizens, and it will chill participation in the census. As discussed above, the U.S. Constitution requires an “actual enumeration” and using government data to exclude individuals from being counted violates that requirement. In addition, the government’s data on immigration status have not been established as reliable and error-free, as evidenced by the federal government detaining U.S. citizens for immigration violations and its recent investigations of voter fraud. There is a very real risk that large numbers of U.S. citizens could be excluded; in 2018, the Census Bureau tried to match 2010 Census responses to administrative data but in 9 percent of cases (28 million out of 309 million people counted in 2010), the matches failed or there was insufficient information. For these reasons, federal data should not be used to determine census respondents’ immigration status.

The Department also seeks feedback on adding a direct question about immigration to the census. NEA opposes adding such a question. First, as discussed above, the U.S. Constitution requires the counting of immigrants in the census, so asking a question with the purpose of excluding immigrants is unconstitutional. Because the only justification for adding a question regarding immigration to the census is an unconstitutional justification, the Department cannot propose after-the-fact justifications for the addition of such a question. Second, numerous studies have found that asking about immigration status on the census reduces response rates to the census, leads people to omit members of their household, and makes census data less valid. Given that one in four children in the U.S. lives with at least one immigrant parent, it is likely that children in these households will be severely undercounted, even if the children themselves are citizens or permanent residents.

The Proposed Residence Criteria Would Undermine Accurate Enumeration and Residence Assignment

As noted on the Census Bureau’s page, Why We Conduct the Decennial Census of Population and Housing, the goal of the decennial census is to “count everyone once, only once, and in the right place.” The guidance for the 2020 census was designed to accomplish this goal by capturing each person’s “usual residence,” or the “place where a person lives and sleeps most of the time,” on a single “Census Day,” with clear rules about how more complicated situations should be handled. These situations include, but are not limited to, traveling or moving on Census Day, being born or dying on Census Day, and how to count college students, people experiencing homelessness, U.S. military personnel, and people at institutions such as hospitals or correctional facilities (see Table 1 below for more information).

The Department proposes replacing the Census Day with an “enumeration period” between January 3 and April 1, with “usual residence” defined as “the residence at which they have lawfully spent the greatest number of days during the census enumeration period,” and additional clarification that “A person is considered to have spent a day at a location if he or she spent at least one hour physically and lawfully present there that day.” As argued above, the Constitutional intent of the decennial census is not to count people who are “lawfully” present in the country but rather all individuals living in the United States. Beyond this, the Department provides no reasoning for moving away from the traditional method of anchoring residence around a single day, and its arbitrary justification for starting the period on January 3 is that it would match the halfway point of the congressional term.

Switching to an enumeration period would make it considerably more difficult to “count everyone once, only once, and in the right place.” Many people who spend all or part of the winter away from their usual place of residence would be counted in their winter locations rather than where they live most of the year, pay taxes, and vote, which would inaccurately shift the population count toward warmer, southern states. People who move late in the enumeration period may feel compelled to count themselves at their prior residence, even if they should rightly be counted at the location they moved to. Minor students at boarding schools would be counted as living in those locations rather than at their parents' or guardians' home, which was the guidance for the 2020 census. Most perplexingly, children who split their time between residences (e.g., children who move between two homes if their parents are divorced), could be counted as a "resident" of a home-based daycare facility if that is the place where they spend the greatest number of days as defined by the Department.

Further confusing the issue, the Department adds that "the Census Bureau intends to consider tax records filed after April 1 insofar as they are probative of persons' 'usual residence' from January 3 to April 1." This assumes that individuals can be connected to tax records and that those tax records accurately reflect a person’s “usual residence,” yet neither of these conditions is universally true. Many people, including children and college students, seasonal and mobile workers, people paid informally, and people without stable tax filings, may be counted incorrectly or missed entirely.

Analysis of IRS filings compared to the population shows that about 87-90% of the U.S. population is covered by individual income tax filings. Adding other tax information returns improves coverage somewhat, but dependent children are the most likely age group to inaccurately captured through IRS administrative data. In the words of one researcher, “The measurement of the non-filing population represented by children is, by far, the most problematic… Based on experience working with tax panel data, we know that [Social Security number (SSN)] reporting errors on tax returns are most common for dependent children.” These problems are also not equally distributed within the child population. A Census Bureau study of California children found that “older children, those of Non-Hispanic Black mothers and Hispanic mothers, children or parents with lower English proficiency, children whose mothers did not complete high school, and families with lower income-to-poverty ratio were less frequently reported in IRS 1040 records than other groups.”

In addition, tax records require a current mailing address, which may differ from an individual’s current residential address. People may, for example, file taxes using a post office box, their tax preparer's address, or an address different from their usual address. They may also move, so address information on their W-2 may be out-of-date, or they may be servicemembers for whom W-2s reflect duty stations rather than residential addresses. The Department offers no recognition of these critical problems and no guidance for how discrepancies between census responses and tax records will be resolved. Indeed, internal documents appear to show that the Treasury Department objected to this approach, stating that tax records “may be used as corroborating evidence of an individual’s residence, but no particular tax record is controlling, and the evidentiary weight assigned to such information should reflect the purpose, timing, and reliability of the particular data element.”

Guidance is also lacking for several situations that affect the count of children and college students. Table 1 provides a comparison between the enacted 2020 census residence guidance and the Department's proposal for the 2030 census. The shaded boxes highlight situations impacting counts of children for which the Department has provided no specific guidance. For the 2020 census, the Department provided clear instructions about how to count "babies and children of all ages, including biological, step, and adopted children, as well as grandchildren," how to count foster children, how to count college students, how to count people in residential school-related facilities, and how to count people in juvenile facilities. Additional guidance covered any individual, including children, who were away from their usual residence, lived or stayed in more than one place, moved on or around Census Day, and/or were experiencing homelessness. Absent clear guidance for these situations, it is highly likely that the undercounting of children and young adults will worsen significantly compared to past censuses.

