Sample Research Report 

Hispanics, Mulattoes, and Native Hawaiian
or Other Pacific Islanders, (Oh My!):
The Census Bureau and Racial/Ethnic Formation

            Originally, I intended to analyze the incredibly rich lexicon used to categorize and describe immigrant and minority groups. I wanted to analyze several key terms historically and etymologically and then contrast them with dated words that are no longer “politically correct” with expressions that convey group pride and solidarity. My interest in the topic of racial and ethnic classification stems from being constantly barraged with questions of “What are you?” and, “What are we supposed to call you people?” These questions are important (despite the obvious lack of subtlety) since a person may believe that their entire identity begins with their response.

            What I found, however, was that both currently and historically, mainstream racial and ethnic nomenclature has, by default, fallen squarely on the shoulders of the Office of Management and Budget (OMB), who draws the racial boundaries, and the Census Bureau, who collects, processes, and digests the information. While investigating the history of both bureaucratic entities I uncovered a plethora of strange practices and quasi-racist policies that have become embedded in how this country classifies its citizens. Granted, this should have come as no surprise, because the same Constitution that requires the tabulation of its citizens also defined slaves as 3/5ths of a person. I will henceforth look apprehensively at any statistical chart on race and ethnicity and question the terms that are used to describe individuals—even when it is the individuals who are asked to describe themselves. The evasion of politics into racial formation in the US is so dense and convoluted that a compendium of information that ranges from amusing to offensive will only scratch the surface of the question: “What shapes the way we think about race and ethnicity and what are the consequences of our conclusions?”

The 1930 Census: Chock-Full of Peculiarities

            A list of a few of the oddities will help set the stage for a discussion of the Census. The best place to start is with the 1930 Census, since it is rather unique in the history of enumeration:

§         It was the only Census that offered “Mexican” as race (more on this later). (Perlmann 33).

§         Also available as an option for race in 1930 was “Hindu.”

§         A person of mixed “White” and “Negro” blood was to be considered Negro, no matter how small the percentage of “Negro” blood (Hinckley 11).

§         Similarly, a person of mixed Indian and Negro was to be listed as Negro unless the person was a majority Indian and/or “generally accepted as an Indian in the community” (ibid).

§         Any mixture of white and any other race was to be reported according to the race of the non-White parent. Mixtures of colored blood were to be listed according to the father’s race, except as noted above.

§         1930 was the last Census that asked an individual to report their year of immigration along with the birthplace location of the mother and father in order to determine ethnicity and nationality (Hinckley 13).

§         While surveying immigrants, the 1930 Census also asked what language a person spoke in the home before coming to the US, the so-called “Mother Tongue” (ibid.).

            Luckily, racial classification and the Census Bureau have evolved since 1930. This constant morphing likely ensures that in 70 years some college student will talk about the 2000 Census as the “Bad Ol’ Days.” And with good reason: the 2000 Census protocols already seem antiquated. For instance, according to current OMB classification guidelines, Hispanics may belong to any race. Before the 2000 Census form asked for a respondent’s race it asked if the respondent was Spanish/Hispanic/Latino. If yes, the person was prompted to check one of the three predetermined subcategories: Mexican, Mexican Am., Chicano; Puerto Rican; or Cuban. If none of these subcategories applied, the respondent could write in their other Spanish/Hispanic/ Latino Group. Only after this question is completed is an individual prompted to choose their race from a plethora of categories, some of which are broad (White; Black, African Am., or Negro), most of which are narrowed down to a single country (Chinese; Japanese; Korean; etc.). The award for the statistically smallest and most specific prelisted category goes to Native Hawaiian. Space is given for American Indians or Alaskan Natives to write down their tribe, and if no category presented is sufficient to the respondent, there is space to write in Some Other Race. More than 90 percent of the people who checked Some Other Race were Hispanic, which is less than half of the people who used the initial question to define themselves as Hispanic (Perlmann 55). Another source of confusion and miscalculation was the 40 percent of Hispanics who did not check any of the boxes for the major races (Perlmann 57). The question of how to categorize and define Hispanics will assuredly continue to plague the OMB as they prepare to issue their classification recommendations to the Census Bureau in preparation for 2010.

