Comments Re: Proposals from the Federal Interagency Working Group for Revision of the Standards for Maintaining, Collecting, and Presenting Federal Data on Race and Ethnicity (82 FR 12242)
View a PDF of these comments online here.
Recipient: Jennifer Park, Office of the U.S. Chief Statistician
Dear Ms. Park:
On behalf of The Leadership Conference on Civil and Human Rights and the 80 undersigned organizations, we appreciate this opportunity to provide comments in response to proposals from the Federal Interagency Working Group (“Working Group”) for Revision of the Standards for Maintaining, Collecting, and Presenting Federal Data on Race and Ethnicity (“Standards”), published in the Federal Register on March 1, 2017 (the “Notice”). The Leadership Conference is a coalition charged by its diverse membership of more than 200 national organizations to promote and protect the civil and human rights of all persons in the United States.
The Leadership Conference provides a powerful unified voice for the many constituencies of the coalition: persons of color, women, children, individuals with disabilities, LGBTQ individuals, older Americans, labor unions, major religious groups, civil libertarians, and human rights organizations. Our coalition views an accurate and fair census, and the collection of useful, objective data about our nation’s people, housing, economy, and communities generally, to be among the most important civil rights issues of our day. We have a long record of first-hand experience working in support of previous censuses. For the 2010 Census, we undertook the most comprehensive and extensive effort by a stakeholder organization to promote participation in historically hard-to-count communities and to mobilize local advocates in support of the census by highlighting the civil rights and social justice implications and community benefits of an accurate count. In addition, The Leadership Conference conducts advocacy on various data-related issues, such as sufficient funding for the American Community Survey (ACS).
The Leadership Conference commends OMB for recognizing that continued racial and ethnic change in the United States requires an evolution in the statistical policy governing how we measure the composition of our population. That policy must strike a balance between the compelling individual interest in identifying oneself and society’s interest in ensuring compliance with laws that uphold the civil and constitutional rights of all people. In fact, OMB highlights these primary goals in the historical basis for the Standards: “Development of these Federal data standards stemmed in large measure from new responsibilities to enforce civil rights laws. Data were needed to monitor equal access to housing, education, employment opportunities, etc. for population groups that historically had experienced discrimination and differential treatment because of their race or ethnicity.”[i]
Since the last significant revisions to the OMB Standards 20 years ago, data collection methods have evolved, offering respondents more options for providing race and national origin or ethnicity data. Notably, digital response platforms offer greater flexibility to federal agencies and federal grant recipients in eliciting detailed responses. It is important to maintain consistency, to the extent operationally feasible, in the choices offered respondents using different modes of participation, including Internet, telephone, and paper questionnaire. However, the Standards should recognize the promise of technological advances as more respondents move to on-line participation in surveys and censuses and survey staff use electronic devices to collect information, by setting high expectations for collecting and presenting richer data when newer collection methods allow for such improvements over traditional methods.
I. Issues for Comment in the Federal Register Notice (“Notice”)
As a coalition representing a broad, diverse set of communities, The Leadership Conference is best positioned to offer a civil rights perspective on themes that are common to the issues presented for comment in the Federal Register notice. We also offer feedback on proposals or issues for which there is consensus within the coalition, and suggest issues that we believe require further consideration and development by OMB.
As an overarching matter, we first address one prevalent theme in the Working Group interim report and OMB Notice: the “cost” of revising the Standards. While cost is a valid factor for OMB, as a steward of federal dollars, to consider, The Leadership Conference believes that the nation must have the tools necessary to uphold and advance constitutional guarantees of equal rights and effectively enforce those rights through relevant laws. The “cost” of discrimination in major social institutions such as the labor force, health care system, education, housing, and financial services — both to individuals who are denied equal access and opportunities and to society as a whole — is significant, long term and pervasive, and outweighs the cost to federal agencies of adapting their statistical practices to reflect the nation’s demographic change and provide the tools for effective, rigorous administration of civil rights laws.