Two examples illustrate these undercounting concerns very clearly: children who live in households headed by someone other than a parent and college students. Currently about 9 million children, or 1 in 8, live in a household headed by someone other than a parent. Most of these households are headed by grandparents, but they may also be headed by another relative, foster parent, or a non-relative. Census Bureau research shows that young children were less likely to be counted the further their householder’s relational “distance” from them, with coverage problems estimated at 16% for children living with grandparents, 22% for those living with other relatives, and 25% for those living with an unrelated adult.

College students have been counted where they live while in school since 1950, a rule that was upheld in court in 1971. The proposed residence criteria offers no specific guidance regarding college students, so they would fall under the general rule regarding the enumeration period. In most cases, students would be counted where they attend school, however, conflicts are likely to arise when trying to match to IRS data, as many are listed as dependents on their parents’ tax records. One estimate shows that in the counties with the 38 largest flagship and land-grant universities, the IRS count of individuals is only 73.5% of the American Community Survey population estimate, compared to 85.8% in all other counties. For both college students and children who live in households headed by someone other than a parent, the proposed criteria do not provide sufficient guidance and will inevitably lead to inaccurate enumeration of these populations, both of which are of high interest to NEA and its members.

Table 1. Residence Criteria for Living Situations for which People Commonly Request Clarification

Situation	2020 Census: Specific Guidance Provided	2030 Census: Specific Guidance Proposed
People Away from their Usual Residence on Census Day/during the Enumeration Period	Yes	No
Visitors on Census Day/during the Enumeration Period	Yes	No
Foreign Citizens in the United States	Yes	Yes
People Living Outside the United States	Yes	Yes
People Who Live or Stay in More than One Place	Yes	No
People Moving into or out of a Residence around Census Day/during the Enumeration Period	Yes	No
People who are Born or who Die around Census Day/during the Enumeration Period	Yes	Yes
Relatives and Nonrelatives	Yes	Minimal guidance is provided for "dependents," with no definition of this term; no guidance for other situations
People in Residential School-Related Facilities	Yes	No
College Students and Staff Living in College Housing	Yes	No
People in Health Care Facilities	Yes	No
People in Housing for Older Adults	Yes	No
U.S. Military Personnel	Yes	Yes
Merchant Marine Personnel on U.S. Flag Maritime/Merchant Vessels	Yes	Yes
People in Correctional Facilities for Adults	Yes	Yes
People in Group Homes and Residential Treatment Centers for Adults	Yes	No
People in Juvenile Facilities	Yes	Partial - no guidance for juveniles in non-correctional facilities
People in Transitory Locations	Yes	Yes
People in Workers' Residential Facilities	Yes	No
People in Religious-Related Residential Facilities	Yes	No
People in Shelters and People Experiencing Homelessness	Yes	No

The Proposed Residence Criteria Would Undermine Census Accuracy, Public Funding, and Education Planning

The residence criteria used for prior censuses have been used to determine where people are counted, not whether they are counted. If the Department's proposal is enacted, the 2030 census would intentionally undercount the nation's population in ways that would disproportionately impact those groups that are already historically undercounted: Black/African American, Hispanic/Latino, American Indian or Alaska Native, and children, particularly children under 5. Estimates of the overall undercount if the proposed residence criteria are used range from 15 million to 16.5 million people, or anywhere from 1 in 25 to nearly 1 in 20 U.S. residents. It would also count millions of people in the wrong place due to the replacement of Census Day with an enumeration period and missing or unclear guidance for complicated situations.

Implementing the Department's proposed residence criteria will not make these uncounted or miscounted people disappear. They will still live in the United States and will still use public schools, emergency rooms, roads, and public transit in the cities and counties where they live. What will have disappeared are the accurate data necessary for the public sector and private companies to adequately plan for, fund, and provide these vital services. While the Department's proposal argues that "the sole Constitutional purpose of the decennial census is to apportion the seats in the U.S. House of Representatives," the Census Bureau acknowledges that census data are used by “business owners … to make decisions, such as where to open new stores, restaurants, factories, and offices; where to expand operations; where to recruit employees; and which products and services to offer” and “to benefit your community” by determining the allocation of federal funding to states, counties, and communities.

As of FY2023, more than $2.24 trillion in federal spending was guided by decennial census data, including $129.8 billion in education spending on programs such as Title I, Head Start, services for students with disabilities, School Nutrition Programs, and Career and Technical Education Grants. In addition to its use in federal spending formulas, state and local education agencies use the census in their own funding allocations and to project school-age populations. The latter data are critical in helping decisionmakers decide when and where to open new K-12 schools and higher education institutions, to guide staffing decisions, to set up school transportation routes, and to determine the need for other services that are fundamental to school operations, such as roads, emergency services, and health services. Making any of these highly consequential decisions using data that intentionally omits and suppresses responses from millions of U.S. residents will be incredibly costly to federal, state, and local governments, will place undue burdens on the taxpayers that support these governments’ work, and will have negative impacts that resonate for years and possibly decades.

