Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Race and ethnicity concepts, which have evolved continually in the United States, must be periodically examined in order to ensure that the nation is keeping pace with its changing composition. Following each census, it is appropriate to examine the current set of race and ethnicity concepts to assess their adequacy in describing the population. Results from the 2010 Census reveal that more than 15 million people, the majority of whom were of Hispanic origin, did not report any of the standard race categories utilized by the US federal government. Another 14 million people did not answer the race question at all. Over the past few censuses, the number not reporting a standard race category has increased, while the race and ethnicity concepts have not adapted enough to address factors that may be fueling this trend. Policymakers in federal agencies, local governments, business, universities, and think tanks need data based on race and ethnicity concepts that accommodate the continued rapid growth of the Hispanic population, and of groups who trace their ancestry to, or identify with, an increasingly diverse set of world population groups.

The classification of race and ethnicity has a long history of change in the United States—as evidenced by the measurement of these concepts in every decennial census since 1790. These concepts have evolved from census to census, influenced by political, social, and demographic phenomena in American society such as slavery, the civil rights movement, and immigration patterns, as well as by scientific and pseudo-scientific concepts of race and ethnicity. We argued previously that in order for constructs of race and ethnicity to be socially, or at least statistically, useful they would have three properties: (1) be recognized by society and the individual, (2) categorize individuals into the same groups over a long period of time, and (3) be predictive of social and economic opportunity (Humes and Hogan 2009). To the extent that the current or any categorizations of race do not meet these criteria, the data collected based upon them will not provide an adequate picture of society and its needs.

This chapter examines current race and ethnicity concepts used by the US federal government and comments on their applicability to the current US population. We provide an overview of race and ethnicity in the United States. We also examine segments of the population where the application of the current race and ethnicity concepts may be most problematic and investigate additional social dimensions that may impact individual’s acceptance or rejection of current race and ethnicity standards. Additionally, we discuss possible future research directions regarding race and ethnicity concepts. Throughout this chapter, data from the decennial census and the American Community Survey are used to provide insights into the race and ethnicity concepts utilized by the US federal government.

Current Race and Ethnicity Concepts

In 1997, the US Office of Management and Budget (OMB) issued the current standards for the collection and tabulation of federal data on race and ethnicity.Footnote 1 OMB standards state four key principles: (1) there are five minimum categories for data on race: White, Black or African American, American Indian or Alaska Native, Asian, and Native Hawaiian and Other Pacific Islander; (2) the reporting of more than one race is accepted; (3) there are two minimum categories for data on ethnicity: Hispanic or Latino and not Hispanic or Latino; and (4) race and ethnicity are two separate and distinct concepts—therefore, people of Hispanic or Latino origin may be any race. The US Census Bureau, along with other federal agencies, must adhere to OMB standards. The general purpose of the OMB standards is to provide relatively consistent statistics on race and ethnicity across the federal statistical agencies for such uses as the enforcement of civil rights laws and monitoring equal access to education, employment, housing, etc.

The 1997 OMB standards differ from the first set of government-wide race and ethnicity standards issued in 1977. The two biggest changes were: (1) the Asian and Pacific Islander category was split into two groups and (2) multiple-race reporting was introduced. These are the only two major changes that have occurred in the race and ethnicity concepts used by the federal government in more than 30 years. Are these changes enough to adequately reflect the race and ethnicity composition of America as it has taken shape over the past 30 years? While it is desirable for change in race and ethnicity concepts to occur slowly over time, in order to minimize the disruption of the historical time series, static race and ethnicity standards may not capture sufficient data about evolving major population groups.

While only two major changes were introduced in the 1997 OMB revision, other significant race and ethnicity conceptual issues were being discussed among demographers. Most notable were the consideration of an additional race category for people of Middle Eastern and North African heritage, as well as the combining of the separate concepts of race and ethnicity into one. Although these issues were tabled as needing further research, they show that population group identification issues were changing in substantial ways.

Race and Ethnicity Distribution Patterns

Data from the decennial census and from the American Community Survey show how the American public responds to the current concepts of race and ethnicity used by the federal government and suggest whether these concepts adequately describe the US population.

