“Accurate Racial/Ethnic Data Should Drive Category Review” by Raul Yzaguirre and Sonia M. Perez (January-February 1995 P&R Issue)

But why is it important to know such data? Why should research and advocacy organizations, or those that represent community-based agencies, like my own, care about how the Census and other government agencies collect these data? How do proposed changes to racial and ethnic categories improve or reduce the quality and accuracy of data about significant segments of the population. like Latinos? And how can government forms best capture not only demographic characteristics about a population, but also cultural or ethnic identity?

As the only official source of detailed data on the U.S. population, Census data serve many important purposes. From the perspective of the U.S. Latino community -the only ethnic group for whom data are independently compiled —such data confirm the role we are playing in the workforce, in schools and in society as a whole. These data also allow researchers, schools, states and municipalities,and others to assess and analyze the social, health and economic status of specific population groups. For instance, we know through the collection of Census data that the poverty rates for Latinos have steadily increased since 1980, even when Black and White poverty rates experienced declines. From a research and pubiic policy perspective, we count on these data to determine how policy makers and other officials can more effectively address the issues that affect their constituencies. Schools, industries and employers require these data for planning purposes in order to match their needs to changing population characteristics.

And, of course, these data are crucial to the racial and ethnic populations themselves. As frequent users of these data, we at the National Council of La Raza. as well as our community-based affiliates, rely on timely and accurate information about U.S. Latinos to enhance our research, advocacy and ability to adequarelx provide needed serxices. Without such data, NCLR affiliates would be poorly positioned to either design and implement needed programs or offer services to their Latino communities. In a political climate which suggests that individuals will have to rely more on private “charities” and less on public policy interxeniions. such data are critical. For NCLR. knowledge of Latino socio-economic status through these specific data allows us to focus our advocacy efforts on what is most needed to improve the status of Hispanics.

Accurate racial and ethnic data are also necessary for protection of our civil rights. Decades ago, the nation understood that in order to remedy social and economic inequality, and ensure the rights of groups that had been historically discriminated against, we needed civil rights protections —which can only be implemented and monitored with the collection of such data. Many civil rights laws rely on race and ethnic data to ensure equitable compliance and effective enforcement in key areas, including education, voting, employment and housing.

But we and other researchers and advocates recognize that the existing racial and ethnic categories are narrow and limited. With respect to Hispanics, results from the 1990 Census show that the Census questions on race and Hispanic origin are problematic for several reasons. First, Hispanics are a multi-racial population and race and ethnicity are not easily recognized or distinguish-able by observable characteristics. Observers who identify Hispanics based on physical characteristics find that such identification may differ from the way the person identifies him/herself.

Second, many Hispanics, especially those who are not U.S.-mainland born, have difficulty classifying themselves by race. For many Hispanic subgroups, particularly those from Central and South America and the Caribbean, choosing one race category is problematic. since they self-identify principally by subgroup (i.e., ethnicity) descend from multiracial origins. Moreover, in the case of Central and South Americans who may also represent a number of indigenous Indian groups, the classification “Indian” may be selected. A related issue concerns Puerto Ricans who, when they live in Puerto Rico, are not asked about their race; yet, if they move to the U.S. mainland, are required to answer a race question on the Census form.

Third, the category “Other Race” is primarily checked by Hispanics, who must then be reclassified by the Census Bureau. According to the General Accounting Office, fully 98% of respondents who classified themselves as “Other Race” in the 1990 Census were Hispanic. In addition, research suggests that Hispanics and other population groups who may not identify with any of the options offered may have simply skipped or ignored the question. If follow-up was not done for that household or if the allocation procedure was used- the practice of assigning a characteristic to a respondent who did not answer the question or whose answer was incomplete or illegible—this may have contributed to the misclassification of Hispanics and others and, therefore, to the undercount or inaccurate count of these populations.

Moreover, national and other data collection efforts inconsistently use race and ethnic terms. Such inconsistency is problematic for Hispanics, who are considered, in most daily and practical applications, a “race” but who, on Census forms, are designated an “ethnic group.” The reliance on the concept of race, as defined in the U.S. by using a “Black-White” paradigm, ignores the growing importance of large Amencan ethnic groups—like Hispanics.