In addition, the availability of data that more accurately and comprehensively describes the population of diverse communities across the county will help policymakers at all levels of government make far more informed, and therefore fiscally prudent decisions, about how to spend limited public resources most effectively, to improve opportunities for all Americans in a way that strengthens our economy and social institutions as a whole. OMB can take steps to mitigate the burden of complying with new Standards by working with agencies to promote common, consistent data collection instruments and methods whenever practicable and setting reasonable timetables for compliance, taking into account the complexity and cost of various collection activities.
There also are cost savings associated with proposed changes to the Standards that are more easily quantifiable. For example, the Census Bureau’s 2015 National Content Test (NCT) found significant decreases in reporting of Some Other Race (SOR), as well as in item nonresponse, when race and ethnicity data are collected through a combined question format, as compared to the separate question format favored in the current OMB Standards. The improved response patterns will reduce the costs associated with reassigning SOR and missing responses to a distinct race category, while also mitigating broader costs, in policy terms, by improving data quality.
A. Questionnaire Format and Nonresponse
1. Separate versus combined question format
The Leadership Conference supports revising the Standards to favor the collection of race and ethnicity data through a combined question format, instead of through separate questions for ethnicity and race, as long as such a format allows respondents to select more than one race and ethnicity, as well as more than one subgroup for each race and ethnicity. However, use of a combined question should not imply that there are no distinctions between race and ethnicity. The OMB Standards should maintain categories for race and ethnicity, while acknowledging that some respondents only identify with an ethnicity (that is, Hispanic and possibly MENA) and not with one of the distinct race categories. Furthermore, the Guidance accompanying the Standards should direct federal agencies to make clear to respondents that the categories offered reflect both races and ethnicities and that they can identify with as many of those categories as they wish. (The Census Bureau’s research and testing of question wording and terminology associated with maximum, accurate responses in a combined questions should be illustrative for other agencies and organizations. We urge the Census Bureau to continue testing optimal wording and terminology for the 2020 Census and ACS.)
In our comments in response to the first Federal Register notice announcing OMB’s intention to review the Standards and setting the scope of the review, The Leadership Conference noted that it was not able to support a preference for a combined question format without a full analysis of whether such an approach would yield data of comparable or better quality than data collected through separate questions for all race and ethnic groups and subgroups. We also expressed concern about the potential loss of race data through a combined question format. Since that time, the Census Bureau released a full analysis of the 2015 National Content Test (NCT) results.[ii] The analysis showed that a greater proportion of respondents identified with one or more distinct race or ethnicity categories in a combined question than in separate questions, as evidenced by a significant drop in Some Other Race reporting to about one percent, in a combined question with detailed checkboxes. Further, there were no statistically meaningful differences in reporting for the Black, Asian, and Native Hawaiian and Pacific Islander race categories in a combined question approach. As expected, reporting in the non-Hispanic White alone category dropped significantly with combined question formats, as compared to separate race and ethnicity questions. Similarly, more than 70 percent of respondents identifying as Hispanics in a combined question format did not select another major race or ethnicity (i.e. MENA) category, suggesting that a significant segment of the Hispanic population identifies primarily with this designation. We emphasize again, however, that data collection instruments should make clear that respondents might identify with and select more than one category when reporting their race and/or ethnicity.
We also believe that federal agencies should use consistent methods of data collection to the fullest extent possible. In its interim report, the Working Group expressed concern about the ability of some agencies to transition from a separate to combined question format, especially if the agencies routinely rely on administrative records to generate race and ethnicity statistics. The Working Group suggests that the Standards conceivably could allow a variety of approaches to collecting the data, so that not all agencies would be required to change question format. We are concerned that inconsistent practices could lead to data that are not comparable in quality or measurement, and we discourage a standard that does not clearly favor one format over another when data are self-reported.
2. Factors to consider in relation to information quality
All federal race and ethnicity data collection activities are important and even vital to monitoring and enforcing civil rights laws and advancing equality of access and opportunity for all Americans. We also recognize that many agencies rely on a range of data “building blocks” — including federal, state, and even local administrative records, and data generated by businesses and organizations — to meet their obligations under civil rights and other laws, or simply to better understand their own practices and to advance diversity goals and parity in outcomes. Many of the policy revisions under consideration will, directly or indirectly, affect the way those institutions maintain and collect race and ethnicity data.