2010 Census Race and Ethnicity Overview

According to the 2010 Census, 308.7 million people resided in the United States on April 1, 2010. This total included 50.5 million Hispanics, who composed 16 % of the total population (see Table 2.1). Additionally, 97 % of the total population (299.7 million) were classified into one of the single race groups. Individuals who were one race are referred to as the race-alone population.Footnote 2 The largest group reported White alone (223.6 million), accounting for 72 % of all people living in the United States.Footnote 3 The Black or African American alone population was 38.9 million and represented 13 % of the total population.Footnote 4 There were 2.9 million people categorized as American Indian and Alaska Native alone (0.9 %). Approximately 14.7 million (about 5 % of the total population) were Asian alone. The smallest race group was Native Hawaiian and Other Pacific Islander alone (0.5 million) and represented 0.2 % of the total population. The remainder of the population who were one race—19.1 million (6 % of the total population)—were classified as Some Other Race alone. People who were more than one race numbered 9.0 million in the 2010 Census and made up about 3 % of the total population.Footnote 5

Table 2.1 Population by Hispanic or Latino origin and by race for the United States, 2010
Full size table

Figure 2.1 shows the size in 2010 and the percentage change between 2000 and 2010 for selected race and ethnic groups.Footnote 6 In the United States, all race and ethnic groups increased in population size between 2000 and 2010, but they grew at different rates. Between 2000 and 2010, the Hispanic population grew relatively fast, increasing by 43 %. The non-Hispanic Asian-alone population experienced fast growth and also increased by 43 % between 2000 and 2010, more than any other race group. The non-Hispanic Native Hawaiian and Other Pacific Islander-alone population, the smallest race group, also grew substantially between 2000 and 2010, increasing by more than one-third. The non-Hispanic Two or More Races population was also one of the fastest-growing groups over the decade. This population increased almost one-third between 2000 and 2010.Footnote 7

Fig. 2.1
figure 1

Numeric size in 2010 and percentage change between 2000 and 2010 by selected race and Hispanic origin group (For information on confidentiality protection, non-sampling error, and definitions, see http://www.census.gov/prod/cen2010/sf1.pdf)

Full size image

Within the non-Hispanic population, several groups grew relatively slowly. The non-Hispanic Black-alone population grew 11 % between 2000 and 2010, far slower than Hispanics, Asians, Native Hawaiians and Other Pacific Islanders, and the Two or More Races population. Additionally, 9 % growth in the non-Hispanic American Indian and Alaska Native-alone population occurred over the decade. The non-Hispanic White-alone population grew at the slowest rate between 2000 and 2010 (1 %). Further, while the non-Hispanic White-alone population increased numerically from 194.6 to 196.8 million over the 10-year period, its proportion of the total population declined from 69 to 64 %.

Racial Distribution Among Hispanics in the 2010 Census

An examination of the pattern of responses provided to the 2010 Census questions on race and Hispanic origin provides important insights. Table 2.2 presents information on the type of response (or no response) given to the 2010 Census race question cross-tabulated by the type of response (or no response) given to the 2010 Census question on Hispanic origin—prior to the application of any data editing or allocation procedures.Footnote 8 In terms of reporting race, the majority of the total population (90 %) provided one or more OMB race categories only. Another 5 % of the total population provided a response to the race question that could not be racially categorized, which resulted in their responses being classified as Some Other Race only. Less than 1 % of the total population reported a mixture of OMB race categories and responses that were classified as Some Other Race. Those who did not respond at all to the race question represented 4.5 % of the total population.

Table 2.2 Responses to the Hispanic origin question and the race question, 2010
Full size table

The pattern of response to the race question changes dramatically when taking into account the type of response to the Hispanic-origin question (prior to the application of any data editing or allocation procedures). Among those who reported not being Hispanic, virtually all (99 %) provided OMB race responses. Among those who reported being Hispanic, 53 % reported OMB race categories only, 31 % provided responses classified as Some Other Race only, 4 % provided a mixture of OMB race(s) and Some Other Race responses, and 13 % did not respond to the race question at all. It is striking that almost half of all those who reported being Hispanic either did not provide a race response that was classifiable, did not respond to the race question at all, or provided a mixture of OMB race categories along with responses that could not be racially classified. This is significant given that the Hispanic population is the largest and one of the fastest-growing minority groups in the United States.