The Department asserts in its proposal that the American Community Survey (ACS) and other federal datasets can be used in place of the decennial census for purposes other than congressional apportionment. Yet the accompanying Preliminary Regulatory Impact Analysis (PRIA) states that “the ACS has several limitations compared with the decennial census,” including its lack of estimates for small geographic areas, specifically those with populations under 65,000. Even more concerningly, the PRIA notes:

"The accuracy of ACS total-population estimates may decline because the ACS will not have an updated full count of the U.S. population. Between censuses, the Census Bureau creates intercensal population estimates and revises them when new decennial census data become available. The ACS relies on those intercensal estimates to generate its sampling weights. Without new decennial census data to recalibrate those estimates, the ACS sampling weights, and therefore its population estimates, will become less reliable [emphasis added]."

In other words, the Census Bureau requires accurate decennial census data to ensure the reliability of the ACS. A less reliable ACS further expands the set of funding and planning decisions that would be negatively impacted by the Department’s proposal. In the realm of education policy, ACS data are used to plan supports for emergent multilingual learners and students with disabilities, determine student poverty rates, calculate student loan repayment models, align higher education programs to workforce needs, and much more. Within NEA, ACS data are used to track the number, demographics, and other characteristics of educators, including their salaries. Unreliable ACS data will make it far more difficult to advocate on behalf of our 3 million members.

The ACS is not the only statistical collection that relies on an accurate decennial census to ensure valid and reliable data. The decennial census, either directly or through the ACS, is the source of sampling weights and denominators for innumerous surveys, polls, and data collections used to inform every sector of society, including core federal statistical projects such as the Current Population Survey and the Consumer Expenditure Survey, which is used to calculate the Consumer Price Index. The private sector cannot compensate for this loss of high-quality data as it, too, uses the decennial census to benchmark its data collections.

In sum, NEA urges the Department to withdraw proposed 15 CFR part 60. The proposed criteria would make unlawful presence a condition of residence, introduce an insufficiently justified enumeration period and one hour threshold, and give administrative records a role that may conflict with individuals’ actual living arrangements. These changes risk both omissions and incorrect resident assignments, with consequences for congressional apportionment, education funding, civil rights enforcement, and the reliability of federal statistical programs. The Department should preserve the longstanding goal of counting each person once, in the correct location, and adopt any methodological changes only after transparent research, expert consultation, and rigorous testing demonstrate that they will improve rather than undermine census accuracy.

Proposed Regulations for Demographic Questions (15 CFR Part 70)

Prohibiting Race and Ethnicity Questions Would Break with Longstanding Census Practice

Every single census since our country’s founding has included a question regarding race – it is unprecedented to have a census and not include this question. The Census Act of 1790 mandated the collection of race and ethnicity data, charging the census takers to “distinguish the sexes and colors” of people within the country. “All three branches of Government have understood the Constitution to allow Congress, and by extension the Secretary, to use the census for more than simply counting the population. Since 1790, Congress has sought, or permitted the Secretary to seek, information about matters as varied as age, sex, marital status, health, trade, profession, literacy, and value of real estate owned.” Dept. of Commerce v. New York, 588 U.S. 752, 769 (2019). Indeed, the U.S. Supreme Court has “recognized the role of the census as a ‘linchpin of the federal statistical system by collecting data on the characteristics of individuals, households, and housing units throughout the country.’” Id. at 770 (citations and quotations omitted).

Over time, how the decennial census has gathered demographic data has changed to reflect the country’s changing laws, composition, and societal norms. Until 1960, census enumerators recorded their perception of a person’s race rather than individuals recording their own race. In 2000, it became possible to select more than one race, to more accurately reflect a population that increasingly does not fit into single racial or ethnic categories. In line with Statistical Policy Directive No. 15, initial plans for the 2030 Census included combining race and ethnicity into a single question. Yet at no point has the practice been to have no questions about race and ethnicity. Because of the unprecedented nature of this change and its significant consequences, NEA strongly opposes the proposed ban on demographic questions about race and ethnicity.

The Rule Should Preserve Flexibility to Consider Sexual-Orientation Questions and Retain Household Relationship Data

No prior census has asked about individuals’ sexual orientation, although the household relationship question has provided some indication of the number of LGBTQ+ households for decades. The household relationship question has been refined over the years to provide more accurate data, with changes based on rigorous research and testing. Through these data, we know that in 2020, there were 1.16 million same-sex couples in the United States, and that over 20% of those households were raising children. While it is unclear what will happen with the household relationship question under the current proposal, NEA asserts that this question must be maintained, as these data are critical for understanding the size and characteristics of same-sex couple households, including same-sex married couples, particularly in light of the Obergefell U.S. Supreme Court decision and federal and interstate recognition under the Respect for Marriage Act.

What is clear under the current proposal is that the Department would prohibit questions about sexual orientation for the 2030 Census and foreclose the Bureau’s ability to adopt such questions in the future, regardless of research or data needs. These data needs are very much present, as seen by a 2017 letter that Sens. Kamala Harris (D-CA) and Tom Carper (D-DE) submitted to the then-director of the Census Bureau regarding the 2020 Census. The letter stated that an appendix on sexual orientation and gender identity (SOGI) had been included in a report on new subjects planned for the 2020 Census, and that this report was then retracted and replaced by a version without the appendix. The senators add that:

"Enclosed are letters that detail the process by which SOGI was considered as a potential new subject to the Decennial Census and ACS. According to these letters, the Census Bureau received requests in 2016 from the Department of Justice (DOJ), the Department of Housing and Development (HUD), the Department of Health and Human Services (HHS), and the Environmental Protection Agency (EPA) to include SOGI on the 2020 Census and ACS. In particular, DOJ’s November 4, 2016 request to the Census Bureau to include SOGI outlined the legal authority that supports the necessity for DOJ to collect this information."