            The 2000 Census and the OMB guidelines on which it was based provided more questions than answers. Why is Korean, for example, considered a race while, let’s say, Brazilian is not? What are the prerequisites for getting your own box on the form? The Bureau decided in 1997 that Arab and Cape Verde should not be listed as races (Perlmann 45). Cape Verde, in fact, has nearly 3 times the population as Guam—a country that has its own subcategory classification (which was another source of confusion to be discussed below). What necessitates the addition or removal of a race from the Census seems to be a mystery. Equally mysterious is how the terminology used to describe African Americans seems to change every 10 years. In the first Census, Blacks were presumably counted as either Slaves (697,697 total) or within the category called All Other Free Persons (59,511 total) (Halacy 36). Colored and Free Colored appeared in 1820, but the names of slaves were never reported; they were merely listed by age under the name of the owner. The only specific details requested were age, sex, and color, and if the slave was deaf-mute, blind, insane, or idiotic. Although there were slaves in the Free states, the slave tabulations were only completed in slave-holding states.

            Beginning in 1850, Colored switched to Black, while Mulatto made its triumphant debut at the behest of Southern representatives in congress who believed that these “half-breeds” were inferior to both whites and blacks. In 1890, the Census began making further distinctions for Blacks, defining Mulatto, Quadroon, and Octoroon as half, one-quarter, and one-eighth Negro Blood, respectively. Then in 1900, most of these categories disappeared and were replaced, simply, with Black (Negro or of Negro descent). Mulatto remained for 20 more years, but the definition was changed from the old meaning, one-half Negro, to the new definition: any traceable amount of Negro blood.

            The Census Bureau introduced self identification in 1960, and has since relied on this practice when calculating the ethnic/racial makeup of the populace. This was largely done as a cost-saving measure since it was cheaper to have people fill out their own forms and let a computer process the information than paying a person to do the same. Prior to 1960 Census, enumerators used “objective indicators” to determine race (Skerry 46). This process was hardly objective: one of the main methods used to determine ethnic background was language use. Even the hint of an accent led the enumerator to make a subjective judgment, while the ethnic background millions of immigrants who spoke English was, by default, American. Another objective indicator was the birthplace of an individual or his/her parents. Since this method excluded third generation immigrants from being counted in the appropriate ethnic group, the inquiry also noted the birthplace of an individual’s four grandparents. But as a rule, enumerators typically relied on skin color and other phenotypic significations to determine race.

            Although it was not a novel concept to let individuals define themselves, the transition to self identification marked a new and problematic chapter in the history of the OMB and the Census Bureau. The most noticeable immediate impact of self identification was the large statistical increase of American Indians who had previously been identified by Census takers as White (Perlmann 34). Since the inception of self identification, “Our concept of the self has become increasingly psychologized” (Skerry 48). Skerry notes that this transition has left Census-taking in a state of disarray. In 1990, when the Census Bureau allowed a write-in category for race, it received nearly 8 million responses, which it compiled into 300 coded racial categories and 600 American Indian tribes (Skerry 50). To further complicate matters, there were more than 250,000 multiracial write-in responses. To save time, the Census Bureau recoded these responses based on the first race written; an individual who defined their race as White/Black became White, while the individual who wrote in Black/White became Black. Over 40 percent of the total write-in responses were people who wrote in an ethnic group as their Other Race, such as Irish, Arab, Iranian, or Jamaican (ibid). These individuals were also subsequently reclassified by the Census Bureau. Clearly, there was some confusion among respondents as to the distinction between race and ethnicity. An even better example of this confusion was seen by the noticeable number of people who identified themselves as Guamanian in the subcategory box under Asian or Pacific Islander and then wrote Guatemalan under the ancestry question. Also, there were many individuals who listed their ancestry as American Indian and their race as Asian Indian or Pakistani (Skerry 57). Since all of these errors required recoding by the Census Bureau, giving people the option to self identify was clearly the beginning of a work in progress—if not a misnomer altogether.

