Denver University Economic and Social Inequalities in the United States Questions

1. Massey and Denton focus on the links between residential segregation and economic and social inequalities.  Describe the measures of spatial inequality that Saenz and Morales present?

a. Dissimilarity Index

b. Isolation index

2. What are the differences in earnings inequality Saenz and Morales present?

3.From these differences what can one assume about wealth, home ownership, and intergenerational transmission of wealth?

CHAPTER 6
Demography of Race and Ethnicity
Rogelio Saenz and M. Cristina Morales
One of the most permanent features of many societies, especially the United States, is
racial and ethnic stratification. Many immigrant groups have been integrated into the
different dimensions of American life, while others have remained relatively marginalized. The road toward inclusion is particularly difficult for groups that initially gained
entrance to the United States through involuntary means (e.g., warfare and conquest)
and for those with more pronounced racial and cultural distinctions compared to the
dominant group (McLemore and Romo 1998). Such patterns set apart the experiences
of African Americans, American Indians, Mexican Americans, and Puerto Ricans,
groups that have been labeled as ‘‘colonized groups’’ due to the aggression surrounding
their initial incorporation into the United States, their racial and cultural distinctions,
and their long-term location on the lower rungs of the American social and economic
hierarchy.
The unique experience of these and other minority groups has major implications
for the United States population. Race and ethnicity are important dimensions in
understanding the demography of the United States, for racial and ethnic groups vary
tremendously with respect to population composition, population processes, as well as
their life chances and access to opportunity structures. Referring to the social world of
African Americans, Weeks (2002:411) notes that ‘‘being of black-African origin in the
United States is associated with higher probabilities of death, lower levels of education,
lower levels of occupational status, lower incomes, and higher levels of marital disruption than for the non-Hispanic white population.’’ The inequality of groups in American society along racial and ethnic lines has important implications for the future of the
United States because of the major demographic transformations already underway in
this country. Of the 75.8 million inhabitants that the United States is projected to add to
169
170
Rogelio Saenz and Maria Cristina Morales
its population between 2000 and 2030, non-Hispanic whites should account for less than
one-fifth (18.9%) of the growth (U.S. Bureau of the Census 2000) and comprise less
than half of the total population by 2060.
It is precisely these variations in population change that piqued the interest of
demographers on racial and ethnic matters in the early part of the 20th century. Several
prominent demographers of the time expressed alarm about the rapid growth of poor
nonwhite populations and feared that the quality of the nation’s population would
diminish significantly (Zuberi 2001). However, most of the focus on the demography of
racial and ethnic groups up to the middle of the 20th century was primarily relegated to
the study of African Americans (e.g., Cox 1948; Drake and Cayton 1945; DuBois 1896,
1903, 1909; Frazier 1939, 1949, 1957). It is interesting to note that despite these many
studies, Hauser and Duncan (1959) in The Study of Population devoted only four pages
to race and ethnicity. And a recent inventory (Teachman, Paaselva, and Carvew 1993)
of topics that demographers have addressed in articles published in Demography over
the journal’s first 30 years did not include a category on race and ethnicity.
In an examination we conducted of articles appearing in Demography between 1964
and 2003 (issues 1 to 3), of the 1,676 articles published, 187 were related to race and
ethnicity (including immigration), accounting for approximately 11% of all the articles.
Figure 6.1 shows, however, that most of the interest in the study of race and ethnicity
occurred in the last two decades, with the peak taking place in the 1999 to 2003 period
when about one-fifth of the articles dealt with this topic.
This chapter focuses on the demography of racial and ethnic groups and is broken
down into four sections. First, it examines the conceptualization, substantive concerns,
and relevance of race and ethnicity to demography. Second, it provides an overview of
theoretical perspectives that have been used to understand racial and ethnic groups.
Third, it discusses the methodological issues related to the study of race and ethnicity,
Pct. of Articles on Race/Ethnicity
25.0
19.6
20.0
15.3
15.0
13.9
11.6
10.0
8.6
7.9
7.6
7.0
11.2
5.0
0.0
1964-1968
1974-1978
1969-1973
1984-1988
1979-1983
1994-1998
1989-1993
1964-2003
1999-2003
FIGURE 6.1. Percentage of articles in Demography focusing on race/ethnicity by period, 1964–2003.
Demography of Race and Ethnicity
171
along with key empirical findings. Finally, it focuses on directions for future research
and a few research areas that merit attention. The chapter focuses almost exclusively on
the demography of racial and ethnic groups in the United States.
SUBSTANTIVE CONCERNS
This section provides an overview of the concepts of race and ethnicity and demonstrates the importance of these concepts for demographic research. Specifically, the
discussion focuses on the construction of race and ethnicity and the complexities
associated with defining these concepts. Also illustrated are the distinct experiences of
racial and ethnic groups associated with variations in population composition, population processes, and life chances.
The Construction of Race and Ethnicity
Race and ethnicity are important concepts for demographers. While they are often used
interchangeably, they are distinct terms. The former is associated with physical characteristics, and the latter is related to behavioral or cultural attributes. Despite the
supposed link between physical features and race, race is a social construct, which is
defined by society rather than by genetics (Bonilla-Silva 2001; McFalls 1998).
Nonetheless, race and phenotype continue to be coupled. Historically, in the
United States, the dominant white population identified skin color as the principal
means for sorting people into varying locations on the social and economic hierarchy.
At the outset, persons who resembled the dominant population were labeled as ‘‘us’’
while those who deviated from the ideal white image became the ‘‘other.’’ In the process,
white skin was associated with what was good, and black skin with what was bad.
Even today, the English language serves as a reminder of this distinction. White is
associated with purity and goodness (the ‘‘good guys’’ wear white hats), and black
is associated with impurity and evil (the ‘‘bad guys’’ wear black hats).
In the United States, racism emerged as an ideology to justify the conquest of
American Indians and the enslavement of Africans (Feagin 2001). Whites could live
with themselves if they became convinced that American Indians and African slaves
were subhumans who were mentally and biologically inferior creatures who could
not rule themselves. Even after the hostilities against American Indians and the
enslavement of Africans ended, institutional arrangements were established to maintain
the second-class citizenship of these minorities. These institutional arrangements
include the establishment of reservations and Jim Crow practices.
Although these are the most extreme cases, other groups (e.g., Irish, Southern
and Eastern Europeans, Asians, and Latinos) have been racialized at different times
in the historical past. In the process, these groups have been defined as ‘‘others’’ at one
point or another and have been associated with inferior physical, mental, and moral
attributes in relation to the dominant white population. Similar kinds of social construction have occurred in other societies (with regard to this phenomenon in China, see
Borchigud 1995; Dikotter 1992; Gladney 1994; Khan 1995; with respect to Brazil,
see Skidmore 1995; and with respect to Germany and Italy, see Teitelbaum and
Winter 1998).
172
Rogelio Saenz and Maria Cristina Morales
In its barest form, race and ethnicity can be viewed as ascribed characteristics. At
birth, one is assigned a race and ethnicity based on the attributes of one’s parents. Yet,
racial classification systems associated with the social construction of race and ethnicity
have inconsistencies and ambiguities (Obach 1999). Racial and ethnic categories are
social constructs that vary across place, time, and situations (Eschbach and Gomez
1998; Saenz and Aguirre 1991; Waters 2002; for a discussion of the fluid and situational
ethnic identity of indigenous populations in Mexico, see Tiedje 2002; for a discussion of
the dynamic nature of the Hutu and Tutsi ethnic categories in Burundi, Congo, and
Rwanda, see Longman 1999). As Waters (2002:25) emphasizes, racial and ethnic identities are ‘‘subject to a great deal of flux and change—both intergenerationally, over the
life course, and situationally.’’ For example, groups once considered neither ‘‘white’’
nor ‘‘black,’’ such as the Irish, Lebanese, and Syrians in the United States as well as
Chinese in Mississippi, have gained acceptance as ‘‘whites’’ over time (Gualtieri 2001;
Ignatiev 1995; Warren and Twine 1997).
