What the New Census Updates Mean for Public-Health and Community-Based Organizations
The Office of Management and Budget (OMB) has updated the decennial census form to more accurately capture the demographics of our country. This is great news, but there are always important caveats that come with categorizing people. Community-based organizations (CBOs) and public health organizations can learn from the OMB’s work and from each other when considering equitable practices for collecting demographic data.
An Updated Census for a Contemporary US
This spring, after two years of research involving 35 agencies, over 100 listening sessions, and thousands of US residents, the Office of Management and Budget (OMB) completed its recommendations for updating the 2030 US census form to more accurately capture our country’s racial and ethnic demographics. You can find the OMB’s full report here and their blog brief here.
Among other changes, race and ethnicity have been combined (the separation in prior forms was confusing for Hispanic/Latino* respondents since their heritage did not appear in the Race category) and “Middle Eastern and North African” has been added as a category (people of this heritage were previously encouraged to check “White”); subcategories related to heritage are provided for all race categories; and new language encourages respondents to choose as many options as apply. A person with Dominican and Vietnamese heritage, for instance, might check “Black or African American,” “Hispanic or Latino” and its subcategory “Dominican,” and “Asian” and its subcategory “Vietnamese”. There are also updates and clarifications in definitions and instructions.
These revisions are good news. We live in an inequitable country where race, a nonbiological social construct, greatly affects the experiences and outcomes of a person. The updates mean researchers, advocates, administrators, and policymakers at all levels can act on better data.
For example, maternal health outcomes for Hispanic mothers are on par with maternal outcomes for white mothers, unusual in our country. A breakdown by heritage and birthplace, though, reveals that Latino/Hispanic mothers who were born outside of the U.S. experience significantly better maternal health outcomes than Latino/Hispanic mothers born in the U.S. Those born in the US fare worse than white mothers, and specific country of origin also makes a difference.
This more granular information can be used to influence maternal health policies and other indicators of family health. It can also be used to challenge narratives about level playing fields and to understand the complex nature of structural racism in the U.S.
Categories Are Not Identities
The updates may be good news, but, of course, there are some caveats. There is no getting around that categorization squeezes complex human beings into simple boxes. Because race and ethnicity in the U.S. correlates with inequities, we must name race and ethnicity in order to address those inequities. Yetcategorizing people in this way retains and reinforces the very conditions that lead to those inequities. It’s a quandary.
Mareike Schomerus, a researcher who studies evidence-based policy and the mental models that can shape it, made this simple, powerful point in a talk with the data equity project We All Count: a category is not the same as an identity. People’s identities are self-selected and organic, and are complex, contextual, relational, and dynamic. People’s “categories,” on the other hand, are typically imposed by others—and usually because the subjects lack privilege. People who don’t face discrimination, Schomerus points out, tend to get put in boxes less.
Moreover, once you try to pin down race categories, they become absurdly mutable and meaningless. Various races and ethnicities, grouped in various ways and referenced using various terms, have appeared and disappeared and reappeared on the census since its inception in 1790. Race categories have been used to count people, hide people, protect people, expose people, exclude people, and include people. And how individuals choose to identify their own race changes based historical and social context, aspirations, fears, mindset, and geographic region.
Considering Impact as Well as Intent for Your Organization’s Data Collection
Many CBOs and public health organizations collect demographic information from their clients or patients. While this practice is important for myriad reasons, from grant requirements to service improvements, it may be beneficial for organizations to regularly reflect on the impact their data collection might be having on community members. No matter how good the intentions or useful the results, an unexamined set of questions may be perceived by respondents as tokenizing or intrusive, or set off alarm bells for people with concerns about safety or documentation.
There are a few ways to make sure that you can collect and use demographic information from and for your community more responsibly. First, carefully think through your reasons for collecting the information, and do not collect more personal information than needed. The set of questions in the list below offers an excellent guide for this process. Second, explain clearly to your respondents why you are collecting the information and who will have access to it. And third, bring the results of your collection back to your respondents, ideally in a form specifically intended for them as the audience.
An administrator on the We All Count forum offered this very helpful set of questions organizations can ask themselves while preparing questionnaires for community members:
Why do you want to collect this information?
Is the demographic/identity data what you want to know or is it a proxy for something else?
Are you trying to understand a type of person or a type of lived experience?
Are the benefits of standardizing your system going to be felt equally for people in each of the categories you create?
Are the downsides to standardizing your categories going to be felt equally across the people in your system?
Who does standardizing make this process better for: people working the data, consuming the data products, or the people providing the data?
Data is helpful for moving programs forward, but its collection is not inert. A regular review of organizational data collection practices can yield meaningful improvements for both your data and your community.
* People with Central and South American heritage are called Latino/a/x in the U.S; people whose heritage involves the Spanish language are called Hispanic. Practically, these terms are sometimes used interchangeably. In this article, we use the term “Hispanic/Latino” because the new census uses both terms in the race category (“Hispanic or Latino”).