            Prior to 1990, there were no challenges to the idea that every American could be “readily classified into one and only one race” (Perlmann 34). But the plethora of write-ins led the OMB to allow individuals to check multiple boxes in the 2000 Census. The significance of this change is not to be underestimated. This shift from the idea that every American had only one racial identity was, indeed, radical. The option to check multiple races came not at the hands of a powerful civil rights lobby but because a “small number of persistent advocates who were greatly upset by the status quo [and] effectively challenged the traditional system” found the sympathetic ear of the Clinton Administration (Perlmann 40).

            Yet the novel new approach was not immune to problems. In the 2000 Census, 2.4 percent of the total population identified with two or more races (Perlmann 58). This percentage, which translates into 6.8 million people, is widely lauded as proof that the United States is ethnically diverse and becoming more so, tolerant of racially ambiguous individuals, and accepting of the interracial relationships that have created this new ambiguity. But the ranks of the multicultural and mixed race individuals are not as big as the data suggests. Included in the 2.4 percent are all the Hispanics who checked one of the standard race boxes (such as White or Black) and wrote in a term denoting their Hispanic origin as their second race. Adjusting for this error, the actual amount of the population who consider themselves to be two or more of the five major races is only 1.6 percent. After subtracting Hawaii’s mixed race population (a statistical anomaly at 21.4 percent for several reasons related to their population’s unique racial distribution), the total number of people in the country who claimed two or more race is just over four million.

            The concept of race plays a fundamental role in structuring and representing the social world (Omi 55). As seen in the information above, the OMB and the Census Bureau are instrumental, if not complicit, in this structuring. The categories used to differentiate groups among racial and ethnic lines are “at best imprecise and at worst completely arbitrary” (Omi 55). Nevertheless, it is difficult to abandon commonly held beliefs that are engrained in the social world. Discussion of these distinctions requires the concession that racial/ethnic distinctions are not based on scientific taxonomy. The OMB themselves has reiterated this seemingly obvious point by stating, “These classifications should not be interpreted as being scientific or anthropological in nature” (Skerry 44). The categories are not going anywhere. Perhaps the sordid past of categorizing and classifying will eventually initiate a substantive discussion on race and ethnicity. Until then, one must simply marvel at the peculiarity of a country so fascinated with racial taxonomy.

Editorial Addendum

            Coming to a sufficiently relevant conclusion in a discussion about race and ethnicity is almost impossible. There is a growing number of people who want to abolish the boxes altogether. Not me. This is all far too fascinating. And there is history in the boxes for me: With the exception of my siblings, I have only known two other people “like me,” more specifically, black/white males. One of these guys, Travis, was a friend of mine: we shared homeroom throughout high school. The long-running joke between us was that the school “put both of the darkies together.” In reality, our names were next to each other in alphabetical order—apparently the determining factor in homeroom assignments. But you couldn’t tell that to Travis and me. While the answer sheets with their endless little boxes (including one for race) were being handed out during standardized testing time, Travis and I would always consult each other: “What race are you gonna be this year, J.?” I would respond, “I dunno, T. American Samoan is lookin’ mighty tempting.” Travis always hated the idea of having to fit in a single category, so he would act confused and become angry at the entire process: “American Samoan? They get their own box?!” For some reason, we always made an effort to select the same race. I guess we didn’t want our numbers diluted. Unfortunately for the two of us (and the statistical accuracy of the tabulations), checking more than one box was not allowed until the year after we graduated. My research has reinforced what I already knew: statistically speaking, I am a minority among minorities. I can continue to expect to be the only one around with my background unless I am in a room with Halle Berry, Thandie Newton, Mariah Carey, and Sade—in which case racial categories would be the last thing on my mind.

Works Cited

Halacy, Dan. Census: 190 Years of Counting America. New York: Elsevier/Nelson, 1980.

Hinckley, Kathleen W. Your Guide to the Federal Census. Cincinnati, OH: Betterway, 2002.

Omi, Michael, and Howard Winant. Racial Formation in the United States: From the 1960s to the 1990s. New York: Routledge, 1994.

Perlmann, Joel, and Mary C. Waters, eds. The New Race Question: How the Census Counts Multiracial Individuals. New York: Russell Sage Foundation, 2002.

Skerry, Peter. Counting on the Census? Race, Group Identity, and the Evasion of Politics. Washington: The Brookings Institution, 2000.

[JC]