Immigrants who come to the United States from countries with different racial
classification systems often experience significant alterations in the conception of their
own race and ethnicity. For instance, race is more fluid and malleable in Latin
America than in the United States (Cruz-Janzen 2002; Landale and Oropesa 2002;
Rodriguez 2000). Brazil, for example, has more than 140 racial categories (Rodriguez
2000). Because ‘‘money whitens’’ in many Latin American societies (see de la Cadena
2001; Streicker 1998), ostensibly black individuals who are well off economically may
refer to themselves, and be seen by others, as ‘‘white.’’ When such individuals immigrate to the United States they are faced with the more dichotomous and static notion
of racial classification of the United States (Landale and Oropesa 2002; Rodriguez
2000).
Finally, nationality represents yet another factor that complicates racial and ethnic
identification and further illustrates the fluidity of identity (Waters 1999). For example,
Waters (1999) observes that national identity for West Indian immigrants consists of
multilayered identities including a national origin identity (Trinidadian), a subnational
identity (black), and a supernational regional identity (West Indian). For a discussion of
the construction of ethnicity among Latin American women in Australia, see the work
of Zevallos (2003); for a similar discussion involving ethnic minorities in the borderlands of China, Burma, and Thailand, see the work of Toyota (2003).
White ethnic groups are much more likely to use ethnicity in a voluntary fashion,
having the freedom to reveal or hide it at will. Sociologists (Gans 1979; Waters 1990)
have illustrated the notions of ‘‘voluntary ethnicity’’ and ‘‘symbolic ethnicity’’ among
whites, ethnic forms which represent temporal, ethereal emblems that white ethnics can
don freely. In contrast, minority groups set apart from the mainstream population find
it more difficult to shed their race and ethnicity at will (Nelson and Tienda 1985). In
these instances, particular physical or cultural attributes make it difficult for minority
group members to downplay their race or ethnicity. However, minority group members
may alter their preference for racial or ethnic identities over time. For example, over the
last several decades, the term Negro gave way to the term black, which has increasingly
given way to the term African American. Similarly, persons of Mexican origin have used
a variety of ethnic identities including Mexican, Mexican American, Chicano, Latino,
and Hispanic, with preference for such terms being situational (Saenz and Aguirre
1991). Nonetheless, the complexities of racial and ethnic identification are especially
apparent among people whose parents are from different racial or ethnic groups.
Demography of Race and Ethnicity
173
Multiracial Identities
The last few decades have seen an increase in the prevalence of intermarriage and a
corresponding increase in multiracial persons (Riche 2000). A growing literature has
developed about multiracial individuals and their construction of race and ethnicity
(Chew, Eggebeen, and Uhlenberg 1989; Gatson 2003; Perlmann and Waters 2002; Root,
1992, 1996; Saenz et al. 1995; Spickard 1989; see, for an international perspective,
Christian 2000). Multiracial individuals have numerous options for identifying themselves along racial and ethnic lines. They can select a single identity and discard part of
their background, or they can decide to blend their cultural and racial allegiances to
form a multiracial identity (Snipp 1997b). The numerous options of multiracial individuals are associated with a great degree of fluidity and instability in racial divisions
(Snipp 1997b). Some of the difficulties with the racial classification of multiracial people
include the lack of clarity and logic in distinguishing racial characteristics (Ferrante and
Brown 1999; Glazer 2002; Spickard 1992), the absence of fixed racial boundaries
pointing to the social construction of race (Allman 1996; Ferrante and Brown 1999;
Outlaw 1990; Waters 2002; Zuberi 2001), and tremendous variation in the people
identified as belonging to the same race (Ferrante and Brown 1999; Hummer 1996).
Prior to 2000, U.S. statistical agencies did not recognize multiracial respondents.
Reflecting the increasing presence and voice of multiracial people, the 2000 U.S. census
allowed individuals to select more than one racial group. Although there continues to be
debate about the social and political implications of the identification and enumeration
of multiracial people (Perlmann and Waters 2002) the construction of the multiracial
category is seen by many as a positive move away from the dichotomous and rigid racial
classification system of the United States.
Whiteness and Privilege
While much of the focus of research in race and ethnicity has dealt with minority
groups, some recent work has focused on whites. The ‘‘whiteness’’ literature emphasizes
the extent to which whites gain privileges because of structural arrangements benefiting
them (Bonilla-Silva 2001, 2003; Doane and Bonilla-Silva 2003; Feagin 2001; Feagin and
O’Brien 2003; Feagin and Vera 2000; Frankenberg 1993; Omi and Winant 1984; for a
comparative base involving whiteness in the United States and South Africa, see Steyn
1999). For example, whites are less likely than minority group members to be denied
access to the opportunity structure, to be singled out for suspicious behavior due to the
color of their skin, or to bear psychological wounds resulting from membership in
marginalized minority groups. Jensen (1998), a professor of journalism at the University
of Texas at Austin, provides a personal introspective account of how he has benefited
through white privilege:
But no matter how much I ‘‘fix’’ myself, one thing never changes—I walk through the world
with white privilege. What does that mean? Perhaps most importantly, when I seek admission to a university, apply for a job, or hunt for an apartment, I don’t look threatening.
Almost all of the people evaluating me for those things look like me—they are white. They
see in me a reflection of themselves, and in a racist world that is an advantage. I smile. I am
white. I am one of them. I am not dangerous. Even when I voice critical opinions, I am cut
some slack. After all, I’m white . . . .But, all said, I know I did not get where I am by merit
174
Rogelio Saenz and Maria Cristina Morales
alone. I benefited from, among other things, white privilege. That doesn’t mean that I don’t
deserve my job, or that if I weren’t white I would never have gotten the job. It means simply
that all through my life, I have soaked up benefits for being white. I grew up in fertile farm
country taken by force from non-white indigenous people. I was educated in a well-funded,
virtually all-white public school system in which I learned that white people like me made
this country great. There I also was taught a variety of skills, including how to take
standardized tests written by and for white people. All my life I have been hired for jobs
by white people. I was accepted for graduate school by white people. And I was hired for a
teaching position at the predominantly white University of Texas, which had a white
president, in a college headed by a white dean and in a department with a white chairman
that at the time had one non-white tenured professor . . . .White privilege is not something I
get to decide whether or not I want to keep. Every time I walk into a store at the same time as
a black man and the security guard follows him and leaves me alone to shop, I am benefiting
from white privilege. There is not space here to list all the ways in which white privilege plays
out in our daily lives, but it is clear that I will carry this privilege with me until the day white
supremacy is erased from this society (4C).
A growing literature has also demonstrated the stratification that exists within minority
groups on the basis of skin color. Minority group members are not immune to the racist
images favoring whiteness and often embrace such beliefs (Hall 1994, 1995; Hill 2002;
for a U.S.–Latin American comparative base, see Uhlmann et al. 2002). As such,
lighter-skinned minority group members have been shown to enjoy greater privileges
within their groups as well as gain greater acceptance into the white world (Allen, Telles,
and Hunter 2000; Espino and Franz 2002; Gomez 2000; Hill 2000; Hughes and Hertel
1990; Hunter 2002; Keith and Herring 1991; Murguia and Saenz 2002; Murguia and
Telles 1996; Telles and Murguia 1990).
Racial and Ethnic Variations in Demographic and Socioeconomic
Characteristics
It is clear that racial and ethnic groups have distinct life experiences. These unique
experiences can be seen in their demographic and socioeconomic experiences, particularly their population composition, population processes, and socioeconomic status.
These topics are also applicable for the minority groups of other societies. See, for
instance, the work of Poston and Shu (1987) with regard to China and that of Anderson
and Silver (1989) with regard to Soviet populations.
