Some Critical Thoughts on the Census Bureau’s Proposals to Change the Race and Hispanic Questions

January 11, 2013

By Nancy Lopez

As a sociologist of racial, ethnic and gender stratification, I applaud the Census Bureau’s ongoing efforts to examine how we can collect race and ethnicity data that address our increasingly complex and changing demographics for generations to come. Among the key recommendations of their 2010 Alternative Questionnaire Experiment (AQE) Report is a call for further testing of the combined race and Hispanic origin question format.

Accordingly, the Census will continue testing questionnaire formats that include Hispanic as a racial category (the first and only time that a specific Hispanic origin group was included in the U.S. Census was in 1930 when “Mexican” was included as a racial group). Including Hispanic as a racial category is a significant departure from current Office of Management and Budget (OMB) guidelines that require that Hispanic Origin (ethnicity) is asked as a separate question from Race (racial status). It is important to note that since 2000, individuals may mark one or more race (but only one Hispanic ethnicity).

While the Census engages in further testing and refinement of questionnaire formats for race and ethnicity data collection, it is important that we consider why we collect and analyze race and ethnicity data in the first place: the focus is to assess our progress in Civil Rights enforcement. Data collection on race and ethnicity is used by federal, state and local agencies to monitor discrimination and segregation in housing (Fair Housing Act), labor market participation (Equal Employment Opportunity Commission), political participation (Voting Rights Act, Redistricting), educational attainment (Department of Education), health (Centers for Disease Control and Prevention), and criminal Justice (Department of Justice), among other policy areas.

If we agree that the key purpose of data collection on race and ethnicity is for monitoring our progress in creating a more perfect union for all, then we should consider several questions:

• To what extent is one’s ethnicity, cultural background, national origin, generational status, and ancestry conceptually interchangeable with one’s race or racial status as a social position in society?
• What is the value-added of a given questionnaire format?
• What is lost or improved by keeping or changing the current two-question format?
• Do these data allow us to monitor patterns of inequality among entire categories of people by race and ethnicity?

The “gold standard” for all racial and ethnic data collection should be meaningful use for interrogating inequalities across a variety of social outcomes.

As a native Spanish-speaker who was born and raised in New York City’s public housing projects, I am viscerally aware of the distinction between race and ethnicity. Although I share the same ethnic background of my immigrants from the Dominican Republic, my father, who is light skinned and not of discernible African phenotype, occupies a different racial status than my mother and I who share a common racial status as Black women.

The distinction between race and ethnicity is not trivial. A growing number of scholars have found that distinguishing race from ethnicity is extremely important for monitoring and ameliorating inequalities in housing segregation (Massey and Denton; Logan); health (LaVeist-Ramos et al.); education (Murguia and Telles); criminal justice (Steffensmeier and Demuth); and employment (Rodriquez et al.), etc. If we collapse race and ethnicity as interchangeable concepts, we may miss the opportunity to examine whether there are unique experiences among co-ethnics that may occupy very different racial statuses.

While it is true that the combined questionnaire formats being tested by the Census instruct individuals to mark one or more race and write in an ethnic designation if they desire, it will be challenging to capture the two concepts with one question. Indeed, the AQE report found that while the total number of Hispanics was not reduced by the combined questionnaire formats, the detailed information on national origin groups did decrease. In other words, one of the potential shortcomings of the combined format is that, although it will not necessarily reduce the number of Hispanics that are counted, we will have less information about their racial status and national origin.

In the end, if we depart from the premise that race and ethnicity are two analytically distinct concepts then we will require two different questions. Currently, only the American Community Survey (ACS) includes a question about ancestry. The value added by the extra “real estate” of a having two separate questions on race and ethnicity, not only for Hispanics but also for other demographic groups, surely outweighs the costs of having poor data that will hinder our ability to assess distinct pathways of inequality.

Another value-added question that the Census could pilot is bringing back the parental place of birth question. Given our changing demographics and diverse immigration streams, the collection of these data would allow us to examine patterns of inequality related to national origin and immigrant status (e.g., first generation-immigrants who came as a adults vs. second generation-U.S.-born children of immigrants). The last time that this value-added question was included in the Census was in 1980. These data would allow us to map whether there are unique inequalities faced by foreign-born individuals and their children by national origin.

Again, I applaud the Census for proceeding with caution before recommending any major changes in national data collection systems that will shape how we assess Civil Rights enforcement for generations to come. While I have focused on the experiences of Latinos, the analytical distinction between race and ethnicity is also important for other groups that have experienced historic and ongoing discrimination, including Native Americans, Blacks, Asians and Middle Eastern communities, etc.

It is my hope that the Census also pursues further testing of what is lost and/or gained in terms of interrogating inequalities by keeping Hispanic ethnicity and race or racial status as separate questions. In particular, I am optimistic about the strategic partnerships between the Census, OMB and diverse Civil Rights organizations, scholars, researchers and communities working towards creating a more perfect union for all.

Nancy López is an associate professor of sociology at the University of New Mexico, where she co-founded and directs the Institute for the Study of “Race” & Social Justice housed in the RWJF Center for Health Policy. She can be reached at

Be Sociable, Share!
, , , , ,


No comments yet.

Leave a Reply