The 1990 Census

The controversy over the issue of multiracial identity became intense in 1988. On January 20 of that year, the Office of Management and Budget (OMB), the branch of the government responsible for implementing changes in federal statistical surveys, published in the Federal Register a notice soliciting public comment on potential revisions in Directive No. 15. This directive was implemented in May, 1978, as the government-wide guide for conducting racial/ethnic surveys. The revisions would permit individuals to identify themselves as “other” if they believed they did not fall into one of the four basic official racial categories—black, white, Asian/ Pacific Islander, American Indian and Alaska Native—or in the so-called “ethnic” category, Hispanic. Heretofore, the OMB advised that the category that most closely reflected how the individual was recognized by the larger community should be used in cases where there was any uncertainty. (Although an “other” category has not been used on all statistical surveys, it has been provided on each census since 1910 to increase the response rate to the race question. However, write-in responses in the “other” category are reassigned to one of the traditional racial categories.)

Many interracial couples and multiracial-identified individuals requested that a multiracial or biracial identifier, instead of “other,” be added to the five categories. (On the 1970 census, multiracial offspring were classified in terms of the father’s racial identity; in 1980, the Census Bureau shifted to a formula relying on the identity of the mother. However, a “biracial” or “multiracial” designation was not permitted.) The OMB received overwhelmingly negative responses from the public to the proposed changes to Directive No. 15, particularly the addition of a multiracial identifier. This included some federal agencies, such as the Civil Rights Division of the Department of Justice, the Department of Health and Human Services, the Equal Employment Opportunity Commission, the Office of Personnel Management, and several large corporations. Some of this opposition was based on logistical and financial concerns about the increase in data collection, paperwork, changes in the format of forms and computer programs for data analysis, and data burden on respondents.

Various African American leaders and organizations in particular voiced their opposition to the change. They argued that most, if not all African Americans, have some European, and in many cases, Native American ancestry (although most identify solely with the black community). Consequently, they feared that many individuals would designate themselves as “multiracial” rather than black in order to escape the continuing negative social stigma associated with African Americans. Similar concerns were expressed by individuals and organizations representing other traditional communities of color, including Latinos, Native Americans, and Asian Americans. In addition, opponents argued that the rule of hypodescent, if originally oppressive, has also been a means of mobilizing communities of color in the struggle against white racial privilege. More important, this mechanism has prevented a reduction in the number of individuals who would be counted as members of the traditional communities of color. These numbers were needed to enforce and support civil rights legislation and claims aimed at tracking historical and contemporary patterns of discrimination. They were particularly important in arriving at goals for achieving social and economic equity in the manner of affirmative action.

On November 12, 1988, the AMEA was formed in Berkeley to serve as a national network for the various independent support groups. Its overall goal was to promote healthy images of interracial couples and multiracial individuals. More specifically, its purpose was to increase public awareness about the importance of adding a multiracial identifier to the decennial census, A flurry of telephone calls and correspondence between officials at the OMB and the Census Bureau and various individual groups affiliated with the AMEA ultimately resulted in some clarity as to how multiracial individuals might be accommodated on the 1990 census. Officials said they would specifically code write-in responses in the “other” category as “biracial,” “multiracial,” or some other designation that clearly indicated a blended identity. This would help determine what, if any, changes should be made on the year 2000 census. This was a departure from policy on previous censuses. However, none of the approximately 253,000 multiracial responses in the other category on the 1990 census can be used as a accurate estimate of the actual number of individuals who identify as multiracial. Also, vast numbers of individuals followed the tradition of circling one box because they were unaware of any other alternative.

The Year 2000 Census

Efforts to get the U.S. Census Bureau to make “multiracial” an acceptable means of self-identification were unsuccessful for the 1990 census. Nevertheless, forms for the Operation Desert Shield/Storm Deployment Survey included a multiracial designator for the offspring of returning intermarried veterans. Under the guidance of Project RACE, several states included “multiracial” as an acceptable official means of self-identification. Georgia, Ohio, Illinois, Michigan, Indiana, and Maryland made this option available on all official state forms. Florida and North Carolina included a multiracial identifier on all school forms. A 1994 survey of eight hundred public school districts, conducted by the Education Office for Civil Rights, found that approximately 30 percent of the school districts use a special separate category. Other districts simply use the mother’s racial designation; some use the father’s. The American College Test (ACT), which is the alternative to the Scholastic Aptitude Test (SAT) college entrance exam, included “multiracial” as an acceptable means of identification. Most universities were resistant to any changes in the collection of racial data; however, Williams College in Williamstown, Massachusetts, included a multiracial identifier on its official forms. Beginning in 1989, reports prepared by the Center for Assessment and Demographic Studies at Gallaudet University in Washington, D.C., counted individuals who indicated identification with a multiracial background. Nevertheless, these data were reassigned at the federal level to one of the four official racial categories (along with the Hispanic identifier, when it was given as an option) or added into the figures for each of the racial groups with which multiracial individuals identified. This second method is especially useful when trying to track those whose background includes historically underrepresented groups of color for the purposes of affirmative action.

