Mark Mather
Associate Vice President, U.S. Programs
Federal revisions to race categories will give people better options for identifying themselves and provide data users with a more accurate picture of the U.S. population.
April 8, 2024
Associate Vice President, U.S. Programs
Senior Vice President, Programs
On March 28, 2024, the U.S. Office of Management and Budget (OMB) announced federal revisions to the race and ethnicity data collection standards, also known as Statistical Policy Directive No. 15 (SPD 15). In other words, changes are coming to how we collect race and ethnicity information in the United States.
The last time OMB changed federal race/ethnicity standards was in 1997—more than a quarter century ago. These new revisions will improve the accuracy and equity of the data. In particular, many people who identify as Hispanic/Latino or Middle Eastern or North African (MENA) did not see themselves reflected in the standard race/ethnicity categories. U.S. Census Bureau research showed that more than 90% of those who identified as “some other race” identified as Hispanic or Latino on the 2020 Census ethnicity question. Consequently, “some other race” was the second largest race group in the 2020 Census, with nearly 50 million people selecting it.
The new race categories will give people better options for identifying themselves, providing data users with a more accurate picture of the U.S. population.
Census data accuracy is critical. The data guide how more than $2.8 trillion in federal funds are distributed to states, communities, and tribal governments each year; inform a wide range of government, business, and nonprofit decision-making; and provide a population base for dozens of federal surveys.
The new standards will change how data are collected across all U.S. federal agencies by using a combined race/ethnicity question. A single question will contain at least seven race/ethnicity categories, from which individuals can choose as many as they feel apply:
The Census Bureau is also expected to continue use of the “Some other race” category in the decennial census and American Community Survey (ACS), as required by Congress.
For data users, this change will be the biggest to Census Bureau race and ethnicity data collection since the 2000 Census—the first decennial census that allowed people to mark more than one race. But changes to race/ethnicity questions are nothing new; questions about race have been present on every decennial census since 1790 and have changed from decade to decade.
For example, in the 2020 Census, changes in design, data processing, and coding of race/ethnicity questions between 2010 and 2020 have made it difficult to measure population change for certain groups, especially multiracial populations.
The new race/ethnicity standards will have implications for our understanding the U.S. population. For one, the changes will affect the sizes of population subgroups. The Census Bureau’s 2015 National Content Test results suggest that the changes will increase the number of respondents identifying as Middle Eastern or North African, Hispanic or Latino alone, and Black or African American, and will decrease the number of people identifying as ”white” or “some other race.” Combining the race/ethnicity questions is also expected to reduce the number of missing responses to the race question on the decennial census and ACS forms.
These changes could create challenges for those who need to measure trends for racial and ethnic population groups, because the data may not be comparable over time.
However, the impact of the revisions on measuring trends will depend on the size of different population subgroups in a geographic area. For example, more than 3.5 million Americans identified as Middle Eastern or North African in the 2020 Census—the first census to include MENA examples in a write-in response—accounting for 1.5% of the White alone or in combination population. But in Michigan, the MENA population made up 3.9% of the White alone or in combination population in 2020, so the OMB revisions will have a bigger impact in that state.
The Afro-Latino Coalition has expressed concerns about the new standards; they anticipate that fewer Latinos will also identify as Black in the combined race/ethnicity question, thus undercounting this group at the intersection of race and ethnicity.
Other groups have pointed out that these changes will ultimately help us implement more equitable outcomes in our policies. Such outcomes could include a better understanding of disparities in health and health care that point to how resources can be leveraged more effectively.
Expect to see agencies create action plans within the next 12 to 18 months for implementing these changes. The Census Bureau has already announced that they are working on better understanding SPD 15 and its implications, and plan to keep the public updated regularly. Such action plans will address when agencies change their questions, as well as how they bridge the prior race/ethnicity questions with the new ones. In particular, SPD 15 instructs agencies to “maintain records on data processing procedures (such as coding, editing, and imputation practices)” and make this documentation available to data users.
For agencies starting new data collection efforts, OMB expects these changes to take place immediately.
The road to implementation, however, will certainly be bumpy. The agency-by-agency process for enacting this change means there will be inconsistencies across the federal system for some time. OMB, the Census Bureau, and other federal agencies will have to dedicate resources to educating data users about the new standards.
And, for state agencies that follow federal reporting guidelines when collecting race/ethnicity information, implementing these changes could take even longer. It took some states more than a decade to adopt new birth and death certificates that allowed people to report more than one race.
Changing the race/ethnicity standards has been a decades-long process and considerable testing was conducted to ensure the questions produce results more consistent with how people identify themselves today. That said, race and ethnicity are fluid concepts, and the one thing we can predict is that these standards will certainly evolve in the future.