In short, four separate federal agencies had submitted requests that argued for the need for these data – a need that has not disappeared over the subsequent decade. Yet the Department’s proposal would prohibit the addition of a sexual orientation question on future censuses, without sufficient evidence, legal citations, or explanation for why that prohibition is warranted.

NEA opposes this unnecessary prohibition. Rather than preemptively prohibiting sexual orientation data collection, decision-making about adding new demographic questions to future censuses should be guided by a deliberative process. That should include rigorous research and testing, consultation with federal agencies about their legal and programmatic needs for the data – including needs at small geographic levels – and meaningful engagement with subject-matter experts and impacted communities.

The Department Has Not Adequately Justified Its Proposed Restrictions on Demographic Questions

The Department argues in its proposal that prohibiting – not just failing to include – questions about race and ethnicity, (plus sexual orientation, which has not appeared on any census to date) is appropriate because the purpose of the decennial census is congressional apportionment, and the count should not be “should not be distorted in any way by questions about immaterial personal characteristics.” While we believe that the Department is calling these characteristics “immaterial” with specific reference to congressional apportionment, we demonstrate above that the federal government itself acknowledges that the uses of the census extend far beyond this purpose. Even if one were to accept the Department’s extremely limited scope for the decennial census, decades of legislative and judicial action focused on the racial and ethnic composition of congressional districts indicate that these characteristics are far from “immaterial.”

The Department further asserts that prohibiting these questions will reduce response burden and thus improve response rates, although it provides no evidence to support this argument. In addition, the census “short form” is already very short – only about 10 questions – and research has shown that removing demographic questions does not improve response rates. Furthermore, Census Bureau research has shown that ethnic pride is a motivating factor for some to complete the decennial census, suggesting that prohibiting questions about race and ethnicity may actually suppress rather than improve the response rate.

The Department also claims that these changes are necessary to prevent the disclosure of identifiable data about individuals. The Census Bureau has long made data privacy, particularly the protection of personally identifiable information, a high priority, as this is a basic tenet of research ethics. Initially, these methods were quite simple, focusing on coarsening (i.e., rounding, aggregating, using ranges) and suppression (i.e., not reporting data). While these methods provide a level of disclosure avoidance, they prevent the release of detailed data to maintain the privacy of personal information. However, in the 1970s the Commerce Department began discussing the use of a set of advanced statistical techniques that fall under the umbrella of “noise infusion,” or the injection of random deviations into individual-level data in ways that do not change overall results. These methods were put into use starting with the 1990 Census and have been in use since that time.

In June 2026, the Commerce Department banned the use of noise infusion, leaving coarsening and suppression as the Census Bureau’s two options for disclosure avoidance. A coalition of five statistical associations released a statement soundly opposing this decision and the process used to come to it, stating that “the result of this ill-conceived order will be either less privacy for our personal information, or less usable data, or both,” and “this order is in conflict with the Executive Order on Restoring Gold Standard Science, which requires federal government scientific endeavors to be, among other things, transparent, collaborative and interdisciplinary, subject to unbiased peer review, and without conflicts of interest.” Without evidence to the contrary, it appears that the Department banned a set of standard, widely used disclosure avoidance methods, and then used its self-initiated inability to employ those methods as a justification for prohibiting the collection of data about race, ethnicity, and sexual orientation.

Eliminating Race and Ethnicity Questions Would Undermine Small Area Data and Civil Rights Enforcement

Accurate information about race and ethnicity helps us understand America as it actually is. It helps us see who lives in our communities, understand people's experiences, and recognize differences that broad averages can sometimes hide. Some of the most valuable products of the decennial census are population and subgroup estimates for small geographies, all the way down to census blocks, which are areas delineated by features such as roads, bodies of water, and county boundaries. Census blocks are the building blocks for larger units of analysis, such as block groups, tracts and places, and are the smallest units for which one can get population data by age, sex, and race/ethnicity. Data at this level of specificity are important because aggregated data may obscure lower-level trends – for example, state-level school-age population data may show a decline, while that population may be increasing for specific counties, cities, school districts, or neighborhoods within that state.

Within the field of education, the National Center for Education Statistics (NCES) is responsible for matching the geographic boundaries of school districts to census geographies, as in many states, school district boundaries do not match other political boundaries, such as county lines. Furthermore, many school districts are additionally divided into attendance areas for specific schools plus voting areas for school board seats. Three demographers who have worked with dozens of school districts described how catastrophic it would be if the decennial census did not include block-level population estimates for children by race and ethnicity:

"Reliable data on race and ethnicity of children at the block level is key to not only calculating the demographic trends of these populations but is also critical when attempting to balance attendance area enrollment by race/ethnicity. The calculation of age specific birth rates, student yield factors, and area mortality rates for these groups are also adversely affected, by inaccurate data rendering the forecast results virtually worthless. Inaccurate projections could result in taxpayer money being misspent and children ending up in overcrowded classrooms."

The demographers further stress that the negative financial and educational impacts of inaccurate or missing small-area population estimates will have impacts that last for years, as districts make misinformed decisions about school facilities, staffing, attendance areas, and support systems for specific subgroups of students. School districts are also impacted by decennial census data through their use to “[m]onitor and enforce equal employment opportunities under the Civil Rights Act of 1964” and to “[a]llocate funds to school districts for bilingual services under the Bilingual Education Act.”