Recall that OMB standards mandate that people of Hispanic origin may be any race. This reflects the diverse populations (especially European, African, and indigenous American) that constitute the Spanish speaking world. For the 2010 Census, a new instruction was added immediately preceding the questions on Hispanic origin and race that was not used in Census 2000 (see Fig. 2.2). The instruction stated that “For this census, Hispanic origins are not races.” However, this did not preclude individuals from self-identifying their race as “Latino,” “Mexican,” “Puerto Rican,” “Salvadoran,” or other national origins or ethnicities; in fact, many did so. If the response provided to the race question could not be classified in one or more of the five OMB race groups, it was generally classified in the category Some Other Race. Therefore, responses to the question on race that reflect a Hispanic origin were classified in the Some Other Race category.

Fig. 2.2
figure 2

Reproduction of the questions on Hispanic origin and race from the 2010 census. Source: U.S. Census Bureau, 2010 Census questionnaire

Full size image

Table 2.3 and Fig. 2.3 examine the racial distribution among the Hispanic population by origin, utilizing final 2010 Census data that have undergone data editing and imputation procedures. Just over half of the total Hispanic population was classified (either directly from their responses, or via editing or imputation) as White and no other race, while about one-third were classified as Some Other Race alone. Much smaller proportions of Hispanics were other race groups alone: Black alone (3 %), American Indian and Alaska Native alone (1 %), and Asian alone or Native Hawaiian and Other Pacific Islander alone (0.5 %).Footnote 9

Table 2.3 Hispanic or Latino population by type of Hispanic or Latino origin and race, 2010
Full size table
Fig. 2.3
figure 3

Hispanic or Latino population by origin and race: 2010 (For information on confidentiality protection, non-sampling error, and definitions, see http://www.census.gov/prod/cen2010/sf1.pdf)

Full size image

When examining the racial distribution by origin, we see several trends. In general, large proportions of Hispanics of Central American origin were White alone or were classified as Some Other Race alone. At least half of Hispanics of Mexican origin and at least 60 % of Hispanics of Costa Rican and Nicaraguan origins were White alone. Additionally, about one-fifth of Costa Ricans and one-quarter Nicaraguans were classified as Some Other Race alone. A larger proportion of Mexicans were categorized as Some Other Race alone (39 %).

A different pattern was observed for Hispanics of Guatemalan, Salvadoran, and Honduran origins, where less than half of these populations were White alone (38 %, 40 %, and 43 %, respectively). Forty-three percent of Hondurans were classified as Some Other Race alone, while half of Salvadorans and nearly half of Guatemalans were Some Other Race alone. The proportions of most Central American origin groups who were other races were relatively small.

Origin groups with larger proportions classified as Some Other Race alone could reflect the more complex racial makeup of their countries of origin or a conceptualization of racial and group identity different from that of the OMB. Those with origins in countries that have significant Mestizo populations, as well as Central American indigenous populations, may find the federal government racial classification system not relevant to them. The Panamanian population, however, had a different pattern than those of other Central American origins. Panamanians were the least likely among those of Central American origin to be White alone (33 %) or Some Other Race alone (19 %), and they were the only Central American origin to significantly identify as Black or African American (33 %).

Hispanics of South American origin had a different racial distribution than those of Central American origin. In general, very large proportions of those of South American origin were White alone and small proportions were categorized as Some Other Race alone. At least 70 % of those of Argentinean, Chilean, Colombian, Paraguayan, Uruguayan, and Venezuelan origins were White alone. The proportion classified as Some Other Race alone was 20 % or less for each of these origin groups. The majority of those of Bolivian, Ecuadorian, and Peruvian origins also were White alone, but to a lesser extent than others of South American origin (64, 52, and 55, respectively). Conversely, those of Bolivian, Ecuadorian, and Peruvian origins were the most likely among those of South American origin to be classified as Some Other Race (27 %, 38 %, and 34 %, respectively). Again, this pattern may reflect the significance of Mestizo and the presence of South American indigenous populations in these countries, which may make it difficult for those of these origins to identify with OMB race categories.

Hispanics of Caribbean origin did not exhibit a consistent racial distribution, perhaps reflecting both the racial makeup of the countries of origin and the particular histories of immigration to the United States. About half of those of Puerto Rican origin were White alone, with 28 % classified as Some Other Race alone. A very large proportion of those of Cuban origin (85 %) were White alone, with 6 % classified as Some Other Race alone. Of all Hispanic origin groups, Cubans had the lowest proportion categorized as Some Other Race alone. In contrast, 46 % of those of Dominican origin were categorized as Some Other Race alone, and 30 % were White alone. Of all Hispanic origins, Dominicans had the lowest proportion of White alone.

