Olvera Street is a Los Angeles icon—a thriving Mexican market filled with colorful souvenirs, restaurants and remnants of the oldest buildings in Los Angeles. But though the bright tourist destination teems with visitors, few realize it was once the site of a terrifying raid.
In 1931, police officers grabbed Mexican-Americans in the area, many of them U.S. citizens, and shoved them into waiting vans. Immigration agents blocked exits and arrested around 400 people, who were then deported to Mexico, regardless of their citizenship or immigration status.
The raid was just one incident in a long history of discrimination against Latino people in the United States. Since the 1840s, anti-Latino prejudice has led to illegal deportations, school segregation and even lynching—often-forgotten events that echo the civil-rights violations of African-Americans in the Jim Crow-era South.
The story of Latino-American discrimination largely begins in 1848, when the United States won the Mexican-American War. The Treaty of Guadalupe Hidalgo, which marked the war’s end, granted 55 percent of Mexican territory to the United States. With that land came new citizens. The Mexicans who decided to stay in what was now U.S. territory were granted citizenship and the country gained a considerable Mexican-American population.
As the 19th century wore on, political events in Mexico made emigration to the United States popular. This was welcome news to American employers like the Southern Pacific Railroad, which desperately needed cheap labor to help build new tracks. The railroad and other companies flouted existing immigration laws that banned importing contracted labor and sent recruiters into Mexico to convince Mexicans to emigrate.
Anti-Latino sentiment grew along with immigration. Latinos were barred entry into Anglo establishments and segregated into urban barrios in poor areas. Though Latinos were critical to the U.S. economy and often were American citizens, everything from their language to the color of their skin to their countries of origin could be used as a pretext for discrimination. Anglo-Americans treated them as a foreign underclass and perpetuated stereotypes that those who spoke Spanish were lazy, stupid and undeserving. In some cases, that prejudice turned fatal.
According to historians William D. Carrigan and Clive Webb, mob violence against Spanish-speaking people was common in the late 19th and early 20th centuries. They estimate that the number of Latinos killed by mobs reach well into the thousands, though definitive documentation only exists for 547 cases.
The violence began during California’s Gold Rush just after California became part of the United States. At the time, white miners begrudged former Mexicans a share of the wealth yielded by Californian mines—and sometimes enacted vigilante justice. In 1851, for example, a mob of vigilantes accused Josefa Segovia of murdering a white man. After a fake trial, they marched her through the streets and lynched her. Over 2,000 men gathered to watch, shouting racial slurs. Others were attacked on suspicion of fraternizing with white women or insulting white people.
Even children became the victims of this violence. In 1911, a mob of over 100 people hanged a 14-year-old boy, Antonio Gómez, after he was arrested for murder. Rather than let him serve time in jail, townspeople lynched him and dragged his body through the streets of Thorndale, Texas.
These and other horrific acts of cruelty lasted until the 1920s, when the Mexican government began pressuring the United States to stop the violence. But though mob brutality eventually quelled, hatred of Spanish-speaking Americans did not.
In the late 1920s, anti-Mexican sentiment spiked as the Great Depression began. As the stock market tanked and unemployment grew, Anglo-Americans accused Mexicans and other foreigners of stealing American jobs. Mexican-Americans were discouraged and even forbidden from accepting charitable aid.
As fears about jobs and the economy spread, the United States forcibly removed up to 2 million people of Mexican descent from the country—up to 60 percent of whom were American citizens.
Euphemistically referred to as “repatriations,” the removals were anything but voluntary. Sometimes, private employers drove their employees to the border and kicked them out. In other cases, local governments cut off relief, raided gathering places or offered free train fare to Mexico. Colorado even ordered all of its “Mexicans”—in reality, anyone who spoke Spanish or seemed to be of Latin descent—to leave the state in 1936 and blockaded its southern border to keep people from leaving. Though no formal decree was ever issued by immigration authorities, INS officials deported about 82,000 people during the period.