“Multiracial” was not accepted as a means of self-identification on federal forms during the early 1990s. However, the OMB began a comprehensive review process in 1993 to discuss possible changes in this direction on the year 2000 census. After extensive cognitive research and field testing of sample households, the OMB on October 30, 1997, approved changes that would allow multiracial individuals to identify themselves as such on official forms. (The key findings based on the comprehensive review were that between 1 percent and 1.5 percent of the public would select a multiracial identifier when offered an opportunity to do so.)

Most activists had hoped for a combined format that would include a separate multiracial box but would also allow individuals to check the other boxes representing the various components of their background. The OMB proposed a format that read: “What is this person’s race? Mark |X| one or more races to indicate what this person considers herself/himself to be.” This format was chosen partially in response to the unanimous support it received from the various federal agencies that require data on race and ethnicity. These agencies argued that the mark-one-or-more alternative— unlike the combined format—would require fewer changes in formatting on existing forms and allow for data continuity. More important, the data could be counted in each of the existing official racial categories with which multiracial individuals identified, thus including the historically underrepresented racial components in their background. This would be especially important for the purposes of the continued enforcement of civil rights legislation and in meeting affirmative action guidelines. This format is similar to the one that appears on the Canadian census. Since the 1980s, Canada has allowed individuals to check more than one box on the census race/ethnic ancestry question. In addition, Canadian census data have been used for the purposes of achieving “job equity” in a manner similar to affirmative action and other civil rights mandates in the United States.

The mark-one-or-more format also received strong support from traditional civil rights organizations such as the National Association for the Advancement of Colored People (NAACP), the National Urban League, the Congressional Black Caucus, and the Mexican American Legal Defense and Education Fund (MALDEF). These groups argued that a stand-alone multiracial identifier would lead to a decline in their numbers. (It should be noted, however, that the combined format would have prevented this loss. In addition, that format would have had the advantage of making it possible to count the data in each of the racial groups with which multiracial individuals identify as well as specifically acknowledging the identity of multiracial individuals.) Furthermore, the traditional communities of color expressed concerns about the potential divisiveness of even the appearance of a multiracial box, whether as a stand-alone identifier or in combination with checking multiple boxes. These concerns were very influential in prompting the OMB officials to chose the mark-one-or-more format. In addition, various representatives from traditional communities of color expressed concerns about how the data would actually be tabulated even though they supported the mark-one-or-more format. The OMB officials indicated that they would make recommendations and provide additional guidance with respect to this question before the year 2000 census after consulting with officials from various federal agencies, interested groups, demographers, planners, and social scientists.

The issue continued to generate some controversy in 2010 and beyond, particularly as it pertained to Hispanics. In the twenty-first century, a large and growing number of people come from families with one white and one minority parent; the largest group among these consist of younger persons with Hispanic and white European ancestry. However, the 2010 census, as well as the 2000 census, does not reflect this changing demographic. While the 2000 census was the first to allow respondents to report a multiracial heritage, that census, and the 2010 census, provided respondents with the opportunity to indicate that they were part Hispanic and part something else. The problem is that the census continued to examine race and ethnicity with two questions: The first is race and the second is “Hispanic origins.” This format was consistent with the 1997 OMB guidelines. Looking ahead, the Census Bureau announced that this format would not change on the 2020 census. The result, however, is that the Census Bureau then assumes that any respondent is only Hispanic, ignoring the answer to the race question—even though Hispanics may be of any race. In 2015, the bureau tested a single question that treated “Hispanic origin” like a race and that allowed respondents to report more than one origin. The result was that nearly 30 percent of Hispanics indicated a second ethno-racial origin, usually white. A further result was to reduce the number of Americans, most of them Hispanics, who claim an “other” option on race.

Further Reading

See Also: Affirmative Action; Biracialism; Censuses, U.S.; One-Drop Rule.