Demographic data for small geographies are crucial for identifying, tracking, and addressing the impacts of segregation based on student race and ethnicity. Title VI of the Civil Rights Act of 1964 “prohibits discrimination on the basis of ‘race, color, or national origin’ in programs or activities that receive federal financial assistance,” which in the case of K-12 education includes state educational agencies, school districts, and vocational schools, among others. While districts and schools report data about student demographics directly to NCES and the Department of Education’s Office for Civil Rights (OCR), it is important to put these data in the context of broader community demographics. For example, The Segregation Tracking Project, a collaboration between UCLA and Stanford University, looks at both school segregation, using NCES data, and residential segregation, using census tract data. These comparisons are particularly important as racial and economic segregation have increased since the 1990s as districts were released from court-ordered desegregation plans and school choice policies have increased. Race and ethnicity information can help identify problems such as school segregation, develop solutions, and determine whether those solutions are actually working.

Decennial census data by race and ethnicity are also necessary for enforcement of federal laws, including the Voting Rights Act. Notably, prior to the 2020 census, the Department of Commerce argued that the inclusion of a citizenship question was necessary for the federal government to enforce the Voting Rights Act, a justification that the U.S. Supreme Court found to be specious. In a stark departure, the Department of Commerce now no longer wants to collect the block-level data needed for enforcement of the Voting Rights Act. These block-level data are utilized to create voting districts and ensure that they are representative. Without having block-level census data, it will be impossible to enforce the anti-discrimination provisions of the Voting Rights Act. The Census Bureau’s own website notes that race data are used to “monitor compliance with the Voting Rights Act and enforce bilingual election requirements.” Data does more than confirm that inequality exists. They help us see its scale, understand where it is happening and determine where public resources and policy attention may be most needed.

Eliminating Demographic Questions Would Degrade Federal Statistical Data

As discussed at length above, the decennial census provides the population baseline on which many federal and non-federal datasets rely, and this is also true for counts by race and ethnicity. The denominator used in nearly every statistical rate the federal government publishes by race and ethnicity is a population estimate anchored to the decennial count and carried forward. As stated in the PRIA:

"Numerous key dependencies have been structured around the reliable annual release of intercensal estimates for race and ethnicity, such as population controls for major federal surveys; denominators for crucial social and economic indicators (e.g., fertility/mortality rates, cancer rates, poverty rates, unemployment rates, labor force participation, etc.); required information for state and federal grants and other funding formulas; foundational data to support planning at the local level (infrastructure, real estate, and other investments); and essential inputs for countless research projects across academia, government, and the private sector."

Eliminating these population estimates will have catastrophic effects on the nation for at least a decade. In addition, it will be impossible to accurately weight data to reflect the population of the nation, a state, or a locality, which is standard practice for sample surveys.

In its proposal, the Department points to the ability to use the ACS and other federal administrative data as sources of demographic information and states that its proposal for the decennial census has no bearing on the use of questions about race and ethnicity in the ACS. However, our overall argument that the ACS cannot serve as a sufficient substitute for the decennial census under the current proposal is particularly true regarding the collection of data about race and ethnicity. As stated in the Preliminary Regulatory Impact Analysis (PRIA), “the power of the ACS is the ability to provide characteristics for race and ethnicity groups; whereas population totals for race and ethnicity groups from the decennial census and population estimates are more reliable” [emphasis added]. This is particularly true for small racial and ethnic groups, such as American Indian and Alaska Native, Native Hawaiian and Pacific Islander, and detailed Asian American and Hispanic ethnic origin groups, as these groups are the most difficult to capture accurately through sample surveys such as the ACS. The PRIA further notes that “decennial census counts of the population by race and ethnicity are relied upon, as there are no other comparably accurate, comprehensive, and reliable sources of this information. Omitting race and ethnicity from the 2030 Census questionnaire would require the development of estimates from a base population derived from outdated and/or less accurate data and would deteriorate the accuracy of the estimation method itself” [emphasis added].

NEA urges the Department to withdraw proposed 15 CFR part 70.

Conclusion

The Department of Commerce’s proposed changes to the decennial census would fundamentally undermine the accuracy, completeness, and usefulness of the nation’s most comprehensive population count. As detailed in these comments, the proposed residence criteria raises serious constitutional and statutory concerns, risk excluding millions of people from the apportionment count, and could assign individuals to locations that do not reflect their usual residence. The proposed restrictions on demographic questions would eliminate essential race and ethnicity data, foreclose consideration of sexual orientation questions in future censuses, and weaken the information available to policymakers, educators, researchers, and communities. Taken together, these changes threaten not only the integrity of the decennial census but also the broader statistical infrastructure that supports representative government, equitable public services, and effective public policy.

The consequences would extend far beyond the 2030 Census. Inaccurate population counts and the loss of reliable demographic data would compromise congressional apportionment, distort the allocation of federal funding, impair education planning, hinder the enforcement of federal civil rights laws, and diminish the reliability of the American Community Survey and other federal statistical programs. These harms would be particularly consequential for children and other populations that are already at risk of being undercounted, as well as for the schools, communities, and public institutions that depend on accurate data to determine needs and allocate resources. The resulting gaps in the nation’s statistical foundation could persist for years, affecting decisions long after the census itself has been completed.

The Department has not adequately demonstrated that these sweeping changes are legally permissible, methodologically sound, or likely to improve census accuracy. Nor has it provided the transparent research, rigorous testing, expert consultation, and comprehensive assessment of potential consequences necessary to justify such a significant departure from longstanding census practice. The absence of this evidence is particularly troubling given the scale of the proposed changes and the potential for their effects to extend across multiple federal programs and levels of government.