The impact on Spanish-speaking communities was devastating. Some light-skinned Mexican-Americans attempted to pass themselves off as Spanish, not Mexican, in an attempt to evade enforcement. People with disabilities and active illnesses were removed from hospitals and dumped at the border. As one victim of “repatriation” told Raymond Rodriguez, who wrote a history of the period, Decade of Betrayal, “They might as well have sent us to Mars.”
Others, like Rodriguez’s father, did not wait for raids or enforcement and returned to Mexico independently to escape discrimination and the fear of removal. His wife refused to accompany him and the family never saw him again.
When deportations finally ended around 1936, up to 2 million Mexican-Americans had been “repatriated.” (Because many of the repatriation attempts were informal or conducted by private companies, it is nearly impossible to quantify the exact number of people who were deported.) Around one third of Los Angeles’ Mexican population left the country, as did a third of Texas’ Mexican-born population. Though both the state of California and the city of Los Angeles apologized for repatriation in the early 2000s, the deportations have largely faded from public memory.
Another little-remembered facet of anti-Latino discrimination in the United States is school segregation. Unlike the South, which had explicit laws barring African-American children from white schools, segregation was not enshrined in the laws of the southwestern United States. Nevertheless, Latino people were excluded from restaurants, movie theaters and schools.
Latino students were expected to attend separate “Mexican schools” throughout the southwest beginning in the 1870s. At first, the schools were set up to serve the children of Spanish-speaking laborers at rural ranches. Soon, they spread into cities, too.
By the 1940s, as many as 80 percent of Latino children in places like Orange County, California attended separate schools. Among them was Sylvia Mendez, a young girl who was turned away from an all-white school in the county. Instead of going to the pristine, well-appointed 17th Street Elementary, she was told to attend Hoover Elementary—a dilapidated, two-room shack.
The bare-bones facilities offered to students like Mendez lacked basic supplies and sufficient teachers. Many only provided vocational classes or did not offer a full 12 years of instruction. Children were arbitrarily forced to attend based on factors like their complexion and last name.
Then Mendez’s parents fought back. In 1945, along with four other families, they filed a class action lawsuit against four Orange County school districts. Their goal: Ensure that all children could attend California schools regardless of race.
The case culminated in a two-week-long trial. In court, school officials claimed that Latino students were dirty and infected with diseases that put white students at risk. Besides, they argued, Mexican-American students didn’t speak English and were thus not entitled to attend English-speaking schools. (When asked, officials conceded that they never gave students proficiency tests.) “Mexicans are inferior in personal hygiene, ability and in their economic outlook,” said one official.
Mendez’s attorney countered with testimony from experts in social science. He argued that the policy trampled on Latino children’s Constitutional rights. When Carol Torres, a 14-year-old Latino girl, took the stand, she immediately proved that Mexican-American students in the district could and did speak English.
It took seven months for Judge Paul J. McCormick to render a decision. On February 18, 1946, he ruled that the school districts discriminated against Mexican-American students and violated their Constitutional rights. Though the school districts challenged the ruling, the Ninth Circuit Court of Appeals agreed with McCormick. Thanks to Mendez v. Westminster School District, California officially ended all segregation in its schools.
Mendez, who was eight when the lawsuits began, later told reporters that she thought her parents were fighting for her right to attend a school with a nice playground. But the case accomplished much more than that. Soon, parents in Texas and Arizona successfully challenged school segregation. In 1954, a decade after Mendez was turned away from the whites-only elementary school, the United States Supreme Court ruled that all school segregation based on race was unconstitutional in Brown v. Board of Education.
Though the case was a victory for the Mendez family, Sylvia was harassed and heckled by her fellow students when she attended the white school. Nonetheless, she pushed to succeed and became a nurse and civil-rights activist. She was awarded a Presidential Medal of Freedom in 2010—and now, two Los Angeles-area schools are named after her parents.
Today, an estimated 54 million Latinos live in the U.S. and around 43 million people speak Spanish. But though Latinos are the country’s largest minority, anti-Latino prejudice is still common. In 2016, 52 percent of Latinos surveyed by Pew said they had experienced discrimination. Lynchings, “repatriation” programs and school segregation may be in the past, but anti-Latino discrimination in the U.S. is far from over.
