Disability Data Convening: First of Many Needed Steps

On September 30, 2024, disability advocates, researchers, and experts gathered with government officials in Washington, D.C. and online for a much-anticipated meeting titled “Stakeholder Meeting on Disability Data Needs.” The convening was constructive, and the participants agreed that it represented a first step toward a robust, ongoing working relationship between the disability community and government agencies tasked with collecting data on disability. (The meeting’s live stream, agenda, and presentation slides are publicly posted.)

The stakeholder meeting stemmed from the Census Bureau’s October 2023 proposal to change the questions about disability on the annual American Community Survey (ACS), which is an important source of demographic and socioeconomic data on the U.S. population. Since 2008, the ACS has asked about disability using six yes/no questions called the ACS-6. These ask whether a person has serious difficulty with specific functions: hearing, seeing, cognition, walking or climbing stairs, dressing or bathing, and doing errands alone. The Census Bureau proposed changing to a different set of six questions called the Washington Group Short Set, or WG-SS, which also focus on functional limitations, but ask people to grade on a scale how much difficulty they have with each one. 

The Census Bureau’s public notice of the proposed change in the Federal Register took many in the disability community by surprise. They hadn’t been aware of, or involved in discussions about, the possible change. The bureau received more than 12,000 comments in response to the proposal, the vast majority of which opposed the change. Concerns centered on the lack of engagement with the disability community and on estimates that the proposed changes would exacerbate the ACS undercount of the number of people counted as disabled by about 40 percent. In particular, commenters noted that both the ACS-6 and the WG-SS questions disproportionately undercount people with chronic conditions, people who use communication assistance devices, people with psychosocial and cognitive disabilities, and people with other specific types of disabilities. Overwhelmingly, commenters, like The Leadership Conference, asked that the bureau pause its plan to revise the ACS disability questions “until it has engaged in meaningful consultation with disability rights advocates and the disability community directly and has demonstrated that any revised questions produce accurate, useful data for all applicable statutory and policy purposes.”    

The director of the Census Bureau, Robert L. Santos, acknowledged the disability community’s concerns in a February 6, 2024 blog post in which he announced that the questions about disability would remain unchanged for at least the 2025 ACS. He also stated that the bureau and other government agencies would commit to broader stakeholder engagement, starting with convening a meeting with “federal agency disability stakeholders, disability community representatives, data users, researchers and disability advocates” before taking any further action. 

The September 30 meeting was the one promised as a first step in Director Santos’s blog post. The gathering was co-hosted by four government agencies and two non-governmental organizations: the Census Bureau, the Office of Management and Budget (OMB), the National Center for Health Statistics (NCHS), the National Council on Disability (NCD), The Leadership Conference Education Fund, and the Consortium for Constituents with Disabilities’ Health Task Force. Meeta Anand, the senior program director of the census & data equity program at The Leadership Conference and The Education Fund, opened and facilitated the convening. The day included presentations, panel discussions with Q&A sessions, and an hour-long period for comments from in-person and online participants.

Panelists included advocates and researchers — many of them people with disabilities — and experts and officials from the Census Bureau and other government agencies. Their perspectives recognized the multiple definitions of disability, from a medical model that views disability as a physical impairment, to a social model that views disability as a challenge created by barriers in society. Participants and panelists explored how different measures of disability might be needed depending on the purposes for which government and/or researchers use the data. 

Government experts explained their agencies’ structures and the process and criteria for changing the questions on the ACS, including the considerations that led to the proposal to change from the ACS-6 to the WG-SS questions. Nearly everyone agreed that both the current ACS-6 questions and the WG-SS questions tend to undercount people with specific types of disabilities.

Advocates emphasized the importance of the principle “nothing about us without us” — that disabled people must be central to any discussion of how to gather data about people with disabilities or anything else regarding disability policy. As Marissa Ditkowsky of the National Partnership for Women & Families stated, the September 30 meeting must not be simply a “checked box,” but rather an investment in an ongoing working relationship between the government agencies and the disability community. Advocates also emphasized the diversity of people with disabilities and the importance of complete and accurate data to ensure the visibility of disabled people in society and the fair allocation of vital resources. There was interest in further testing to develop a better way to ask about disability and agreement that the disability community must be included in the process of determining that better approach.

Panelists and attendees identified numerous open issues, in addition to the catalyzing question of how disability should be measured on the ACS. These included the need for more transparency about how ACS data are used for resource allocation and other purposes and for more information about which measures of disability are used in the range of surveys and data collections across the federal government. Participants highlighted questions about what additional studies are needed to evaluate alternative measurements, how to better capture data about children with disabilities, whether and how to distinguish between “disability” and “functional limitation” in data collection, and whether data on disability should be treated as “population” rather than “health” data.

The day concluded with several important commitments. Elizabeth Poehler, the chief of the Census Bureau’s Enumeration Methods and Requirements Branch, Decennial Statistical Studies Division, confirmed that the bureau is not considering changing the disability questions on the ACS for the 2025 or 2026 surveys. At the meeting and in an October 28 blog post, Census Bureau Director Santos affirmed that no changes would be made to the disability questions in the 2025 and 2026 ACS, and that “nothing will change without continued engagement with and input from the disability community.” He noted that there is not one simple solution and that questions about how to define and measure disability go beyond the ACS and require a “government-of-the-whole” and “community-of-the-whole” approach. 

As next steps toward fulfilling its commitment to inclusive engagement, the bureau must continue to provide multiple opportunities for engagement with the disabled community that meet various access needs, including town halls, opportunities for public comment, and plain language resources. To build a stronger engagement infrastructure, it should also:

• Form a working group of experts from across the bureau, which should meet regularly with people with disabilities and disability advocates, researchers, and data users, similar to the Young Children Working Group that the bureau formed in 2022;

• Appoint a subject matter expert on disability to serve as a liaison between the bureau and its advisory committees;
• Designate a point person to work with disability stakeholders; and 
• Prioritize hiring people with disabilities. 

In addition, the Census Bureau, other federal statistical agencies, and OMB’s Office of Information and Regulatory Affairs (OIRA) should also consider taking other steps, some of which were discussed in the September 30 meeting. These could include: 

• Conducting an in-depth study of how different federal agencies use disability data. The study could identify the strengths and gaps of current disability data collection and map different agencies’ data needs against currently available data; 
• Establishing a federal statistical policy directive on disability, like the one on race and ethnicity;
• Continued coordination with the Disability Data Interagency Working Group (DDIWG), a National Science and Technology Council group that is tasked with leading the way for the federal government to improve federal disability data;
• Considering reclassifying disability as a demographic rather than a health characteristic; and 
• Expanding the collection of disability statistics through other Census Bureau surveys. 

The September 30 meeting highlighted that effective improvements in disability data collection will depend on deep engagement with the broader disability community. This kind of collaboration will lead to more accurate and inclusive data that will reflect the diversity and needs of the disabled community.