Nevertheless, we believe the quality of data collected in the decennial census is of paramount importance for several reasons, and that data collection and reporting Standards that help ensure the highest quality census data should be given significant, if not primary, weight.
- First, the census provides a benchmark for all other statistical data the federal government and many other public and private entities collect. Census data provide a frame for surveys and are used to weight data produced through a variety of methods, including surveys and the compilation of administrative data. Given the foundational nature of census data, OMB should give significant weight to statistical protocols and practices that allow the decennial census to gather the highest quality and most comprehensive information about race and ethnicity possible.
- Second, the decennial census is a constitutionally required statistical undertaking, conducted for the purpose of implementing a democratic system of governance based on “we, the people” — that is, the population. Census data are the basis for congressional apportionment and for the guarantee of equal representation in the redistricting process at all levels of government under the Fourteenth Amendment. The government seeks to uphold that constitutional right through the Voting Rights Act of 1965, as amended; accurate, detailed race and ethnicity data for small geographic areas are essential tools for enforcing the provisions of the VRA.
- Third, the census is the only statistical activity that reaches every person and every household in the nation. Thus, it is the only opportunity to measure and truly understand the full depth of racial and ethnic diversity in the country, providing a framework for all other data collection, as well as important context for policy debates and decision-making that bear on constitutional rights and also seek to improve the well-being and future of Americans in all of the nation’s communities.
- Finally, the census and related planning activities can serve as important laboratories for collection methodologies and statistical policy approaches that will benefit and inform the work of other federal agencies that collect race and ethnicity data.
Undoubtedly, compliance with a new set of Standards will be more difficult, time-consuming, and/or costly for some entities than others. Feasibility of implementation is important, but we also must recognize that there always will be some resistance and barriers to change. Inconsistency in data collection protocols and data products, and gaps in the availability of comprehensive race and ethnicity data, already exists. Ensuring that the Standards for maintaining, collecting, and presenting these data promote scientifically rigorous measurement of emerging demographic and social changes must be a paramount consideration as OMB finalizes revisions to the Standards.
B. Classification of Middle Eastern or North African Race/Ethnicity
The Leadership Conference supports the addition of a new ethnic category for persons of Middle Eastern and North African (MENA) origin that is geographically based and separate from the White or Caucasian category. We further urge OMB to include MENA as a minimum reporting category for all purposes for which federal agencies collect and publish race and ethnicity data, as long as the data are statistically reliable and can be reported in compliance with data confidentiality standards. We urge OMB to work closely with advocates for all relevant communities to achieve a consensus on (1) the nationalities and transnational groups that should be included in the definition of a MENA category; (2) a determination on whether MENA should be considered a race or ethnicity; and (3) effective instructions to respondents to promote reporting in this category among all persons of MENA origin, whether native or foreign born.
Evaluation of the Census Bureau’s 2015NCT showed that when the Census Bureau offered a distinct MENA category in a combined race and ethnicity question, a significant percentage of respondents (79 percent) identifying with MENA origins selected that category. At the same time, reporting of detailed MENA responses in the other major categories declined significantly. Not surprisingly, the drop was most notable in the White category, where the percent of responses by people identifying with a detailed MENA origin fell from 85.5 percent when no distinct MENA category was offered, to 20 percent when a MENA checkbox was included in a combined question.
These results clearly indicate that respondents of MENA origin or ancestry overwhelmingly do not view themselves as “White” (or, since multiple responses are allowed, only White), which the current Standards contemplate. Because the Standards favor self-identification in race and ethnicity reporting whenever feasible, it is important that the categories that define our country’s racial and ethnic composition reflect the way people see themselves to the best extent possible. This principle becomes more important when the ethnicities in question are associated with discrimination or are singled out in public policy for heightened scrutiny.
In fact, it is essential to have an accurate portrait of communities that have been targets of racial or ethnic discrimination and hate crimes, and that often do not have equal access to social and economic institutions and activities. The first step toward addressing issues of inequality — whether in access to health care, job and contracting opportunities, lending, and affordable housing, in interactions with law enforcement, or in acceptance by the public generally — is to understand objectively the nature and scope of the challenges. To this end, we cannot know what we do not measure accurately. Collecting data that clearly and fully identifies Americans of MENA origin will help government, civic, faith, and private sector leaders meet the needs of these communities and ensure equality of opportunity in all aspects of American society.