For these reasons, NEA urges the Department to withdraw the proposed rule in its entirety, including proposed 15 CF parts 60 and 70. Any future revisions to residence criteria or demographic questions should be developed through a transparent, evidence-based process that includes rigorous methodological testing, meaningful consultation with statistical experts and affected communities, and a thorough evaluation of legal, operational, and distributional consequences. The Department should also preserve the flexibility necessary for the Census Bureau to adapt its methods to changing population characteristics and evolving data needs, rather than codifying restrictive requirements that could impede future improvements.

Ultimately, the decennial census must fulfill its fundamental purpose: to count everyone once, only once, in the right place, while providing the reliable demographic information necessary to understand and serve the nation’s population. NEA urges the Department to protect these principles and preserve the census as an essential foundation for representative government, informed policymaking, educational opportunity, and the equitable distribution of public resources.

NEA appreciates the opportunity to submit these comments and urges the Department to withdraw the proposed rule.

Sincerely,

Stacey Pelika, Ph.D.
Director of Research
Education Policy and Implementation Center
National Education Association

After its termination, CSAC members created a non-governmental organization, the Independent Census Scientific Advisory Committee (I-CSAC), see: https://www.censusscientific.org/.
Notice of Termination of 14 Discretionary Federal Advisory Committees, 90 FR 24383, (June 10, 2025), https://www.federalregister.gov/documents/2025/06/10/2025-10398/notice-of-termination-of-14-discretionary-federal-advisory-committees.
The Census Bureau held a consultation with Tribal Leaders, Tribal Nations, and Tribal-serving Organizations on Oct. 2, 2026. However, this consultation was held with only 11 days’ advance notice and no advanced access to consultation questions. The consultation was terminated after two hours with at least 10 people still in the queue. The National Congress of American Indians summarized the meeting as follows: “Tribal leaders and other commenters stated that this meeting could not be considered government-to-government consultation as the notice was insufficient, the consultation was virtual and lasted only two hours, and not all Tribal leaders were able to provide their comments.” See: National Congress of American Indians, Summary: U.S. Census Bureau Tribal Consultation Regarding the NPRM: Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions, https://cdn.sanity.io/files/raa5sn1v/production/ad77c4b503f1dba6ea16f842ac711d52a03ab7ae.pdf.
National Archives, Code of Federal Regulations, https://www.ecfr.gov/ (last accessed on Sept. 29, 2026).
U.S. Census Bureau, 2020 Census Residence Criteria and Residence Situations, https://www.census.gov/programs-surveys/decennial-census/decade/2020/about/residence-rule.html (last accessed on Sept. 29, 2026).
U.S. Census Bureau, U.S. Decennial Census Measurement of Race and Ethnicity Across the Decades: 1790–2020, https://www.census.gov/library/visualizations/interactive/decennial-census-measurement-of-race-and-ethnicity-across-the-decades-1790-2020.html (last accessed on Sept. 29, 2026).
See, e.g., H.J. Res. 20, 101, 263, 356, 484, 71st Cong. (1931); S. 2366, 96th Cong. (1980); H.R. 6769, S. 2366, 96th Cong. (1980), and H.J.Res. 199, S. 358, §601, 101st Cong. (1989).
C.E. Turney, Law Assistant, Senate Legislative Counsel, Power of Congress to Exclude Aliens from Enumeration for Purposes of Apportionment of Representatives (April 30, 1929), reprinted at 71 Cong. Rec. 1821 (May 23, 1929).
Notably, the decennial census counts residents living in parts of the U.S. that do not have access to full voting representation in congress: Washington, DC, and the five U.S. territories (American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the United States Virgin Islands).
8 U.S.C. § 1101(a)(42).
Nicole Foy, We Found That More Than 170 U.S. Citizens Have Been Held by Immigration Agents. They’ve Been Kicked, Dragged and Detained for Days., ProPublica (Oct. 16, 2025), https://www.propublica.org/article/immigration-dhs-american-citizens-arrested-detained-against-will.
Zach Montague and Shawn Hubler, Judge Restricts D.H.S. Field Operations Aimed at Unearthing Voter Fraud, New York Times (Oct. 21, 2026), https://www.nytimes.com/2026/10/21/us/politics/judge-dhs-voter-fraud-elections.html.
Randy Capps, Jennifer Van Hook, and Julia Gelatt, Millions of U.S. Citizens Could Be Excluded under Trump Plan to Remove Unauthorized Immigrants from Census Data, Migration Policy Institute (July 2020), https://www.migrationpolicy.org/commentary/millions-us-citizens-could-be-excluded-under-trump-plan-remove-unauthorized-immigrants.
2020 Census Barriers, Attitudes, and Motivators Study (CBAMS) Focus Group Final Report, U.S. Census Bureau, p. 50-60 (Jan. 24, 2019) https://www2.census.gov/programs-surveys/decennial/2020/program-management/final-analysis-reports/2020-report-cbams-focus-group.pdf; J. David Brown and Misty L. Heggeness (2026). Citizenship question effects on household survey response. Journal of Policy Analysis and Management, 45, e70004. https://doi.org/10.1002/pam.70004.