The different racial distributions among Hispanic-origin groups reveal that, while just over half of all Hispanics identify with one or more OMB race groups, large proportions did not identify with any of them. This is particularly true for those with origins in countries with significant Mestizo populations and indigenous populations. Individuals with origins in Central American, South American, and Hispanic Caribbean countries whose concepts of race and ethnicity are distinct from those used in the United States may determine that the OMB race categories do not apply to them. Other dimensions can also be examined to explore factors that may indirectly impact identification with the current federal government race categories.

Examining Citizenship Status, Year of Entry, and Racial Distribution Among Hispanics

If lack of identification with OMB standards occurs because these race concepts are incongruent with those in Hispanic-origin countries, citizenship status and year of entry data may provide evidence to support or disprove this theory. Perhaps those who have had more exposure to and immersion in US culture have also had more opportunity to understand the racial constructs used by the federal government. They would be more likely to identify with OMB race categories than newer arrivals. Data from the 2005–2009 American Community Survey offer information on the intersection of citizenship status, year of entry, and racial distribution among Hispanics.

Table 2.4 shows the percentage of the Hispanic population 20 years and older who were classified as Some Other Race alone by citizenship status and selected origin.Footnote 10 When taking citizenship status into account, Hispanics who were not citizens (41 %) were more likely than Hispanic citizens (34 %) to be classified as Some Other Race alone.Footnote 11 Conversely, Fig. 2.4 shows that among Hispanics 20 years and over, citizens were slightly more likely to be classified as White alone or other OMB race groups than non-citizens. This racial distribution could suggest that non-citizens are less likely than citizens to understand the race categories delineated by OMB.

Table 2.4 Percentage of the Hispanic or Latino population 20 years and over classified as some other race alone by citizenship and selected origin, 2005–2009 (in percent)
Full size table
Fig. 2.4
figure 4

Hispanic or Latino population 20 years and over by citizenship and race: 2005–2009 (Data based on sample. For information on confidentiality protection, sampling error, non-sampling error, and definitions, see www.census.gov/acs/www)

Full size image

Additionally, when examining the classification of Some Other Race alone among Hispanics 20 years and over by citizenship status and selected origin, the above pattern holds for those of Mexican origin and of Dominican origin. However, about half of citizens and half of non-citizens of Guatemalan and Salvadoran origin were classified as Some Other Race alone. Therefore, for some origins, controlling for citizenship does not appear to impact the racial distribution. Perhaps an additional dimension contributes to this distribution.

Figure 2.5 shows the racial categorization for foreign-born Hispanics 20 years and over by year of entry. The recency of immigration impacts the racial distribution. Foreign-born Hispanics who entered the United States after 1970 were more likely to be classified as Some Other Race alone than those who entered before the 1970s. Conversely, Hispanics who entered the United States before the 1970s were more likely to be White alone than those who entered after 1970. It is notable that the proportion of foreign-born Hispanics who identified as any of the other race groups was about 4 or 5 % across all categories of year of entry.Footnote 12 The Immigration and Nationality Act of 1965 significantly changed the way the immigration quotas were allocated to different parts of the world, including the nations of the Western Hemisphere, as well as the criteria for admission. Thus, the racial composition of the immigration flow changed. Additionally, the effects of the civil rights movement may have made the United States more attractive to immigrants who were not White.

Fig. 2.5
figure 5

Foreign-born Hispanic or Latino population 20 years and over by year of entry and race: 2005–2009 (Data based on sample. For information on confidentiality protection, sampling error, non-sampling error, and definitions, see www.census.gov/acs/www)

Full size image

Similar patterns emerge when examining the proportion of foreign-born Hispanics 20 years and over who were classified as Some Other Race alone by year of entry and origin. Figure 2.6 shows that the proportions of those of Mexican and Dominican origins who were classified as Some Other Race alone were higher for those who entered the United States after 1970 than for those who entered prior to the 1970s. For those of Salvadoran origin, the proportions classified as Some Other Race alone were higher for those who entered the United States after 1970 than for those who entered before 1960. People of Guatemalan origin who entered the United States from 1970 to 1999 were more likely to be classified as Some Other Race alone than those who entered prior to the 1970s. However, this pattern did not hold for those of Cuban origin. The proportion of Cubans categorized as Some Other Race alone was below 10 % for all year of entry categories, except for the 1980s (10 %). This could reflect the increased racial diversity that existed among the wave of refugees that arrived in the United States via the Mariel, Cuba boatlift in 1980.