POPULATION COMPOSITION.
The distinct life experiences of racial and ethnic groups
are associated with variations in their population compositions. Figure 6.2 shows age/
sex pyramids for 12 U.S. racial and ethnic groups: whites, African Americans, American
Indians, six Asian groups (Asian Indians, Chinese, Filipinos, Japanese, Koreans, and
Vietnamese), and three Latino groups (Mexicans, Puerto Ricans, and Cubans). The
shapes of the pyramids clearly display the vastly different demographic and historical
experiences of the 12 groups.
Here are some general patterns. First, Mexicans and Puerto Ricans have the
youngest populations. Nearly one-third of Mexicans (31.8%) and almost three-tenths
of Puerto Ricans (28.9%) are less than 15 years of age. These patterns reflect their
relatively high fertility levels. Second, Japanese and Cubans are older populations. The
elderly account for approximately one-fifth of their populations, a larger percentage
Age Group
Age Group
Age Group
8
3
2
7
2
3
2
7
7
8
2
3
Pct. of U.S. Total Mexican Population
2
U.S. Mexican Population, 2000.
Pct. of U.S. Total FilipinoPopulation
3
U.S. Filipino Population,2000.
7
8
Pct.of U.S. Total African American Population
8
U.S. African American Population, 2000.
3
2
7
3
2
7
Pct. of U.S. Total Puerto Rican Population
0 to 4
7
10 to 14
0 to 4
2
20 to 24
10 to 14
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
30 to 34
3
U.S. Puerto Rican Population, 2000.
Pct. of U.S. Total Japanese Population
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
20 to 24
8
8
U.S. Japanese Population, 2000.
Pct.of U.S. Total American Indian Population
8
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
U.S.American Indian Population, 2000.
7
8
3
2
7
2
3
Pct. of U.S. Total Cuban Population
2
U.S.Cuban Population, 2000.
Pct. of U.S. Total Korean Population
3
U.S. Korean Population,2000.
7
Pct.of U.S. Total Asian IndianPopulation
8
U.S.Asian Indian Population, 2000.
FIGURE 6.2. Age-sex pyramids for 12 selected racial/ethnic groups, 2000. Source: 2000 1% Public Use Microdata Sample (PUMS).
Pct.of U.S. Total Vietnamese Population
0 to 4
0 to 4
7
10 to 14
10 to 14
2
20 to 24
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
20 to 24
3
U.S.Vietnamese Population, 2000.
7
30 to 34
8
3
Pct. ofU.S.Total Chinese Population
0 to 4
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
8
Males
Females
U.S. Chinese Population, 2000.
Pct.of U.S. Total White Population
80 to 84
0 to 4
10 to 14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
0 to 4
10 to14
20 to 24
30 to 34
40 to 44
50 to 54
60 to 64
70 to 74
80 to 84
Age Group
Age Group
Age Group
Age Group
Age Group
Age Group
Age Group
Age Group
Age Group
U.S. White Population, 2000.
8
Demography of Race and Ethnicity
175
176
Rogelio Saenz and Maria Cristina Morales
than that for persons less than 15 years of age (Japanese, 9.5%; Cubans, 14.6%). These
patterns result from low fertility. Third, immigration has played a significant role in the
shaping of the Asian and Latino groups. This is seen in the bulges associated with the
primary working ages surrounding the 25 to 44 age categories. About two-fifths of
Asian Indians (41.8%) and Vietnamese (37.2%) are 25 to 44 years of age. Fourth, the
groups also vary in their sex ratios (number of males per 100 females). Males outnumber
females among Asian Indians (115.0), Mexicans (111.1), Vietnamese (102.9), and Cubans (101.4). In contrast, females outnumber males in the other eight groups, with the
sex ratios being especially low for Japanese (76.2), Koreans (79.9), Filipinos (82.2), and
African Americans (89.9). The wide variability in sex ratios is primarily associated with
sex selectivity in immigration. Immigration from Mexico and India, for example, is
primarily male. Moreover, the formation of ‘‘war bride’’ marriages between Asian
women and American men is reflected in the low sex ratios of the Japanese, Korean,
and Filipino groups. The low sex ratio of African Americans reflects the high mortality
rates of African American males (see below). Indeed, much of the racial and ethnic
variation in age/sex composition reflects distinctions in population processes. See the
more general discussion in this regard in chapter 1 in this Handbook.
FERTILITY.
Racial and ethnic groups differ in their fertility. Data from the National
Center for Health Statistics (NCHS) (Hamilton, Martin, and Sutton 2003) may be used to
examine these distinctions. Unfortunately, current data are not available for the fertility
rates of specific Asian and Latino populations. NCHS presents data for the ‘‘Asian and
Pacific Islander’’ and ‘‘Latino’’ aggregate groups. Figure 6.3 shows the total fertility rate
(TFR) for five racial and ethnic groups in 2001. Latinas have the highest fertility rate,
with a TFR of 2,737, and African American women have the second highest TFR (2,039);
the TFRs are the lowest among American Indian (1,740), Asian and Pacific Islander
(1,835), and non-Hispanic White (1,840) women. Research has also shown racial and
3000
2737
TFR Per 1,000 Women
2500
2039
2000
1840
1740
1835
1500
1000
500
0
NH White
Afr. Amer.
Amer. Ind.
Asn. PI
Latino
FIGURE 6.3. Total fertility rate by race/ethnic group, 2001. Source: Hamilton et al. 2003.
Demography of Race and Ethnicity
177
ethnic variations in fertility in other societies as well. For example, see research examining
the fertility of aboriginal groups (Suwal and Trovato 1998) and Asians (Halli 1987) in
Canada and ethnic groups in the former Soviet Union (Anderson and Silver 1989).
MORTALITY. Racial and ethnic groups also have different mortality patterns. Data
used here are from the NCHS (Arias et al. 2003) and are based on the same five racial
and ethnic groups. Figure 6.4 presents the age-adjusted death rates (AADR) of males
and females in each of the five racial and ethnic groups in 2001. The AADR, which
adjusts for age differences across groups, indicates the number of deaths (per 100,000
persons) that members of a standard population (the United States) would have in a
given year if they experienced the age-specific death rates of a given racial/ethnic group.
A few patterns emerge. First, across all the groups, females consistently have lower
death rates than males, with the rates of females being approximately two-thirds as high
as those of their male counterparts. Second, African Americans have the highest death
rates. The AADR of African American males is 38% higher than that of non-Hispanic
white males, while the AADR of African American women is 30% higher than that of
non-Hispanic white women. Third, the lowest death rates occur among Asian and
Pacific Islanders (AADR of males, 597.4; AADR of females, 412.0), Latinos (802.5;
544.2), and American Indians (798.9; 594.0), each having lower death rates than those of
non-Hispanic whites. For related discussions see chapter 10 in this Handbook.
The death rates of American Indians and Latinos are inconsistent with the relatively low socioeconomic positions of these groups. This is also the case with the
relatively low fertility rates of American Indians. We suspect that the inconsistencies
involving American Indians are related to their diversity and their high rates of intermarriage. Over the last few decades, a major reason for American Indian high growth is
the greater prevalence of people identifying themselves as American Indian (Eschbach
1,600.0
1,393.7
1,400.0
Male
Female
AADR Per 100,000
1,200.0
1,012.8
1,000.0
800.0
925.5
802.5
798.9
713.5
594.0
600.0
597.4
544.2
412.0
400.0
200.0
0.0
NonHsp. White NonHsp. Afr. Am.
Amer. Ind.
Asian Pac. Isl.
Latino
FIGURE 6.4. Age-adjusted death rates for race/ethnic groups by sex, 2001. Source: Arias et al. 2003.