Migration Policy Institute, Share of All Children in Immigrant vs. U.S.-Born Families, by State, 2000 and 2024, https://www.migrationpolicy.org/data-tool/children-immigrant-us-born-families (last accessed Sept. 30, 2026).
U.S. Census Bureau, Why We Conduct the Decennial Census of Population and Housing, https://www.census.gov/programs-surveys/decennial-census/about/why.html (last accessed on Sept. 30, 2026).
According to the Census Bureau, the idea of “usual residence” “is grounded in the law providing for the first census, the Act of March 1, 1790, expressly specifying that persons be enumerated at their ‘usual place of abode.’” See U.S. Census Bureau, Residence Criteria and Residence Situations for the 2020 Census of the United States, 2020-Census-Residence-Criteria.pdf (last accessed on Sept. 30, 2026).
Census Day has been April 1 since 1930. For a history of Census Days, see: Amelia Kerbis and Gloria Wang Shawber, Why the 2030 Census May Not Live Up to the Previous Decennial, (Oct. 20, 2026), https://handsoffourcens.us/blog/why-the-2030-census-may-not-live-up.html.
Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions, 91 FR 57524, (Sept. 10, 2026), https://www.govinfo.gov/content/pkg/FR-2026-09-10/pdf/2026-18481.pdf.
Ibid.
Amy O’Hara, Beth Jarosz, Daniel Cork. (2026). Tax Records Are Critical Inputs to an Accurate Census.
Failure to Test and Complement Them Could Derail the 2030 Census, Georgetown University Massive Data Institute, https://georgetown.app.box.com/s/mh1kpammw07o6e85un6cr6m1olxnkaze.
James Cilke. (2014). “The Case of the Missing Strangers: What We Know and Don’t Know About Non-Filers.” Paper presented at National Tax Association Annual Conference, Santa Fe, NM. 107th Annual Conference Proceedings, 2014. https://ntanet.org/wp-content/uploads/proceedings/2014/029-cilke-case-missing-strangers-know-don.pdf.
Gloria G. Aldana. (2024). Comparison of Child Reporting in the American Community Survey and Federal Income Tax Returns Based on California Birth Records, U.S. Census Bureau, https://www.census.gov/library/working-papers/2024/adrm/CES-WP-24-55.html.
Josh Kovensky and Layla A. Jones. “Exclusive: Commerce Overruled Treasury Concerns in Push to Narrow Census Count.” Talking Points Memo, Sept. 17, 2026. https://talkingpointsmemo.com/news/exclusive-commerce-overruled-treasury-concerns-in-push-to-narrow-census-count
Mark Mather, Children Are Already Undercounted in the Census. Proposed Changes Could Make It Worse, Population Reference Bureau, (Sept. 30, 2026), https://www.prb.org/news/children-are-already-undercounted-in-the-census-proposed-changes-could-make-it-worse/.
Eric Jensen, Laurel Schwede, Deborah Griffin, and Scott Konicki. (2018). Investigating the 2010 Undercount of Young Children – Analysis of Complex Households. Research and Methodology Directorate, Center for Behavioral Science Methods Research Report Series (Survey Methodology #2018-20), U.S. Census Bureau, http://www.census.gov/content/dam/Census/library/working-papers/2018/adrm/rsm201820.pdf.
For an overview of how students have been counted, see: Amelia Kerbis and Gloria Wang Shawber, Why the 2030 Census May Not Live Up to the Previous Decennial, (Oct. 20, 2026), https://handsoffourcens.us/blog/why-the-2030-census-may-not-live-up.html.
National Redistricting Foundation. Census IRS Rule Threatens College Town Representation, https://redistrictingfoundation.org/policies/census-irs-rule-threatens-college-town-representation/, (last accessed Oct. 22, 2026).
U.S. Census Bureau, 2020 Census Residence Criteria and Residence Situations, https://www.census.gov/programs-surveys/decennial-census/decade/2020/about/residence-rule.html (last accessed on Sept. 29, 2026).
Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions, 91 FR 57524, (Sept. 10, 2026), https://www.govinfo.gov/content/pkg/FR-2026-09-10/pdf/2026-18481.pdf.
For a detailed explanation of problems with the proposed guidance for U.S. military personnel, see Beth Jarosz, Daniel Cork, and James Hower. (2026). Proposed Residence Criteria Suggest Military Personnel May Be Counted in the Wrong Place, Georgetown University Massive Data Institute, https://georgetown.app.box.com/s/89fzeffxlkns32dimpvx3qobcrvkyz1c.
U.S. Census Bureau, Counting Every Voice: Understanding Hard-to-Count and Historically Undercounted Populations, https://www.census.gov/newsroom/blogs/random-samplings/2023/10/understanding-undercounted-populations.html (last accessed Sept. 30, 2026).
Estimate given by former Director of the Census Bureau Robert Santos in Carl Smith, “How Proposed Census Changes Could Affect State and Local Governments,” Governing, Oct. 1, 2026. https://www.governing.com/in-the-weeds/how-proposed-census-changes-could-affect-state-and-local-governments
Justin Gest, “Trump Wants to Decide Who Counts. And Who Doesn’t.” New York Times, Oct. 8, 2026. https://www.nytimes.com/2026/10/08/opinion/trump-census-immigrants-redistricting.html
Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions, 91 FR 57524, (Sept. 10, 2026), https://www.govinfo.gov/content/pkg/FR-2026-09-10/pdf/2026-18481.pdf.
U.S. Census Bureau, Why We Conduct the Decennial Census of Population and Housing, https://www.census.gov/programs-surveys/decennial-census/about/why.html (last accessed Oct. 12, 2026).
Project on Government Oversight, Census Matters: Why an Accurate Count Is Essential to Funding Our Communities, https://www.pogo.org/reports/census-matters-why-an-accurate-count-is-essential-to-funding-our-communities (last accessed on Sept. 30, 2026).