Fig. 2.6
figure 6

Foreign-born Hispanic or Latino population 20 years and over who identified as some other race alone by origin and year of entry: 2005–2009 (Data based on sample. For information on confidentiality protection, sampling error, non-sampling error, and definitions, see www.census.gov/acs/www)

Full size image

Thus, the 2010 Census and the 2005–2009 American Community Survey show that large proportions of the Hispanic population do not identify with OMB race categories. Responses to the race question that were classified as Some Other Race alone varied by Hispanic origin, citizenship status, and year of entry. This finding could reflect the importance of the racial makeup of an individual’s, or their preceding generation’s, country of origin. It could also reflect the social climate encountered by individuals when arriving in the United States, as well as the acculturation of those individuals over time.

2010 Census Race Responses Not Classified or Seeking Group Recognition

A number of people enumerated during the 2010 Census indicated that the OMB race categories did not apply to them and/or that they were seeking recognition for population groups that are not typically tabulated from the decennial census. Table 2.5 displays the most frequently reported responses to the 2010 Census race question that could not be racially or ethnically classified.Footnote 13 By far, the term most commonly reported was simply “American” (nearly two million responses). Other responses, such as “Human,” “Human race,” “No,” “None,” and “N/A” (not applicable), could be interpreted as representing a rejection of the race question itself. Additionally, responses that could not be racially or ethnically classified included religious responses, the most commonly reported being “Jewish” and “Muslim.” Also, some respondents appeared not to know how to identify their race and reported “Unknown” or “Other.” The counts of people reporting a particular term that could not be racially or ethnically classified drops significantly beyond “American,” reflecting the myriad of ways this segment of the population struggled to answer the question on race.

Table 2.5 Top ten uncodable responses to the 2010 Census question on race
Full size table

As discussed earlier, a number of people reported a Hispanic origin as their race. Table 2.6 shows the most frequently reported responses to the race question that were Hispanic origins. There were 7.5 million responses of “Mexican,” the most reported Hispanic origin in response to the question on race. Still, it should be noted that many (53 %) of people of Mexican origin were White alone. Other specific Hispanic-origin groups reported by several hundred thousand people were “Puerto Rican,” “Dominican,” “Salvadoran,” and “Guatemalan.” After reports of “Mexican,” the second and third most frequently reported Hispanic-origin responses to the race question were general terms. Just over four million people reported their race as “Hispanic” and almost two million reported “Latin American.” Thus, the wide range of Hispanic-origin responses to the question on race indicate a broad inability or unwillingness to identify with OMB race categories.

Table 2.6 Top ten Hispanic origin responses to the 2010 Census question on race
Full size table

Leading up to the 2010 Census, a number of organizations launched campaigns instructing their communities how to complete the question on race. A basic goal was to obtain an official count of their community from the 2010 Census, which was not included in the standard census data products, nor was the race question designed to elicit their community-specific responses for tabulation in standard data products. Most notable were the efforts of the Arab and Iranian communities. OMB standards define those with origins in the original peoples of Europe, the Middle East, and North Africa as “White.” Therefore, reports of “Iranian” or Arab nationalities to the 2010 Census question on race were classified and tabulated in the category “White.” Historically, the Census Bureau has not separately tabulated White ethnic groups, thus, counts of White ethnic groups are not included in standard census data products. However, a number of advocates for this community lobbied the Census Bureau to collect ancestry data on the 2010 Census, in order to obtain counts of Arab groups and Iranians, which are traditionally tabulated from the ancestry question formerly asked on the long form of the census and currently asked on the American Community Survey. The Arab American Institute garnered the support of the Democratic National Committee, which passed a resolution calling for the collection of ancestry data in the census; however, this was passed too late to impact the 2010 Census questionnaire design. This community urged its members to write in their group affiliation in the race question, expecting to obtain group counts through a custom tabulation after the 2010 Census.