178
Rogelio Saenz and Maria Cristina Morales
1995; Passel 1997; Snipp 1992, 1997a). This occurs among American Indians living on
reservations as well as those who are generations removed from these locales. The
growth may also be due in part to problems of consistency, particularly among the
offspring of the intermarried, in the identification of people in the census and in birth
and death certificates, which could contribute to the unexpected levels of mortality and
fertility described above. Similar reasons have been shown for the rapid growth of
several of the minority nationalities of China (Poston 1993).
In the case of Latinos, there has been much debate and speculation about the cause
of the ‘‘epidemiological paradox’’ involving the low death rates of Latinos (AbraidoLanza et al. 1999; Echevarria and Frisbie 2001; Forbes et al. 2000; Hayes-Bautista 2003;
Hummer et al. 2000; Landale, Oropesa, and Gorman 1999, 2000; Markides and Coreil
1986; Patel et al. 2004). Because the paradox is especially evident among Mexican
immigrants, it has been suggested that the favorable mortality patterns are associated
with strong levels of social support, the selective nature of immigration to the United
States, and the return of immigrants to Mexico when they become seriously ill
(the ‘‘salmon bias’’) (for a challenge to the paradox at the older ages, see Patel et al. 2004).
There are real difficulties associated with developing mortality rates for a mobile
population such as Mexican immigrants. The low death rates of Latinos could well be a
statistical artifact produced by immigrants who return to Mexico when they become
seriously ill; this is especially the case if the death is not recorded in the United States.
Hence, in the computation of the death rate, these individuals would be part of the
denominator (population at risk of dying) but not the numerator (deaths). The problem
is likely to be the most prevalent in the case of infant mortality since, especially along
the U.S.-Mexico border, Mexican women may deliver babies in the United States and
return to Mexico immediately thereafter. In such cases, infant deaths occurring in
Mexico are not recorded in the United States, causing the infant mortality rate of
Latinos to be artificially low. Data from the NCHS (Arias and Smith 2003) indicate
that in 2000, Latinos (5.6) and whites (5.7) had similar infant mortality rates (number of
deaths to babies less than one year of age per 1,000 live births) compared to a rate of 14.1
for African Americans. For mortality research in other societies, see the research of
Ross and Taylor (2002) showing the high mortality rates of indigenous groups in
Australia compared to nonindigenous groups in Australia and indigenous groups
outside of Australia. See also the work of Trovato (2001) examining the mortality
patterns of aboriginal groups in Canada, New Zealand, and the United States.
Internal and International Migration
Racial and ethnic groups also vary with respect to their propensity for migration within
the United States and international migration to the U.S. Data from the 2000 One
Percent PUMS (U.S. Census Bureau 2003) are used to measure the degree of interstate
migration between 1995 and 2000 among persons five years of age and older who were
living in the United States in 1995. The migration information is obtained through the
use of the five-year migration query, which seeks information about where people were
living in 1995 and 2000. Interstate migrants are those individuals who were living in a
different state in 1995 and 2000. Because age is typically associated with migration, and
because the different racial and ethnic groups differ substantially with respect to age
structure, it is necessary to use age-adjusted interstate migration rates (AAIMR) that
Demography of Race and Ethnicity
179
use the age-specific interstate migration rates of each racial and ethnic group and apply
them to the population in the U.S. in 1995. The AAIMR refers to the number of
interstate migrants in 1995 to 2000 per 1,000 persons five years and older living in the
United States in 1995.
Figure 6.5 (dark bars) shows the AAIMRs for the 12 racial and ethnic groups in the
U.S. Asian Indians (137.2) and Koreans (119.2) are the most geographically mobile
groups with more than 1 in 10 persons five and older living in a different state in 1995
and 2000. Three other groups (Chinese, 99.3; American Indians, 97.9; and whites, 93.9)
also show a considerable amount of geographical movement. On the other hand,
Mexicans (56.1), Cubans (67.4), African Americans (71.1), and Filipinos (73.4) were
the least likely to move across state boundaries between 1995 and 2000. These low
AAIMRs of Mexicans and Cubans likely reflect their historical concentration in the
Southwest and in Florida, respectively.
The 2000 PUMS data may also be used to estimate the number of persons from
the different racial and ethnic groups who were living abroad in 1995 and in the United
States in 2000. Abroad migration rates (AMR) indicate the number of persons who
moved from abroad to the United States between 1995 and 2000 per 1,000 persons
living in the United States in 1995. There is a significant amount of variation in the
volume of international migration across the racial and ethnic groups (Figure 6.5, light
bars). Asian Indians, again, have the greatest amount of movement, with 312.2 Asian
Indians coming to the United States between 1995 and 2000 per 1,000 members of the
group already living in the United States in 1995. Three other Asian groups (Japanese,
197.7; Koreans, 176.7; and Chinese, 170.1), along with Mexicans (112.6), have AMRs
greater than 100. In contrast, three groups (American Indians, 10.5; whites, 11.4; and
African Americans, 21.8) have extremely low levels of international movement between
1995 and 2000.
Rate Per 1,000 Persons in U.S. in 1995
350.0
312.2
AAIMR
AMR
300.0
250.0
197.7
200.0
176.7
170.1
137.2
150.0
119.2
97.9
93.9
100.0
99.3
50.0
21.8
11.4
80.3
73.4
71.1
112.6
98.8
96.1
88.7
80.1
74.0
67.4
56.1
98.5
10.5
ex
ic
an
Pu
er
to
R
ic
an
C
ub
an
M
es
e
n
na
m
et
re
a
Vi
Ko
es
e
in
o
Ja
pa
n
lip
Fi
er
.
Am
er
.I
nd
.
As
ia
n
In
d.
C
hi
ne
se
m
r.A
Af
W
hi
te
0.0
FIGURE 6.5. Age-adjusted interstate migration rates (AAIMR) and abroad migration rate (AMR) by race/ethnic
group, 1995–2000. Source: 2000 1% Public Use Microdata Sample (PUMS).
180
Rogelio Saenz and Maria Cristina Morales
Life Chances
The racial and ethnic groups also vary with respect to socioeconomic opportunities and
life chances, specifically, educational attainment and poverty (for an international
comparison base, see the work of Nattrass and Seekings [2001] on blacks and whites
in South Africa). Educational attainment gauges the extent to which members of
different racial and ethnic groups have access to human capital resources that are crucial
for socioeconomic achievement. Poverty provides an indication of the degree to which
people from different racial and ethnic groups lack the minimal economic resources
required to sustain themselves. Because the groups differ significantly with respect to
nativity, the analyses are conducted separately for native- and foreign-born groups.
Note that due to relatively small numbers, foreign-born American Indians and Puerto
Ricans are excluded from the analysis (here and in subsequent analyses below).
Figure 6.6 provides information on the percentage of persons 25 to 44 years of age in
each racial and ethnic group and nativity group who have completed college (the
equivalent of a bachelor’s degree). There is a noticeable amount of variation across
group on these percentages. Overall, with the exception of Vietnamese, Asians have the
highest educational levels, with Asian Indians (native-born, 72.7%; foreign-born, 71.0%)
positioned at the top alongside native-born Chinese (70.1%). Five other Asian groups
have at least half of their members 25 to 44 years of age holding a college degree: nativeborn Koreans (63.6%), foreign-born Chinese (56.6%), foreign-born Japanese (56.6%),
native-born Japanese (54.8%), and foreign-born Koreans (50.2%). Whites do not fare as
well as many Asian groups and native-born Cubans with respect to college completion. In
contrast, foreign-born Mexicans have by far the lowest level of education, with less than 1
in 20 having a college diploma. Furthermore, five other groups have fewer than one in five
of their members 25 to 44 years of age who are college graduates: native-born American
80
72.7
70
Native-Born
71.0 70.1
Foreign-Born
63.6
Pct. College Grads.