How racial barriers play in the experiences of Mexican Americans has been hotly debated. Some consider Mexican Americans similar to European Americans of a century ago that arrived in the United States with modest backgrounds but were eventually able to participate fully in society. In contrast, others argue that Mexican Americans have been racialized throughout U.S. history and this limits their participation in society. The evidence of persistent educational disadvantages across generations and frequent reports of discrimination and stereotyping support the racialization argument. In this paper, we explore the ways in which race plays a role in the lives of Mexican Americans by examining how education, racial characteristics, social interactions, relate to racial outcomes. We use the Mexican American Study Project, a unique data set based on a 1965 survey of Mexican Americans in Los Angeles and San Antonio combined with surveys of the same respondents and their adult children in 2000, thereby creating a longitudinal and intergenerational data set. First, we found that darker Mexican Americans, therefore appearing more stereotypically Mexican, report more experiences of discrimination. Second, darker men report much more discrimination than lighter men and than women overall. Third, more educated Mexican Americans experience more stereotyping and discrimination than their less-educated counterparts, which is partly due to their greater contact with Whites. Lastly, having greater contact with Whites leads to experiencing more stereotyping and discrimination. Our results are indicative of the ways in which Mexican Americans are racialized in the United States.
Keywords: Mexican Americans, race, racialization, education, skin color, social interaction, Mexican American Study Project
Mexican Americans have lower levels of education than non-Hispanic Whites and Blacks. Some scholars have argued that this is a result of Mexican immigrants having relatively low levels of education especially by standards in the United States, yet this gap is persistent and continues into the fourth generation (Telles & Ortiz, 2008). To explain this, we have argued that the education disadvantage for Mexican Americans largely reflects their treatment as a stigmatized racial group rather than simply being a result of low immigrant human capital or of other causes suggested in the literature (Telles & Ortiz, 2008). This paper investigates the role of race and racialization among Mexican Americans by more directly examining the relationships of education, skin color, and social interactions with racial identity and racial treatment (discrimination and stereotyping).
The role of race in the lives of Mexican Americans has been hotly debated. On the one hand, some argue that Mexican Americans have been racialized throughout their history in the United States (Acuña, 1972; Almaguer, 1994; Barrera, 1979; Foley, 1997; Gomez, 2007; Montejano, 1987; Ngai, 2004; Vasquez, 2010). Their long and continuous history as labor migrants destined to jobs at the bottom of the economic hierarchy and their historic placement at the bottom of the racial hierarchy, preceded by the conquest of the original Mexican inhabitants in what is now the U.S. Southwest, have created a distinct racial category of “Mexican” in the popular imagination. While not as heavily excluded from economic and social integration as African Americans, Mexican origin persons have encountered severe racial barriers, which have structured opportunities for them. These scholars argue that Mexican Americans lag educationally and economically even after several generations in the United States, as a result of this treatment. They have been thus limited to mostly working class jobs and from successfully integrating into middle class society.
On the other hand, others consider Mexican Americans to be similar to European Americans with modest backgrounds ago that arrived in the United States more than a century ago (Alba & Nee, 2003; Bean & Gillian, 2003; Perlmann, 2005). These assimilation theorists argue that while Mexican Americans may be slightly darker, slightly more stigmatized, and slightly more disadvantaged than these prior European groups, these factors will only slightly delay their integration into U.S. society. The key word here is “slightly.” These scholars recognize some of the disadvantages faced by the Mexican origin population but they do not consider these disadvantages sufficiently severe to affect long-term integration. However, the persistent educational disadvantage across generations and frequent reports of discrimination and stereotyping (like those we provided in Generations of Exclusion) challenge this view.
In this paper, we examine the ways in which race plays a role in the lives of Mexican Americans. While we use the same data previously used in Generations of Exclusion, the analysis are entirely new. Here, we study the relationships between racial appearance (such as skin color), education, and social interactions (such as contact with Whites), on the one hand, with racial identity and racial treatment, on the other. This paper thus extends our findings from Generations of Exclusion.