Additionally, members of the Afro-Caribbean community launched campaigns regarding the reporting of race prior to the 2010 Census. OMB standards define those with origins in the Black racial groups of Africa as “Black or African American.” Therefore, reports of Afro-Caribbean and African nationalities are classified and tabulated in the category “Black or African American.” Also, historically, the Census Bureau has not separately tabulated Black ethnic groups, thus, counts of Black ethnic groups are not included in standard census data products. White House briefings on the 2010 Census collection and tabulation of data on race were organized by the Afro-Caribbean community. This community also influenced the introduction of a bill in Congress that required the addition of a “Caribbean” checkbox to the 2010 Census question on race. Again, because of the timing, changes to the 2010 Census race question could not be entertained. This community used social media, among other methods, to reach out to their members, instructing them to write-in their group affiliation in the race question, with the expectation that a custom tabulation of group counts would be obtained after the 2010 Census.

Table 2.7 shows the most common specific write-ins classified as “White” or “Black or African American” responses to the 2010 Census question on race. It is expected that reports of Italian, German, Irish, and Polish would be commonly reported, as these ethnic groups are among the largest in the United States. However, it is less expected to see large numbers reporting Iranian, Arab, and Middle East in the list of the top ten, as these groups represent a relatively small segment of the US population. Additionally, reports of Haitian and Jamaican are included in the list of the top ten. We argue that reports of these relatively small population groups indicate seeking group recognition, as well as a rejection of OMB categories as the sole representation of their racial identity. Unfortunately, the communities discussed above that advocated their members write-in their group affiliation in the race question expected that data from the 2010 Census could provide official counts of their communities. Since the question on race was not designed to collect data on White or Black ethnic groups, these counts do not reflect the size of these communities. The data in Table 2.7 represent only the number of people who went out of their way to report these responses in the 2010 Census race question and should not be construed as accurate counts of these populations in the United States.

Table 2.7 Top ten specific Write-Ins classified as White or Black responses to the 2010 Census question on race
Full size table

Multiple-Race Reporting Over Time by Age Cohort

Examining age cohorts of selected multiple-race combinations provides interesting insights about race reporting over time. The 1997 OMB standards permitted respondents to report more than one race for the first time. However, it is debatable whether or not the reporting of more than one race meets the fundamental requirements for a useful classification system. The individual may indeed recognize his or her background, but it is unclear whether these groups are recognized by their social networks. Evidence also indicated that the classification into multiple-race groups is not stable over time (Bentley et al. 2003). Finally, it is too soon to determine whether these multiple-race categories are predictive of social or economic opportunity. This analysis examines race reporting by age cohort for the two largest multiple-race populations among the OMB race categories.

In terms of race reporting over time, Table 2.8 shows interesting data for age cohorts for the White and Black population. The data shown are the White and Black population age distribution in 2000 and the counts of these age cohorts in the 2010 Census, by 10-year age groups. For example, among those White and Black, 231,361 people were reported as under 5 years old in 2000 and 296,497 were reported as 10–14 years old in 2010. Therefore, we see that the Census 2000 under 5 years old age cohort grew by 28 % in the past 10 years. While the overall population counts for each age cohort are relatively small, virtually all of the age cohorts experienced substantial growth over the past 10 years. In total, these White and Black age cohorts increased by nearly one-third in the past 10 years.

Table 2.8 White and Black population by age cohort: 2000 and 2010
Full size table

Table 2.9 presents similar statistics for the White and Asian population. The trend in the growth of White and Asian population age cohorts was similar to the growth seen in the White and Black population age cohorts. For example, among those White and Asian, 149,628 people were reported as under 5 years old in 2000 and 196,692 were reported as 10–14 years old in 2010. This age cohort increased by nearly one-third in the past 10 years. Significant growth was exhibited in most of the White and Asian age cohorts, particularly up through the cohort that was 30–34 years old in 2000. Overall, the White and Asian age cohorts increased just over one-quarter in the past 10 years.

Table 2.9 White and Asian population by age cohort: 2000 and 2010
Full size table

The growth in the age cohorts among the White and Black population and the White and Asian population between decennial censuses is not easily explained. Population change over time is attributed to three basic phenomena: births, deaths, and immigration. Since we examined age cohorts over time, births are not a factor. Mortality is low for most of these age groups and in any case deaths would reduce the size of the cohort. We know from the 2007–2009 American Community Survey that about 7 % of the White and Asian population and about 5 % of the White and Black population are foreign born. Therefore, immigration probably had a minimal impact on the growth of these age cohorts.