C. Additional Minimum Reporting Categories
We believe the Standards should state more clearly and directly that the major race and ethnicity categories represent the minimum reporting categories agencies must use if they collect this information through self-reporting methods. Furthermore, recognizing that the “U.S. population has continued to become more racially and ethnically diverse,”[iii] we believe OMB should require, rather than simply encourage, agencies to collect and report more detailed data whenever possible.
Within the primary race and ethnicity categories, people of different origins (or subgroups) often have vastly different experiences and outcomes, as evidenced by key socio-economic indicators published for detailed subgroups, including data on educational attainment, incidence of disease and access to health care, employment, income, and other important measures of well-being. Identifying and understanding these differences allows policymakers and civic leaders to fashion remedies for disparate outcomes that address root causes and factors, such as immigration status, language barriers, cultural behaviors, and geographic isolation. Collecting and reporting data only on broader race or ethnicity categories masks important differences within these population groups and hinders efforts to improve circumstances for all Americans.
The current Standards suggest that agencies are permitted to collect more detailed information, but many federal agencies have not taken steps to do so. Without a stronger statement of federal goals such a requirement would represent, we are unlikely to see improvement in the scope of information available for important policy and programmatic purposes. The Standards should establish a preference for collection of subgroup, or disaggregated, data whenever feasible and require agencies to seek an exception to this directive and explain their rationale based on methodological, operational, or cost barriers.
Our support for stronger guidelines on the collection of detailed race and ethnicity data is compatible with a possible change in preferred question format for collecting these data, from separate questions for race and ethnicity, to a combined question format. The 2015 NCT results demonstrated that using a combined question format with checkboxes that allow respondents to identify a subgroup easily yielded a level of detailed reporting comparable to or better than similar reporting when separate race and ethnicity questions are used.
With respect to setting minimum categories for the collection of detailed data, the Standards should recognize that the composition of the population continues to change overall and that the prevalence of certain races, national origins or ethnicities in some geographic areas but not others requires flexibility in determining the most useful combination of subgroups for which agencies might collect data, depending on the size and scope of the collection activity. We urge OMB to consult with representatives for each population group to determine whether the Standards should specify which subgroups require the collection of more detailed data, whenever feasible. As a general matter, we believe the Standards should recognize that the composition of the population will continue to change and that periodic review of the collection of detailed data is warranted.
We do not believe that requiring the collection of detailed race and ethnicity data would impose an additional response burden on the public. Respondents are free to identify with as few or as many race and ethnicity categories as they wish, and many will welcome the opportunity to choose either a specific national origin or ethnicity or more than one category. Some respondents, in fact, may not identify strongly with a major race or ethnicity category (such as Asian or Hispanic), but might instead “see” themselves more clearly as a member of a subgroup (such as Chinese or Mexican).
D. Relevance of Terminology
1. Adding South or Central American subgroups to the description of the American Indian or Alaska Native (AIAN) category
The Leadership Conference notes that advocates for AIAN communities and South and Central American persons in the United States have, for some time, urged OMB to research and address the most appropriate way to classify indigenous persons, in order to preserve and improve the quality of data collected on American Indians and Alaska Natives, while recognizing the need to measure other indigenous population groups accurately. We hope that OMB will take seriously the Interagency Working Group’s recommendation for further study of ways to encourage better reporting of South and Central American subgroups, and we urge OMB to pursue that research, in consultation with the affected communities, in a timely way.
2. Removal of the terms “Negro” and “Far East”
The Leadership Conference agrees with the Working Group proposals to remove the terms “Negro” and “Far East” from the Standards; the changes are, in fact, long overdue. Both terms are outdated and rarely used in current public discourse. The terms also are offensive to some people in these communities. We note that The Leadership Conference joined the National Urban League and the National Coalition on Black Civic Participation in issuing a statement at the start of the 2010 Census, urging the Census Bureau to drop the term “Negro” in the race question in all future data collection activities.