Mexican Americans and Race in History and Sociology
The issue of race among Mexican Americans is contested in many ways. The racial heritage of Mexicans is mixed, with varying mixtures of European, Indigenous, and African ancestry. As a result, Mexicans are heterogeneous in their racial characteristics, ranging from having light to dark skin and eye color with many in the brown and mestizo middle. Outsiders tend not to see Mexicans as White or Black. Rather they are viewed through the stereotypic lens of being non-white or brown and largely indigenous-looking. Still much about the racial status of Mexicans is debated. Two issues in particular are—one is whether Mexican is a racial category and, two is whether Mexicans are white or non-white.
Mexican Americans themselves often provide ambiguous responses to race questions, perhaps reflecting their own uncertainty about their race as well as ambivalence about being non-white (Gomez, 1992). Historically, Mexican Americans responded to questions about ethnic background with labels such Latin American or Spanish, as we showed with 1965 data in Generations of Exclusion (Telles & Ortiz, 2008). This reinforced European ancestry in responses about group membership and a distancing from indigenous heritage. Up to the 1960s, Mexican American leaders, such as those in the League of United Latin American Citizens (LULAC), emphasized the Spanish/European/White heritage of Mexican Americans, in attempts to secure rights as first class citizens and despite their treatment as non-white in American society (Gross, 2003; Haney-Lopez, 2006).
However, a new generation of “Chicano” activists in the 1960s radicalized the Mexican American movement for civil rights, leading to an affirmation their indigenous or non-white roots while advocating equal opportunity for all, regardless of race (Haney-Lopez, 2003; Muñoz, 1989). Since then, many Mexican Americans have embraced non-white notions of who they are. Today, many political elites position themselves as Hispanic and White (Haney-Lopez, 2003) while many academics, legal scholars, and activists position themselves as Chicano or Latino and non-white (Delgado, 2004). Among the general population, Mexican is often used as a response to the question “what is your race?” (Gross, 2003), thus reflecting a popular understanding that Mexican is a racial category distinct from Whites, Blacks, or Asians.
Sociologists have also debated how to define Mexicans racially. Using the census parlance of race and ethnicity, many rely on the official definition of Mexican as an ethnic group and that Mexicans can be of any race (Alba & Nee, 2003). This perspective of defining Mexican as an ethnic group aligns with notions that Mexicans are similar to previous European ethnic groups. For these scholars, ethnic groups are treated in more benign ways than racially distinct groups. Since Mexicans are not considered a racial group and thought to differ only slightly from Europeans, they should follow similar patterns of incorporation (Alba & Nee, 2003), easily move into honorary White status (Bonilla-Silva, 2003; Haney-Lopez, 2006), and subsequently incorporate fully into mainstream society (Alba & Nee, 2003).
Intermarriage adds another layer of complexity to the question of whether Mexicans are a racial category. Intermarriage of Mexicans with European Americans (or Whites) has been the most common type of intermarriage, leading to the speculation that this will also serve to quickly move Mexicans into being White (Alba, 2009; Alba & Islam, 2009). Yet children of Mexican-White marriages, while having lighter skin, may not actually abandon their Mexican identification (Jiménez, 2010). Moreover, intermarriage increasingly involves other racial groups like Blacks and Asians, especially in multi-racial places, like Los Angeles. While the children of these intermarriages may lose some connection to being Mexican as a result of a having a Black or Asian parent, they do not move closer to being White, so they should continue to be racially ambiguous and non-white.
Mexican Americans in the Census
The United States government, in its efforts to count persons and their characteristics, has played a major role in how Mexicans are defined and classified, and these definitions have shifted significantly over the years. There are two key issues about the classification of Mexicans—one is whether individuals are asked directly about being Mexican (or Hispanic) origin, and two is how the census collects and analyzes racial information for Mexicans (and Hispanics).