60
56.6
50.2
50
45.9
41.8
39.4
40
30
54.8 56.6
38.1
30.2
25.2
25.2
22.4
18.9
20
14.4
13.8
11.9
13.7
10
4.6
ub
an
C
an
R
ic
n
Pu
er
to
e
ex
ic
a
M
am
es
re
an
et
n
Vi
Ko
Am
er
.
Am
er
.I
nd
.
As
ia
n
In
d.
C
hi
ne
se
Fi
lip
in
o
Ja
pa
ne
se
Af
r.
W
hi
te
0
FIGURE 6.6. Percentage of persons 25 to 44 years of age who are college graduates by race/ethnic group and
nativity, 2000. Source: 2000 1% Public Use Microdata Sample (PUMS).
Demography of Race and Ethnicity
181
Indians (11.9%), native-born Puerto Ricans (13.7%), native-born Mexicans (13.8%),
native-born African Americans (14.4%), and native-born Vietnamese (18.9%). These
groups have traditionally been associated with lower levels of socioeconomic status and
are frequently referred to as minorities.
The poverty rates of the racial and ethnic groups may also be analyzed by nativity.
The 2000 census asked individuals in households to report their incomes from all
sources in the last complete calendar year (1999). The income of households and
families was then compared to a poverty threshold based on household size, composition, and presence of children. Those in households or families with incomes below the
poverty threshold were designated as being in poverty in 1999. Figure 6.7 presents
the poverty rates for the 22 groups of interest. Selected Asian groups and whites
have the lowest poverty rates. Fewer than one in a dozen native-born Japanese
(4.6%), foreign-born Filipinos (5.5%), and native-born whites (7.9%) were living in
poverty in 1999. Moreover, five other groups had fewer than one in nine of their
members impoverished: native-born Filipinos (8.0%), native-born Asian Indians
(8.5%), foreign-born Asian Indians (9.2%), native-born Chinese (10.2%), and foreignborn whites (10.4%). In contrast, the six groups with the lowest levels of education, and
which are also viewed as minority groups, had the highest poverty rates: native-born
American Indians (26.0%), foreign-born Mexicans (26.0%), native-born African Americans (25.0%), native-born Puerto Ricans (25.0%), native-born Mexicans (21.6%), and
native-born Vietnamese (18.1%).
In this section the concepts of race and ethnicity have been discussed, and their
importance for demographic analyses has been illustrated. It was shown that racial and
ethnic groups differ significantly with respect to population composition, population
processes, and life chances.
30.0
Native-Born
26.0
25.0
26.0
25.0
Foreign-Born
25.0
Pct. in Poverty
21.6
20.0
18.1
17.7
16.4
15.3
14.3
15.0
15.4
15.0
12.6
10.4
9.2
10.0
10.2
8.5
7.9
12.6
8.0
5.5
5.0
4.6
ub
an
ic
R
C
an
n
ca
ex
i
Pu
er
to
M
es
e
na
m
et
re
an
Vi
Ko
se
pa
ne
e
in
o
Ja
lip
Fi
ne
s
ia
n
C
hi
In
d.
d.
In
As
m
er
.
Am
r.A
Af
W
hi
te
er
.
0.0
FIGURE 6.7. Percentage of persons in poverty by race/ethnic group and nativity, 1999. Source: 2000 1% Public
Use Microdata Sample (PUMS).
182
Rogelio Saenz and Maria Cristina Morales
THEORETICAL ISSUES
This section provides a general overview of the theoretical perspectives that have been
used to explain racial and ethnic variations in demographic, social, and economic
patterns. They may be grouped into two general categories—assimilation and
structural.
Assimilation Perspective
The roots of the assimilationist perspective may be traced to the Chicago School of
Sociology and the work of Robert Park (1950). Milton Gordon (1964), drawing on
Park’s ideas, developed the most popular exposition of the assimilationist perspective.
He emphasized two key aspects of assimilation—cultural and structural. Gordon
viewed assimilation as proceeding across eight subprocesses. The first subprocess is
cultural assimilation (or acculturation), which involves minority group members learning the culture of the majority group. This phase is followed by structural assimilation,
which Gordon distinguished into secondary structural assimilation and primary structural assimilation. Secondary structural assimilation involves members of the minority
group coming into contact with majority group members in impersonal relationships in
institutional and organizational settings. Primary structural assimilation involves the
establishment of warm interpersonal relationships between minority and majority group
members in the form of friendship groups.
For Gordon, primary structural assimilation represents the most crucial assimilation subprocess. Once this stage takes place, subsequent assimilation subprocesses are
expected to ensue automatically (McLemore and Romo 1998). These subprocesses
include marital assimilation (amalgamation), identificational assimilation (ethnic identification), attitudinal receptional assimilation (absence of prejudice), behavioral receptional assimilation (absence of discrimination), and civic assimilation (absence of value
or power conflicts). Gordon’s assimilationist perspective suggests that once minority
group members achieve primary structural assimilation, they are likely to intermarry
with members of the majority group, to shed their ethnic identities in favor of an
American identity, to be less likely to experience prejudice and discrimination, and to
hold universal—as opposed to particularistic—values and interests.
Because the assimilationist perspective is primarily based on the experiences of
European immigrant groups, critics of this approach have questioned the extent to
which the perspective applies to non-European groups. It generally took approximately
three generations for European groups to be integrated into American society (McLemore and Romo 1998). However, some minority groups, such as African Americans,
Mexican Americans, and Puerto Ricans, have been in this country for generations and
have not been integrated. It has also been pointed out that ambiguity and controversy
surrounding the concept of assimilation reinforces oppression (Yinger 1994). For
instance, Alba (1999:9) argues that ‘‘today, assimilation is often depicted in terms of
demands that minority individuals abandon their native cultures to accept the majority
one, a demand that can be viewed as placing them in a position of inferiority and
disadvantage.’’ Therefore, the assimilationist perspective privileges the customs and
values of some groups who emulate the majority group while devaluing the experiences
of other groups that have not easily been integrated (Landsman and Katkin 1998).
Demography of Race and Ethnicity
183
MODIFICATION TO THE ASSIMILATION PERSPECTIVE.
Despite critiques directed
against it, the assimilationist perspective has not been abandoned, but has been modified (see Yinger 1994). For example, sociologists have recently paid an increasing
amount of attention to the sons and daughters of immigrants—the second-generation—and have developed important insights that have helped better understand the
assimilation process. In particular, Portes and his colleagues expanded the assimilation
perspective in their development of the segmented assimilation perspective (Portes 1995;
Portes and MacLeod 1996; Portes and Rumbaut 1996, 2001; Portes and Zhou 1993).
This perspective suggests that immigrants today may be assimilated into one of three
possible paths: (1) acculturation and subsequent assimilation into the white middle
class, (2) assimilation into the underclass, which is marked by permanent poverty, and
(3) the preservation of solidarity within the immigrant community, which promotes
economic mobility. Portes and his associates (Portes and MacLeod 1996; Portes and
Rumbaut 1996; Portes and Zhou 1993) argue that the route that immigrants take
depends on their access to resources within their families and communities. It has also
been suggested that the downward mobility path is associated with skin color, concentration in central cities, and the general lack of access to the opportunity structure in
local labor markets (Landale, Oropesa, and Llanes 1998).
The assimilationist perspective has been used to examine racial and ethnic group
differences in a variety of demographic patterns including fertility (Bean, Swicegood,
and Berg 2000; Carter 2000; Pagnini 1997; Stephen and Bean 1992; Ford 1990),
migration (Fang and Brown 1999), residential segregation (Alba and Logan 1993;
Gross and Massey 1991; Logan, Alba, and Leung 1996; White and Omer 1997),
marriage and divorce (Bean, Berg, and Van Hook 1996), and intermarriage (Hwang,
Saenz, and Aguirre 1997; Kulczycki and Lobo 2002; Qian and Lichter 2001). It has also
been shown to have utility for understanding the fertility of minorities in other societies.