Shifts in reporting race remains the most likely explanation for the growth of these age cohorts. Much of the significant increase in age cohorts for both the White and Black population and the White and Asian population occurred for those who were young adults in 2010. This could represent those who did not report their own race in Census 2000, as they were children or had just entered adulthood. However, for the 2010 Census, perhaps they reported their own race and chose to report more than one. Increased awareness and acceptance of reporting multiple races since Census 2000 could have significantly impacted the growth of the age cohorts between the decennial censuses. This possibility, that race reporting can vary significantly over time among a population group, questions the fundamental usefulness of multiple-race reporting in the current racial classification system used by the federal government. It is too early to determine whether this is a secular trend (increased reporting over time) or the effect of age (increased reporting as one grows older).

New Directions for Race and Ethnicity Data Collection

The Census Bureau and the federal statistical system face many challenges, including a growing list of groups who find the current race and ethnic classification system confusing, if not irrelevant, or who wish to see their own specific group as a category on the US decennial census questionnaire. The research objectives of the Census Bureau are to design questionnaire items that will increase reporting in the standard race and ethnic categories established by OMB, elicit the reporting of detailed race and ethnic groups, lower item nonresponse, and increase accuracy and reliability (Humes 2009). The Census Bureau engaged in such research most recently in the 2010 Census Alternative Questionnaire Experiment (AQE).

The first and primary component of the AQE was mailout/mailback questionnaires focusing on three areas of research. The first research area includes several features: (1) testing the use of modified examples in the race and Hispanic-origin questions; (2) testing the removal of the term “Negro” from the “Black, African American, or Negro” checkbox response category; and (3) testing the use of a modified instruction that permits multiple responses to the Hispanic-origin question. The second research area focuses on several exploratory approaches to combining the race and Hispanic-origin questions into one item. The third research area focuses on (1) ways to clarify that the detailed Asian checkbox categories and the detailed Native Hawaiian and Other Pacific Islander checkbox categories are part of the two broader OMB race groups, and (2) ways to limit the use of the term “race” in the race question. Additionally, two features from the first research area, testing the removal of the term “Negro” and testing modified examples in the race question, are also tested in this research area.

The second component of the AQE was a telephone reinterview study conducted with a sample of AQE mail respondents. This research assessed the accuracy and the reliability of both the control and the alternative race and Hispanic-origin questionnaires by exploring responses to a number of probing questions.

The third component of the AQE was a series of focus groups that were conducted to complement the quantitative analyses. The focus groups sought to identify the source of response anomalies that emerged from the AQE mailout/mailback questionnaires, as well as to identify trends in race and Hispanic origin reporting, giving us a better understanding of response patterns. Preliminary results from the mailout/mailback, the reinterview, and focus groups are being assessed to determine successful strategies to use during the 2020 Census research program.

We will focus here on the area of research that involves combining the race and ethnicity questions into one, as well as the focus group research, as they best illustrate how the process of adapting federal statistics to a changing and multi-cultural society will continue into the next decade.

A primary purpose of the AQE is to test alternative approaches to combining the Hispanic-origin and race questions into one item. As discussed earlier, although the OMB race classification system works well for many respondents, there are others, particularly those of Hispanic origin, who do not identify with OMB race categories. Thus, the Census Bureau is forced to statistically allocate an increasing number of people to a specific OMB race category when preparing special tabulations for the administration of federal programs. With the projected steady growth of the Hispanic population, the “Some Other Race alone” population is expected to continue increasing since a large proportion of Hispanic-origin ethnicity responses correlated with Some Other Race racial identification. Therefore, in light of the growing “Some Other Race” population in consecutive decennial censuses, new approaches to combining the race and Hispanic-origin questions into one item were tested in the AQE. In all of the combined-question experimental panels, respondents were allowed to mark all responses that applied to them.

For example, the design of one panel represents a “detailed” approach to the combined question. This panel combines the two questions, lists examples for all OMB groups, has write-in areas for each OMB group and “Other,” and retains all of the checkbox response categories on the 2010 Census control panel. A simple instruction is used that directs respondents to mark one or more boxes and to write in a specific race or origin. The use of both of the terms “race” and “origin” was included to represent both of the existing OMB concepts. This version provides an opportunity for all OMB race and ethnic groups to report detailed ethnic information in their own write-in areas—for which many groups have recently lobbied the Census Bureau and Congress.