Ceci Villa Ross, 2023, Uses of Decennial Census Programs Data in Federal Funds Distribution: Fiscal Year 2021. https://www.census.gov/library/working-papers/2023/dec/census-data-federal-funds.html
U.S. Census Bureau, Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions: Preliminary Regulatory Impact Analysis, https://www.regulations.gov/document/USBC-2026-0628-0002.
1 Stat. 101 §1 (1790).
For an overview, see: Anna Brown and Ziyao Tian, (Nov. 3, 2025), The Changing Categories the U.S. Census Has Used to Measure Race, Pew Research Center, https://www.pewresearch.org/short-reads/2025/11/03/the-changing-categories-the-us-has-used-to-measure-race/.
U.S. Office of Management and Budget, The 2024 Statistical Policy Directive No. 15, https://spd15revision.gov/content/spd15revision/en/2024-spd15.html (last accessed Oct. 15, 2026).
Danielle Taylor, (July 10, 2020), The Context and Evolution of Data Collection for Same-Sex Married Couple Households, U.S. Census Bureau, https://www.census.gov/library/visualizations/2020/demo/the_context_and_evolution_of_data_collection_for_same_sex_married_couple_households.html.
UCLA Williams Institute, (Sept. 23, 2025), 2020 Census: 57% of Same-Sex Couples in the U.S. Are Married, 22% of Them Are Raising Children, https://williamsinstitute.law.ucla.edu/press/2020-census-press-release/.
Steve Lee, (May 23, 2027), US Sens. Harris, Carper Question Census Bureau on Removal of ‘Sexual Orientation and Gender Identity’ from 2020 Census, San Diego LGBT Weekly, https://lgbtweekly.jeffjungblut.com/2017/05/23/us-sens-harris-carper-question-census-bureau-on-removal-of-sexual-orientation-and-gender-identity-from-2020-census/.
Ibid.
Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions, 91 FR 57524, (Sept. 10, 2026), https://www.govinfo.gov/content/pkg/FR-2026-09-10/pdf/2026-18481.pdf.
For an overview, see: David T. Canon. 1999. Race, Redistricting, and Representation: The Unintended Consequences of Black Majority Districts. Chicago: University of Chicago Press.
Jeanette Y. Ziegenfuss, Casey A. Easterday, Jennifer M. Dinh, Meghan M. JaKa, Thomas E. Kottke, and Marna Canterbury. “Impact of Demographic Survey Questions on Response Rate and Measurement: A Randomized Experiment,” Survey Practice 14(1). https://doi.org/10.29115/SP-2021-0010
U.S. Census Bureau, 2020 Census Barriers, Attitudes, and Motivators Study (CBAMS) Focus Group Final Report, (Jan. 24, 2019), https://www2.census.gov/programs-surveys/decennial/2020/program-management/final-analysis-reports/2020-report-cbams-focus-group.pdf.
U.S. Department of Commerce. (1978). Statistical Policy Working Paper 2: Report on Statistical Disclosure and Disclosure-Avoidance Techniques. https://statspolicy.gov/assets/fcsm/files/docs/spwp2.pdf.
Aubrianna Osorio. Change to Census Data Processing Methods Threatens Health Policy Research and Federal Data Systems, Georgetown University McCourt School of Public Policy, Center for Children and Families, (July 30, 2026), https://ccf.georgetown.edu/2026/07/30/change-to-census-data-processing-methods-threatens-health-policy-research-and-federal-data-systems/.
Department of Commerce, Office of Privacy and Open Government, Disclosure Avoidance for Statistical Products, DAO 216-26, (June 4, 2026), https://www.commerce.gov/opog/disclosure-avoidance-statistical-products.
APC, APDU, COPAFS, ICPSR, and PAA Statement on Commerce’s Disclosure Avoidance for Statistical Products, (June 17, 2026), https://www.populationassociation.org/blogs/paa-web2/2026/06/17/statement-on-commerces-disclosure-avoidance-for-st.
U.S. Census Bureau, What are census blocks?, https://www.census.gov/newsroom/blogs/random-samplings/2011/07/what-are-census-blocks.html (last accessed Oct. 18, 2026).
U.S. Census Bureau. School District Geographic Relationship Files, https://nces.ed.gov/programs/edge/Geographic/RelationshipFiles (last accessed Oct. 18, 2026).
Jerry McKibben, Matthew Cropper, and Zoran Stojakovic, The Importance of Small Area Census Data for School Demographics, Count All Kids, May 21, 2021, https://countallkids.org/resources/the-importance-of-small-area-census-data-for-school-demographics/
Id.
U.S. Government Accountability Office. June 2022. Student Population Has Significantly Diversified, But Many Schools Remain Divided Along Racial, Ethnic, and Economic Lines. https://www.gao.gov/assets/gao-22-104737.pdf
OCR data are collected every other school year rather than annually.
The Segregation Tracking Project 2.0. Help & FAQ. https://edopportunity.org/segregation/faq/#data-source (last accessed Oct. 18, 2026).
Carrie Spector. May 6, 2024. 70 Years after Brown v. Board of Education, New Research Shows Rise in School Segregation, Stanford Graduate School of Education, https://ed.stanford.edu/news/70-years-after-brown-v-board-education-new-research-shows-rise-school-segregation.
Dept. of Commerce v. New York, 588 U.S. 752, 782-785 (2019).
U.S. Census Bureau, Why We Ask Questions About . . . Race, https://www.census.gov/acs/www/about/why-we-ask-each-question/race/ (last accessed on Sept. 28, 2026).
U.S. Census Bureau, Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions: Preliminary Regulatory Impact Analysis, https://www.regulations.gov/document/USBC-2026-0628-0002.
Ibid.
Ibid.

Download the Federal Comment (pdf)

Suggested Results

USBC-2026-0628; Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions

Table 1. Residence Criteria for Living Situations for which People Commonly Request Clarification

Great public schools for every student