See, for instance, Chang’s (2003) work dealing with the major minorities in China.
Structural Perspectives
Structural perspectives attribute the demographic, social, and economic standing of
minority groups to macro-level or societal-level phenomena. In this respect, forces
beyond the individual are shown to be primarily responsible for the observed demographic and socioeconomic patterns of minority groups. The structural perspectives
that sociologists have developed to understand the stratification of minority groups are
particularly useful. These include Blauner’s internal colonialism model, Blalock’s group
size perspective, Blau’s structural perspective, Wilson’s structural perspective, Massey
and Denton’s structural perspective on segregation and poverty, and Bonilla-Silva’s
racialized social system perspective. These orientations generally seek to identify the
forces and mechanisms that keep minority groups located at the bottom of the social
and economic ladder of society.
The 1960s and 1970s represent a critical period in the United States for the
development of scholarship related to the social and economic standing of minority
groups. Blauner (1969, 1972) developed the internal colonialism model, which reflected
the spirit of the era. Blauner applies ideas involving the colonization of Third-World
people at the hands of European colonizers to the experiences of minority groups that
have been colonized in the United States, i.e., groups that were originally incorporated
184
Rogelio Saenz and Maria Cristina Morales
through aggression. These groups include American Indians (conquest), African Americans (slavery), Mexican Americans (warfare), and Puerto Ricans (warfare). According
to Blauner, these groups exist in internal colonies in the United States. Moreover, they
are subjugated through the mechanisms of racism, cultural oppression, and lack of selfadministration. Institutional forces are in place to keep colonized minorities within the
confines of their colonies where they live as second-class citizens.
During the same period, Blalock (1967) developed the relative group size perspective. He reasoned that minority groups experience higher levels of discrimination and
inequality in areas where they comprise a larger share of the population. In such
instances, the minority group represents a threat to the existing power structure, with
the majority group erecting barriers and obstacles to ensure that the minority group
does not become upwardly mobile and gain access to power. The relative group size
perspective has received a considerable amount of empirical support with respect to
earnings (Frisbie and Neidert 1977; Tienda and Lii 1987), inequality (Fossett and
Siebert 1997), poverty (Saenz 1997; Swanson et al. 1994), and interracial attitudes
(Fossett and Kiecolt 1989).
Blau provided important insights for the study of the demography of racial and
ethnic groups with his theoretical developments pertaining to the links between population structure and intergroup relations (Blau 1977, 1994). His structural contexts of
opportunities perspective describe how macrostructural forces influence and constrain
peoples’ associations and choices (1994). Blau applied this macrosociological perspective
to demonstrate that individual choices such as selecting a marriage mate and occupation
are highly constrained and reflect the social structure. In the case of spouse selection,
for example, even though people may prefer to marry an in-group member, structural
constraints often prevent them from realizing their desires. One of the primary
structural constraints is the size of one’s group. Minority groups tend to have higher inmarriage rates in places where their own group is large and have higher out-marriage
rates in areas where their group is small. A significant amount of research on intermarriage has provided support for Blau’s perspective (Anderson and Saenz 1994; Blau and
Schwartz 1997; Cready and Saenz 1997; Hwang, Saenz, and Aguirre 1997; Kalmijn 1998).
Wilson has also influenced the study of the demography of minorities. He set off
considerable discussion and debate following the publication of his book The Declining
Significance of Race: Blacks and Changing American Institutions (Wilson 1978), in which
he argued that structural changes in the economy, along with civil rights legislation, led
to a declining significance of race and an increasing significance of class. Wilson
observed that the massive loss of manufacturing jobs between World War II and the
late 1970s resulted in the loss of jobs that African Americans had relied on to achieve
some degree of upward mobility. He also asserted that only a segment of the African
American population was able to take advantage of opportunities that came about
through the advent of civil rights legislation. Wilson observed that there was a bifurcation in the African American community with the increasing separation of the middle
class and the ‘‘underclass.’’ He noted that middle-class African Americans by and large
moved to suburbs, leaving behind poor African Americans in central cities. Although
Wilson did not deny that race was still a factor in the low socioeconomic standing of
African Americans, he saw class as becoming a particularly important factor in explaining the condition of African Americans.
Wilson’s (1978) thesis on the declining significance of race stirred a massive amount
of debate and discussion (see also Wilson 1996). His ideas have found some support.
Demography of Race and Ethnicity
185
For example, Sakamoto and his colleagues (2000), using data from the Integrated
Public Use Microdata Sample (IPUMS), observe that the relationship between race
and earnings declined over time for all groups except Latinos. Other research has also
found links between social class and occupational attainment (Hout 1984; Sakamoto
and Tzeng 1999). However, others have challenged Wilson’s assertion that the significance of race is declining (Herring 1989; Horton 1995; Morris 1996; Pattillo 1999; Willie
1978, 1989; Wilson 2000). For example, Feagin and Sikes (1994) observe that middleclass African Americans report a significant amount of racial discrimination even in
corporate America. And Oliver and Shapiro (1995) clearly document the historical
legacies of racism in their focus on wealth (accumulated assets and debts) as opposed
to current income. Their study shows that slightly more than three-fifths of African
American households do not have any wealth, twice the rate of their white counterparts.
They show massive racial gaps in wealth across social classes as well.
Massey and Denton (1993) provide another structural perspective for explaining the
social and economic plight of African Americans. Their primary focus is on the links
between residential segregation and the elevated poverty rates among African Americans.
Massey and Denton assert that the residential segregation of African Americans is part of
a well-conceived plan, deeply embedded in the structure of American society, to keep
African Americans away from whites, with federal policies, banking institutions, and the
real estate industry helping to maintain this condition. The segregation of African
Americans results in the absence of links to the opportunity structure. African Americans
living in ghettoes lack access to most amenities that Americans take for granted. The
following passage illustrates Massey and Denton’s (1993:217) thoughts on the roots and
consequences of the well-entrenched residential segregation of African Americans.
After persisting for more than fifty years, the black ghetto will not be dismantled by passing
a few amendments to existing laws or by implementing a smattering of bureaucratic reforms.
The ghetto is part and parcel of modern American society; it was manufactured by whites
earlier in the century to isolate and control growing urban black populations, and it is
maintained today by a set of institutions, attitudes, and practices that are deeply embedded
in the structure of American life. Indeed, as conditions in the ghetto have worsened and as
poor blacks have adapted socially and culturally to this deteriorating environment, the
ghetto has assumed even greater importance as an institutional tool for isolating the byproducts of racial oppression: crime, drugs, violence, illiteracy, poverty, despair, and their
growing economic costs.
There are theoretical developments emerging directly from the race and ethnicity
literature that have important implications for the study of the demography of race and
ethnicity. Of particular importance is Bonilla-Silva’s (1997, 2001, 2003) racialized social
system perspective. He challenges prevailing thinking that the continued inequality
between minority and majority groups in the United States is due to prejudice and
related phenomena centered at the individual level and that racial and ethnic relations
have improved significantly over the last several decades. Bonilla-Silva emphasizes the
structural interpretations of racism and the new racism that has emerged in the postCivil Rights era. He asserts that racism is entrenched in the structure of American
society and affects all segments of life. Whites benefit from their position in the
stratification system and thus, consciously as well as unconsciously, exhibit behaviors
and attitudes that support the existing stratification system. One important aspect of his
work involves the use of a multimethod, triangulation approach, utilizing both traditional survey methods and in-depth interviews, to get at whites’ ‘‘true, more innermost’’
186
Rogelio Saenz and Maria Cristina Morales
feelings concerning race relations. Bonilla-Silva’s research suggests that whites have
developed a sophisticated and elusive language to mask their true feelings regarding
African Americans in the post-Civil Rights era.