Another design panel represents a “streamlined” approach to the combined question (see Fig. 2.7). This version also provides examples for all OMB groups, as well as an opportunity for members of all OMB race and ethnic groups to report detailed ethnic information in their own specified write-in areas. This approach removes all national origin and ethnic checkboxes, which simplifies and streamlines the presentation of the combined question. All groups that are national origin checkboxes on the 2010 Census control panel have been added as examples to their respective checkbox response categories. This permits the removal of the individual checkboxes, yet still allows the groups to be listed on the questionnaire in the form of additional examples.

Fig. 2.7
figure 7

2010 Census alternative questionnaire experiment streamlined question

Full size image

Yet another panel represents a “very streamlined” approach to the combined question. This approach also removes all national origin checkboxes, which simplifies and streamlines the question. This panel also brings equity to all OMB race and ethnic groups by providing one shared write-in area for reporting all detailed race and ethnic responses.

Preliminary results from the AQE focus group research support further testing of the combined question strategy and the special design features associated with these experimental panels. Focus groups with members of the Afro-Caribbean community and the Middle Eastern and North African community about the experimental race and Hispanic-origin question panels confirmed that their racial identity is not always consistent with the OMB standards. Many Middle Eastern and North African participants did not identify as “White.” It is interesting to note that many members of most other focus groups representing different races and ethnicities also did not view the Middle Eastern and North African population as “White” and wondered why they were classified as such. Similarly, Afro-Caribbean members of the focus groups confirmed that terminology does matter: many do not identify as “African American.” Focus group participants also questioned why other race/ethnic groups could provide detailed responses (e.g., “Chinese” or “Mexican”) while the Black population and the White population were not given the opportunity to identify their specific heritage. These are some of the racial identification issues that need to be addressed as research plans are developed for the 2020 Census.

The AQE represents the beginning of the 2020 Census content testing. The AQE was designed to assess strategies for improving race and Hispanic-origin reporting (e.g., combined question, multiple response option to the Hispanic-origin question, modified example strategies, etc.), rather than to identify specific question panels to place in the 2020 Census content testing (Hill 2008).

Conclusion

Evidence from the decennial census and the American Community Survey shows that applying the 1997 OMB standards to data collection efforts is becoming increasingly problematic. Since a significant proportion of Hispanic respondents do not identify with any of the five OMB race groups, the “Some Other Race” population is expected to swell for future data collection efforts. Further, the greater the proportion of the US population who do not identify with the OMB race groups in the decennial census, the greater the impact on other federal statistical programs that rely on census data. For example, most federal statistical programs do not include a “Some Other Race” category. In order to meet the requirements of those programs, the Census Bureau must allocate those classified as “Some Other Race alone” to one of the five OMB race groups.

Additionally, a number of population groups dispute the way the OMB standards categorize populations from their area of geographic origin. A number of Middle Eastern and North African community groups are protesting their OMB racial categorization as “White,” while other groups such as the Afro-Caribbean community are seeking recognition on the census questionnaire as a distinct group separate from African Americans. Further, the fluidity of race itself is very evident among those who identify as more than one race, even as race reporting shifts significantly over time.

The results of the AQE provide a basis for future research, analysis, and discussion of the race and Hispanic-origin questions for the 2020 Census. AQE experimental panels produced promising results that could initiate a dialog about the future standards and measurement of race and ethnicity. Any request to open the 1997 OMB standards for review would need to be well rooted in statistical evidence and stakeholder support. Further, any change to the OMB race and ethnicity standards would impact the entire federal statistical system.

All statistical classification systems must by their nature impose a simplification on a more complex reality. Throughout its history, the federal government’s concepts of “race” have always been a mixture of race, color, national origin, and other elements. Chinese, Mexican, and Hindu have all been used as racial concepts, as have “Mulatto” and “Part Hawaiian” (Humes and Hogan 2009). Although one might argue that many of these are not “races,” they largely reflected both society’s and the individual’s concepts of group identity. Group identity is increasingly complex, with society’s view of the individual perhaps differing from his or her own. Does American society treat a Black Spanish-speaking immigrant from the Dominican Republic as Black or Hispanic or both? When this person reports his race as Dominican, is he attempting to identify with a larger Hispanic group, or is his response no different from a French-speaking Haitian who reports his national origin rather than Black? Will their grandchildren be treated by society as Dominican or Haitian or as Black? Will, in a few generations, the distinction between Vietnamese and Korean, in terms of racial classification, be no different than that between Swede and Italian? The authors of this chapter cannot answer these questions, but we know that these are the right questions to ask.