3. Continued relevance, and definition of, “principal minority race” in reporting data
The Leadership Conference recognizes the historical significance of designating Black or African Americans as a “principal minority race” for the purpose of presenting data on race and ethnicity. African Americans have been the targets of individual and systemic discrimination and violence since before the nation’s founding, and this discrimination was embedded in the Constitution, creating a legacy of inequality that the nation still struggles to address effectively.
We note, however, that since OMB first promulgated standards on the classification of race and ethnicity (for statistical purposes) 40 years ago, and even since OMB last revised the Standards in a significant way 20 years ago, the U.S. has experienced enormous demographic change, which itself has contributed to an alarming increase in racial and ethnic discrimination and hate crimes against people of color. Persons of Hispanic origin now comprise the second largest race or ethnic population group in the country (nearly one-fifth of the population). Asian Americans are now the fastest growing race group, with 43 percent growth between the 2000 and 2010 Censuses. The Native Hawaiian and Pacific Islander (NHPI) population grew three times faster than the U.S. population as a whole during that decade. Hispanic origin, Asian, and NHPI status often makes these populations vulnerable to discrimination and violence based on perceived immigration status, which regrettably is on the rise. In addition, the American Indian and Alaska Native population continues to urbanize at a fast pace (30 percent growth from 2000 to 2010),[iv] increasing their proportion of the population base in some communities.
Given these trends, the term “minority” as a descriptor of all race or ethnicity categories other than White is no longer accurate or relevant in a growing number of states and localities where non-White race or ethnicities — individually, collectively, or in some combination — comprise a majority of the population. Equally pertinent, some people believe that the term “minority” implies a less important or less equal status of persons within a non-White race or ethnicity category.
Accordingly, we urge OMB to revise the Standards to favor more clearly the presentation of data for all standard (major) race and ethnicity categories whenever statistically possible. We believe the designation of an “All Other Races” category is no longer appropriate, as it renders many people “invisible” for important policy and programmatic purposes. Instead, the Standards should direct agencies that are unable to report data for specific race or ethnicity categories due to unacceptable quality or statistical reliability, confidentiality concerns, or absence of legal requirements, to note clearly (with the use of an asterisk or footnote, for example) the reason(s) why data are not included for those groups.
II. Topics for Additional Consideration
A. Guidance for Comparing Data Over Time
The Leadership Conference respectfully urges OMB to develop clear guidance for federal agencies and all other data users on scientifically sound, consistent protocols for comparing race and ethnicity data across time and categories. Implementation, monitoring, and enforcement of many civil rights laws, in particular, require analysis of racial and ethnic trends with respect to access and opportunity. As it did after revising the Standards in 1997, OMB should work with a wide range of stakeholders — but especially those involved in implementation, enforcement, and monitoring of civil rights laws, as well as with the redistricting process — to develop guidance for tabulation and “bridging” between data collected under different versions of the Standards.
“Bridging” data collected under the current Standards and, in the future, under revised Standards (if OMB chooses to update the Standards) will be particularly challenging not only because the Standards might include a new reporting category (i.e. MENA), but also because federal agencies might collect race and ethnicity data through a combined question that does not distinguish between these concepts and does not require the respondent to select both a race and ethnicity. Nevertheless, OMB must ensure that data collected in accordance with the Standards will be useful in carrying out laws and programs, especially in the civil rights arena. The data must allow agencies to monitor compliance with civil rights laws, which in many cases requires comparisons of past practices with current ones, as well as the establishment of benchmarks and identification of statistically significant changes in the data.
B. Process for Promulgating Final Standards
Finally, we want to express our concern about OMB’s intention not to seek further public comments on the final Standards it plans to issue later this year. We are mindful of the depth and scope of difficult issues OMB and the Working Group must evaluate and resolve before finalizing any revisions to the Standards, as well as the Census Bureau’s immutable timetable for 2018 End-to-End Census Test preparation and submission of 2020 Census and ACS questions to Congress by April 1, 2018. We also appreciate OMB’s earlier decision to modify its initial schedule for review and revision of the Standards, to allow for full public vetting of the 2015 NCT results and subsequent opportunity for thoughtful public comment on proposed changes to the way federal agencies collect and report race and ethnicity data.
