As the federal government prepares to conduct the 2020 Census, critics of the Census Bureau are pushing to make fundamental changes to how it collects its data. This month, Ward Connerly and Mike Gonzalez argued in The Washington Post that the agency should remove questions used to monitor race and ethnicity.
This is a bad idea — based on incorrect information — that would do more harm than good to our country.
First, the Census Bureau’s race and ethnic classifications do not overlook the growing mixed-race population in the United States, as Connerly and Gonzalez suggested. In fact, the government specifically redesigned the 2000 decennial to let Americans more easily self-identify with multiple racial and ethnic groups. This information can be found in any of the countless statistical reports routinely issued by the Census Bureau describing our ever-changing population.
Questions regarding racial and ethnic self-identification have been included in each U.S. Census dating back to the first in 1790. The specific wording of those questions and the level of specificity requested have, of course, been revised significantly over 220 years, reflecting our evolving understanding and respect for the cultural diversity of our nation.
Regardless, racial and ethnic self-identification is an essential component of the identities of millions of Americans. It is a valuable proxy indicator of their life experiences, and researchers have found that race and ethnicity are consistently associated with numerous measures of social well-being.
This information is also routinely used to expose politically motivated attempts to gerrymander congressional districts. This month’s court order to redraw North Carolina districts, which appear to have been designed in part to limit the representation of minority groups, is the most recent example illustrating the importance of objective, nonpartisan statistical information to ensure equal protection of the rights of all Americans. Removing this information from the decennial census would make it easier for us to ignore the social discrimination, health and economic disparities that persist in our nation.
Of course, social researchers recognize the imperfect measurement of race, ethnicity and most of the other social constructs that we study. There are many legitimate criticisms of the existing measures, and ongoing efforts in government, academia and private enterprise continue to develop more rigorous and useful measures. The Census Bureau has historically served as a leader in these efforts, typically making changes to census questions only after lengthy periods of careful research, experimentation and public comment. Questions regarding racial and ethnic self-identification will undoubtedly continue to evolve. Those changes will, we can hope, be made in the name of public service and based on nonpartisan research.
The 2020 Census is already under considerable stress due to the cancellation of large-scale pretests and other essential development work. This is no time to impose untested changes to the questionnaire.
TIMOTHY P. JOHNSON AND ROGER TOURANGEAU, WASHINGTON
Editor’s note: Timothy P. Johnson is director of the Survey Research Laboratory at the University of Illinois at Chicago and president of the American Association for Public Opinion Research. Roger Tourangeau is former president of the American Association for Public Opinion Research, and vice president and senior statistical fellow at Westat. They wrote this for The Washington Post.