METHODS, MEASURES, AND EMPIRICAL
FINDINGS
This section covers methods and measures related to racial and ethnic groups and brings
in, where appropriate, empirical findings. In particular, it discusses the data, measures,
and analytical procedures used to study racial and ethnic groups.
Data Issues
As noted earlier, race and ethnicity are social constructs that shift over time, space, the
life course, and with altering situations. This conception of race and ethnicity stands in
sharp contrast to how demographers use race and ethnicity in their research (Waters
2002). Their statistical models, for instance, examine variations in race/ethnicity across
groups rather than within individuals. Furthermore, the lack of attention to the social
construction of race and ethnicity affects population projections about the future racial
and ethnic composition of any population. Indeed, population projections assume that
racial and ethnic boundaries and patterns of identification remain static into the future
(Bean et al. 1997; Hirschman 2002; Perlmann 2002; Waters 2002; for a discussion of
population projections involving Jews and Arabs in Jerusalem, see DellaPergola 2001).
This conventional population projection approach also assumes that people belong
to a single racial or ethnic group (Bean et al. 1997), that the current meaning of racial
and ethnic categories will continue to be meaningful in the future (Alba 1999), and that
the impact of intermarriage is not significant enough to have one generation change
categories over time (Alba 1999; Edmonston, Lee, and Passel 2002; Hirschman 2002;
Waters 2002). Although it is difficult to project the significance and stability of racial
and ethnic groups into the future, it is well known that a segment of the population
includes multiracial individuals and that some racial and ethnic groups have relatively
high intermarriage rates (see below). These are issues that demographers will need to
address in their presentation and interpretation of population projections.
There are other racial and ethnic classification issues worth mentioning. For
example, in the collection of census data one individual usually provides information
on all household members. While this practice is likely to produce fairly accurate data in
many instances, there is the potential for inaccuracies in households composed of
unrelated individuals. In addition, there may also be differences in how children from
intermarried couples are classified racially and ethnically by their parents.
The accuracy in the racial and ethnic classification of individuals is a particular
problem in mortality research. Death certificates, which represent the data base for
much mortality research, may be filled out by people who did not know the deceased.
Hospital personnel and funeral directors may inaccurately identify the deceased in terms
of their race or ethnicity (Farley 1996). Similar problems emerge in the study of fertility
when the father of the infant is not known or reported. Differences are also likely to
exist in how infants from intermarried couples are designated with respect to their race
and ethnicity.
Demography of Race and Ethnicity
187
There are other methodological problems related to the measurement of demographic phenomena. For example, in mortality research, death rates assume that the
numerator (deaths) and the denominator (population at risk) refer to the same population. However, the numbers used to generate these rates usually come from different
data sources: vital statistics records for deaths (the numerators) and census data or
population estimates for the population at risk (the denominators). This conventional
approach is problematic when race and ethnicity are fluid or when people from certain
racial and ethnic groups are highly geographically mobile. Sullivan and her colleagues
(1984) describe the problems associated with estimating mortality rates for the Mexican
American population. Such problems may also be associated with the epidemiological
paradox among Latinos. Although this problem has been discussed primarily with
respect to the computation of infant mortality rates, it is likely that the problem also
affects the computation of fertility rates when immigrant women come to the United
States to deliver their babies and then return to Mexico. In such instances, the births are
in the numerator, but the mothers are not in the denominator. This may result in the
fertility rates of Mexican-origin women being artificially inflated.
Measures Associated with Race/Ethnicity and Inequality
There are a number of measures of race and ethnicity and inequality. Data from the
2000 One Percent PUMS and related sources are used to illustrate these measures. The
approaches for identifying the racial and ethnic membership of individuals are first
presented.
RACE AND ETHNIC GROUP IDENTIFICATION.
Most of the data associated with the
classification of people along racial and ethnic lines are driven by governmental decisions (Office of Management and Budget [OMB]) about the appropriate categories to
be used in generating the data. (For an excellent overview of the historical measurement
of race in the U.S. census, see Snipp [2003] and Lee [1993].) Recently the Office of
Management and Budget (1997; see also OMB 1977) issued a revision of the racial
categories on which U.S. federal statistics are to be collected. This mandate established
six categories: white; Black or African American; American Indian or Alaskan Native;
Asian; Native Hawaiian or other Pacific Islander; and Hispanic or Latino. Hispanics/
Latinos are considered to be an ethnic group, with members belonging to any race. As
such, two questions are used to obtain information on racial identification and Hispanic/Latino identification. The Office of Management and Budget (2000) also mandated that individuals be allowed to select two or more racial categories.
Table 6.1 (Panel A) shows the distribution of the U.S. population across the six
OMB-designated categories as reported in the 2000 census (Grieco and Cassidy 2001).
Of the 281.4 million counted in the 2000 census, non-Hispanic whites represent the
largest group. Of those who reported only one race, approximately 69% are nonHispanic whites. Latinos and non-Hispanic African Americans each accounted for
about 12% of the national population, with non-Hispanic Asians making up slightly
less than 4%. Multiracial individuals (persons who reported more than one race)
accounted for 2.4% (for a total of 6.8 million people including 4.6 non-Hispanics and
2.2 Hispanics) of the overall U.S. population.
188
Rogelio Saenz and Maria Cristina Morales
TABLE 6.1. Population of the United States by Racial/Ethnic Group and Hispanic/Latino Population
by Racial Classification, 2000.
Panel A. U.S. Population
Racial/Ethnic Group
Population
% U.S. Pop.
Non-Hispanic/Latino
White alone
Black or African American alone
American Indian and Alaskan Native alone
Asian alone
Native Hawaiian and other Pacific Islander alone
Some other race alone
Two or more races
Hispanic/Latino
Total
194,552,774
33,947,837
2,068,883
10,123,169
353,509
467,770
4,602,146
35,305,818
281,421,906
69.1
12.1
0.7
3.6
0.1
0.2
1.6
12.5
100.0
Panel B. U.S. Hispanic/Latino Population by Race
Race
Population
% Hsp. or Lat. Pop.
White alone
Black or African American alone
American Indian and Alaskan Native alone
Asian alone
Native Hawaiian and other Pacific Islander alone
Some other race alone
Two or more races
Total Hispanic/Latino population
16,907,852
710,353
407,073
119,829
45,326
14,891,303
2,224,082
35,305,818
47.9
2.0
1.2
0.3
0.1
42.2
6.3
100.0
Source: Grieco and Cassidy (2001).
Table 6.1 (Panel B) reports the distribution of Latinos by race. Close to half
(47.9%) of Latinos classified themselves as white, and more than two-fifths (42.2%)
classified themselves as ‘‘other.’’ There is a significant amount of variation in the racial
classification of Latino subgroups. Table 6.2 shows the racial distribution of Latinos in
the three largest Latino subgroups (Mexicans, Puerto Ricans, and Cubans). Cubans
stand out with respect to their preference for the white racial identity with nearly 85%
identifying themselves as white compared to slightly less than half of Mexicans (47.3%)
and Puerto Ricans (47.0%). Mexicans (45.4%) and Puerto Ricans (38.4%) are more
likely than Cubans (7.2%) to choose the ‘‘other’’ racial category. Yet, a noticeable
percentage of Puerto Ricans (5.8%) and Cubans (3.6%) identify themselves as black
TABLE 6.2. Percentage Distribution of Latinos by Race for Selected Ethnic Groups, 2000.
Race
White alone
Black or African American alone
American Indian and Alaskan Native alone
Asian alone
Native Hawaiian and other Pacific Islander alone
Some other race alone
Two or more races
Source: 2000 1% PUMS.
Mexican
Puerto Rican
Cuban
47.3
0.7
1.1
0.2
0.1
45.4
5.2
47.0
5.8
0.5
0.4
0.2
38.4
7.8
84.5
3.6
0.1
0.2
0.1
7.2
4.3
Demography of Race and Ethnicity
189
compared to a smaller percentage of Mexicans (0.7%). Finally, Puerto Ricans (7.8%) are
more likely than Mexicans (5.2%) and Cubans (4.3%) to be multiracial.