However, the Interagency Working Group’s Interim Report on review of the Standards, as well as OMB’s Federal Register notice, make it clear that both entities must address and resolve many complex issues, and weigh many factors, before making final decisions on modifications to this statistical policy. The significance of the policy to the implementation and enforcement of civil rights laws, which themselves are designed to uphold important constitutional rights, makes broad public acceptance of the final Standards vital. Therefore, we strongly urge OMB to provide an opportunity for comment on proposed revisions to the Standards before the updated policy is finalized.
Further, we hope that OMB will continue to reach out to stakeholders and data users in the interim months, as it considers feedback in response to the current Notice and works to resolve and develop feasible guidelines on several key, outstanding issues that will inform our understanding of the nation’s diverse and ever-changing population for many years to come. Those issues include the definition for a new MENA category (should OMB decide to add this category) and the extent to which federal agencies will be required to collect detailed data whenever feasible. Continued dialogue, as OMB reviews the comments submitted in response to the current Notice, will help OMB reach conclusions that have broad support from a wide range of stakeholders, including Congress, federal agencies, state and local civic leaders and institutions, civil rights advocates and experts, researchers, and business leaders.
We look forward to working with you and your colleagues in the coming months, to ensure that the voices of the civil and human rights community continue to be heard in this important, ongoing national conversation. If you have any questions about these comments, please contact Corrine Yu, Managing Policy Director, at 202-466-5670 or [email protected]
The Leadership Conference on Civil and Human Rights
9to5, National Association of Working Women
Advancement Project California
Alliance for Strong Families and Communities
American Anthropological Association
American Association of University Women (AAUW)
American Federation of State, County & Municipal Employees (AFSCME)
American-Arab Anti-Discrimination Committee
Asian & Pacific Islander American Health Forum
Asian Americans Advancing Justice – AAJC
Asian and Pacific Islander American Vote (APIAVote)
Association of Asian Pacific Community Health Organizations (AAPCHO)
The Center for Asian Pacific American Women
Center for Law and Social Policy (CLASP)
Disability Rights Education & Defense Fund
Equal Justice Society
Equal Rights Advocates
Family Equality Council
Hispanic Association of Colleges and Universities
Human Rights Advocates
Human Rights Campaign
Institute for Science and Human Values
Justice in Aging
Labor Council for Latin American Advancement
Laotian American National Alliance
Latino Census Network
Lawyers’ Committee for Civil Rights Under Law
League of United Latin American Citizens
League of Women Voters of the United States
Legal Aid at Work
NAACP Legal Defense and Educational Fund, Inc.
National Association of Latino Elected and Appointed Officials (NALEO) Educational Fund.
National Association of Social Workers
National Black Justice Coalition
National Center for Lesbian Rights
National Center for Transgender Equality
National Coalition for Asian Pacific American Community Development
National Congress of American Indians
National Council of Asian Pacific Americans (NCAPA)
National Council of Asian Pacific Islander Physicians
National Council of Jewish Women
National Council of La Raza
National Education Association
National Employment Lawyers Association
National Health Law Program
National Hispanic Media Coalition
National Hispanic Medical Association
National Immigration Law Center
National Indian Education Association
National Institute for Latino Policy (NiLP)
National Latina Institute for Reproductive Health
National LGBTQ Task Force
National Network for Arab American Communities
National Partnership for Women & Families
National Women’s Law Center
NETWORK Lobby for Catholic Social Justice
OCA – Asian Pacific American Advocates
OCA Greater Houston
Partnership for America’s Children
Pride at Work
Prison Policy Initiative
United Church of Christ, OC Inc.
The Voter Participation Center
Women’s Intercultural Network
Women’s Voices Women Vote Action Fund
[i] Federal Register, Vol. 82, No. 39, March 1, 2017, p. 12243.
[ii] 2015 National Content Test: Race and Ethnicity Analysis Report, Version 1.0, U.S. Census Bureau, February 28, 2017.
[iii] Ibid. p. 12244.