Any decisions made regarding the race and Hispanic origin questions must be carefully weighed and evaluated, especially if such changes are being considered for the 2000 Census.
NCLR:
Advocates the selection of identifiers that research and testing determine would ensure the most accurate response rate. It is critical that the racial and/or ethnic identifiers proposed for the 2000 Census and beyond be based on well-researched analysis and findings. While it is important to update the Census and revise it to reflect changing demographics, it is also important not to adopt terms simply because they may be preferable to some according to the political and social climate of the time.
Favors additional research to test different versions of the race and Hispanic origin questions. With respect to the Hispanic population, we believe focus group research conducted regionally and by Hispanic subgroup would assist in disentangling the nuances related to these questions. In addition, we encourage further investigation of the growing Central and South American populations, who may have a different perspective regarding the identification of race.
Does not support including Hispanic as a racial designation, rather than as a separate ethnic category.
Hispanics are a multiracial population and do not constitute a single race. NCLR recognizes that the terms “race and “ethnicity” are frequently used interchangeably and would support a question that included both terms. However, many Hispanics might not respond to a question that listed “Hispanic” solely as a race. On Census and other federal forms, we recommend maintaining use of the term “Hispanic” to denote an ethnic group.
Would be inclined to support the combination of the race and Hispanic origin questions into a question relabeled “Race/Ethnicity,” if testing indicates that such a question solicits a greater and more accurate response rate to the Census survey compared to the 1990 Census. As a way to reduce the large proportion of Hispanics who marked “other race”in the 1990 Census or who did not answer the race question, NCLR suggests testing a combination of the questions. Before large-scale changes are made, however, it is critical that the Census Bureau and the Office of Management and Budget determine which version of the questions accommodates the largest number of respondents and provides the most accurate data.
Strongly urges the Census to continue to collect Hispanic subgroup data. Hispanics are not a monolithic ethnic group. In fact, there is much diversity among the four principal Hispanic subgroups—Mexican American, Puerto Rican, Cuban, and Central and South American. Data for each group are critical to addressing and improving their status and should continue to be gathered. As certain groups increase in number, additional data collection may be warranted.
Does not support an open-ended question regarding race and ethnicity. An open-ended question would negatively affect the quality of the data collected for purposes of understanding the status of Americans relative to each other. Moreover, data that could not be aggregated would hinder the efforts of existing civil rights laws meant to protect specific “minority” groups from employment, housing and other forms of discrimination.
Does not support the addition of a “multiracial” category. As a multi-racial population, Hispanics understand and appreciate the complexity of issues of identity, and know that race and ethnicity are not necessarily distinguishable by physical characteristics. However, the original intent in collecting racial and ethnic data resulted, in part, from the lack of information on particular groups and the concern over the discrimination these groups face. Careful consideration must be given to the implications such categories would have on civil rights issues involving equitable representation and affirmative action, for which Census data are often used to determine coverage according to affected group. Double-counting of an individual (for twice-protected groups) may present a problem; additionally, such a category might encourage misclassification of people who may mark it without truly identifying as such. Furthermore, such a category would limit the usefulness of data collected about “multiracial” residents, since it would be difficult to address the needs of a profoundly heterogeneous group. For example. how would public policy address the theoretical 15% unemployment rate of “multiracials”?
Suggests a review of the purpose of the “other race” category. The purpose of the “other race” category is unclear. Although this category is presented for individuals who do not identify with the options presented, we understand that the Census Bureau currently reclassifies these data into one of the other categories. We wonder about the effectiveness of such a category.

Finally, in the challenge to document race/ethnicity, a major problem exists in trying to address the concerns related to identity expressed by the growing number of multiracial/multiethnic individuals. As the federal government seeks to be responsive to these concerns, we should all remember the need for and uses of accurate, reliable data about our countrys make-up. As a first step in understanding and addressing how identity fits into the redesign of racial ethnic categories, we should promote and support additional research in this area. For example, to learn about the existing biracial1 multi-racial population and how the public would respond to such “labels,” a special survey might be added to the Census’ yearly surveys. Such information may shed some light for the direction of future changes to government data collection efforts.

National data on Latinos and other racial/ethnic groups who have been marginalized in the U.S. are crucial to our progress and must continue to be collected. While we work to integrate our perspectives into how these data are collected, we should, above all, seek to ensure a reliable count based on quality data, as well as on an accurate reflection of the nation’s demographics.

Raul Yzaguirre is President of the National Council of La Raza, a Hispanic research and advocacy Organization (810 First St. NE, #300, Washington, DC 20002-4205). NCLR also represents more than 180 community-based “affiliates” that together provide services to more than two million Latinos across the country annually.

Sonia M Perez is NCLR Poverty Project Director.

Notes:

The lead article in our November/December issue, “Racial/Ethnic Categories: Do They Matter?” (an edited version of a July 25, 1994 New Yorker feature by Lawrence Wright, titled “One Drop of Blood), raised a critical scientific personal and political issue. We asked a number of social scientists and activists to offer their comments. This first round of the Symposium presents contributions by Raul Yzaguirre and Sonia M. Perez, Juanita Tamaro Lott. Libero Della Piana, john powell, Samuel L. Myers, Jr. and Reynolds Farley. The March/April issue of Poverty & Race will contain Part II of this Symposium. For new readers, a copy of the original article is available with a SASE.