Data from the 2000 One Percent PUMS are used to examine variation in the
presence of multiracial people across racial and ethnic groups. To conduct this exercise,
for the non-Hispanic racial groups, combined racial groups have been established which
consist of individuals who chose a given race regardless of how many races they select.
For example, the combined white group consists of individuals who report they were
only white, as well as multiracial individuals whose race classification included white.
Figure 6.8 shows the variation across groups regarding the prevalence of multiracial
individuals. More than three-fourths of non-Hispanics who chose the ‘‘Other’’ racial
category (77.0%) are multiracial, as are slightly more than half of Native Hawaiians and
Other Pacific Islanders (51.3%) and American Indians and Alaskan Natives (43.3%).
Asians (13.0%) represent the only other group with more than one-tenth of group
members reporting multiple races. On the other hand, non-Hispanic whites (2.0%)
and blacks (4.3%) have the lowest percentages of their members reporting two or
more racial categories.
OTHER RACE AND ETHNICITY MEASURES: NATIVITY AND CITIZENSHIP STATUS.
There
is another dimension of race and ethnicity, namely, nativity and naturalization citizenship status among the foreign-born. Data from the 2000 One Percent PUMS are used to
develop actual measures for 12 groups—whites, African Americans, American Indians,
six groups of Asians (Asian Indians, Chinese, Filipinos, Japanese, Koreans, and Vietnamese), and three groups of Latinos (Mexicans, Puerto Ricans, and Cubans). Note
that the first nine groups are non-Hispanic (for the sake of simplicity, the non-Hispanic
designator will not be used). These racial and ethnic groups are the focus of analyses in
the remainder of the chapter, except where data are not available for specific groups.
90.0
77.1
80.0
Pct. Multiracial
70.0
60.0
51.3
50.0
43.3
40.0
30.0
20.0
13.0
10.0
2.0
5.2
4.3
7.8
4.3
0.0
White
Afr. Amer. Amer. Ind. &
AK Native
Asian
Nat. HI& Oth.
Pac. Isl.
Other
Mexican Puerto Rican
Cuban
FIGURE 6.8. Percent of persons who are multiracial by racial/ethnic groups, 2000. Source: 2000 1% Public Use
Microdata Sample (PUMS).
190
Rogelio Saenz and Maria Cristina Morales
The 2000 census asked individuals to report their state and country of birth. Based
on these data, persons may be classified into the native-born (including those born
abroad to American citizens) and the foreign-born. Using these classifications, the
percentages of members of each of the 12 groups who are foreign-born are computed
(see Table 6.3). Six Asian and one Latino group have upward of two-thirds of their
members who are foreign-born: Koreans (77.4%), Vietnamese (76.4%), Asian Indians
(75.7%), Chinese (70.5%), Filipinos (69.0%), and Cubans (68.9%). Approximately twofifths of Mexicans (41.3%) and Japanese (40.2%) were born outside of the United States.
In contrast, relatively few African Americans (5.5%), whites (3.5%), Puerto Ricans
(1.4%), and American Indians (1.0%) are foreign-born.
The 2000 One Percent PUMS data also permit the computation of naturalization
rates for the foreign-born members of each of the 12 groups. The naturalization rate
refers to the percentage of foreign-born individuals who have become U.S. naturalized
citizens. Six groups have naturalization rates above 50%: Filipinos (62.2%), Cubans
(60.6%), Vietnamese (58.9%), whites (54.8%), Chinese (52.5%), and Koreans (51.0%).
On the other hand, Mexicans (22.1%) and Japanese (25.3%) have the lowest naturalization rates.
OTHER RACE AND ETHNICITY MEASURES: LANGUAGE AND INTERMARRIAGE.
The final
two dimensions of race and ethnicity to be examined are language (an indicator of
acculturation) and intermarriage (an indicator of assimilation). The 2000 census asked
individuals who were at least five years of age two questions related to language,
namely, whether a language other than English was spoken at home and, for these
individuals, their level of fluency in English. Using this information in the 2000 One
Percent PUMS, respondents may be categorized into three categories: (1) monolingual
English speakers (those speaking English at home); (2) bilingual (those speaking a
language other than English at home and who speak English ‘‘well’’ or ‘‘very well’’);
and (3) monolingual non-English speakers (those speaking a language other than
English at home and who speak English ‘‘not well’’ or ‘‘not at all’’). This is not a perfect
measure of language abilities. Indeed, individuals who are truly bilingual are classified
as monolingual English speakers if they speak only English at home. In addition,
respondents may inaccurately assess the language abilities of other household members.
Because language abilities are likely to be related to age and nativity, analyses are
restricted to persons 25 to 44 years of age and conducted separately for the native- and
foreign-born groups. Note that because of small numbers of foreign-born American
Indians and Puerto Ricans, they are not included in subsequent analyses involving the
foreign-born. Table 6.3 shows the percentage distribution of native- and foreign-born
individuals in the 12 groups across the three language categories. Among the nativeborn groups, with the exception of Vietnamese and members of the three Latino groups,
most people in the other groups are likely to be monolingual English speakers. Five
groups have more than three-fourths of their members speaking English at home:
African Americans (96.8%), whites (96.6%), Japanese (89.6%), Filipinos (75.5%), and
American Indians (75.2%). However, the greatest portion of Puerto Ricans (71.0%),
Cubans (68.1%), Mexicans (56.9%), and Vietnamese (48.7%) are bilingual speakers and
also show the greatest tendency to be monolingual non-English speakers. Among
foreign-born persons, except for Mexicans, most members are bilingual speakers with
percentages ranging from 59.3% among whites to 84.8% among Filipinos. On the other
hand, the majority (51.7%) of Mexicans are monolingual Spanish speakers.
8.8
3.1
6.7
4.7
96.8
2.8
0.4
51.4
43
5.6
96.6
3.2
0.3
31.9
59.3
8.8
N/A
N/A
N/A
N/A
5.5
44.6
African
Amer.
3.5
54.8
White
47.2
45.1
N/A
N/A
75.2
23.5
1.3
53.6
34.3
12.1
1
44.8
Amer.
Indian
3.9
9.2
4.3
3.6
12.7
30.4
*
*
*
*
*
75.5
23.1
1.4
12.5
84.8
2.8
69
62.2
Filipino
61.6
36.4
2
6.2
71.4
22.4
70.5
52.5
Chinese
*
55
43.5
1.5
12.8
82.5
4.8
75.7
38.7
Asian
Indian
8.6
33
*
*
89.6
9.7
0.7
9
66.9
24.1
40.2
25.3
Japanese
5.8
19.2
*
*
67.4
30.4
2.2
11.4
63.7
24.9
77.4
51
Korean
2.5
7.4
*
*
47.1
48.7
4.2
4.5
66.2
29.3
76.4
58.9
Vietnamese
23.4
23.9
4.6
3.2
40
56.9
3.2
5.3
43
51.7
41.3
22.1
Mexican
20.2
20.3
N/A
N/A
20.6
71
8.4
11.5
58.2
30.3
1.4
38.8
Puerto
Rican
41.2
36.7
9.8
6.2
31
68.1
0.9
5.2
66.1
28.7
68.9
60.6
Cuban
Source: 2000 1% PUMS.
Note: NB¼native-born; FB¼foreign-born.
*
Because of small sample sizes, the percentages for the six NB groups have been aggregated into NB Asian: 25.2% of Asian NB males are married to white females; 39.5% of NB Asian
females are married to white males.
% Foreign-born
% of FB naturalized citizens
Language patterns (Pop. 25– 44)
% NB English
% NB bilingual
% NB native language
% FB English
% FB bilingual
% FB native language
Intermarriage with whites (