Asking about Hispanic origin is relatively straightforward. Every census since 1970 has included a question on Hispanic origin. The most recent (2010) wording of this question is: Is this person of Hispanic, Latino, or Spanish origin? The response categories have generally included Mexican (along with Mexican-Am. and Chicano), Puerto Rican, Cuban, and other. Since 1990, individuals were asked to fill in the country of origin when responding other Hispanic.1 Prior to 1970, Mexicans (and Hispanics) were not asked directly about being Hispanic origin. Rather Information about place of birth, parents' place of birth, and mother tongue was collected in those censuses and these characteristics were used to count and describe the Mexican origin population.
The issue of how race is collected and analyzed for Mexicans (and Hispanics) is much more complicated (Gibson & Jung, 2005). The general trend over time has been a shift from no classification to Mexican as a race, to Mexicans as White, to Mexicans as any race. Mexicans have resided in the U.S. since the mid-nineteenth century, yet up to the 1920 census, the Census Bureau made no mention of Mexicans or how to classify them. However, it appears that enumerators themselves made attempts to distinguish Mexicans from others since an unusually high number of “mulattoes” with Spanish surnames were counted in western states in 1880 (Hochschild & Powell, 2008).
The first time that Mexicans are officially counted is the 1930 census. That year, Mexican was listed as a racial category, the one and only time that this occurred. Also, enumerators were employed to collect census information and individuals did not respond for themselves. The instructions provided to enumerators provide insights into how Mexicans were viewed at the time. They read as follows:
Mexicans.-Practically all Mexican laborers are of a racial mixture difficult to classify, though usually well recognized in the localities where they are found. In order to obtain separate figures for this racial group, it has been decided that all person born in Mexico, or having parents born in Mexico, who are not definitely white, Negro, Indian, Chinese, or Japanese, should be returned as Mexican (“Mex”)2
These instructions indicate the understanding that Mexicans were mixed race but clearly not White or perceived as White. The use of “laborers” in the first line of this instruction suggest that class may have played a role into the use of Mexican in that laborers might have been classified as Mexican but higher status Mexicans might have been classified as White (Hochschild & Powell 2008).
In response to protests from the Mexican government and LULAC about using Mexican as a racial category, the Census Bureau changed the official designation of Mexicans to White (Gross, 2003; Hochschild & Powell 2008). Consequently, the 1940 and 1950 census provided the following instructions to enumerators: “Mexicans are to be regarded as white unless definitely of Indian or other nonwhite race.”3 This clearly shows the shift from Mexican as a race to Mexicans as White. The population of Spanish mother tongue—defined from place of birth, parents' place of birth, and mother tongue—was counted and described in official publications (Gibson & Jung, 2005).4
Enumerators did not completed forms in the 1960 census, rather census forms were mailed to households and completed by individuals,. This made it possible for Mexicans (and Hispanics) to respond using any racial category. But the Census Bureau in both 1960 and 1970 continued to define Mexicans (and Hispanics) as racially White. Therefore Mexicans (and Hispanics), who responded other to the race question, had their answers changed (or recoded) to White. This served to ignore what individuals reported about themselves. Starting in 1980, the Census Bureau stopped defining Mexicans (and Hispanics) as White and defined them as being of any race. This meant that they stopped changing responses as other race provided by Mexicans (and Hispanics). Consequently from 1970 to 1980, there is a sharp increase in the overall number of individuals reporting that they are other race, largely attributed to large percentage of Hispanics who choose other race (Gibson & Jung, 2005). More than 40 of Hispanics answered other to the race question in 1990 (Rodriguez, 2000) and more than 45 percent of Mexicans reported that they are other race in 2000 (Bonilla-Silva, 2003).5
Of course, what is rarely acknowledged or reported is that when Mexicans report their race as other, they subsequently add Mexican in the explanation to this response—de facto, naming Mexican as their race. Census officials raise the concern that Mexicans, and Hispanics more generally, choose racial responses that do not fit officials' definitions of race. They find this so objectionable that they have sometimes argued that Latinos are “confused” or find it “difficult” to understand the race question on the census (Rodriguez, 2000).6 When we consider how the Census Bureau has changed the racial classification of Mexicans from none, to Mexican, to White, to any race, it could be argued that government bureaucrats are confused.
Mexican Americans as Non-Whites
Race is a social construct but one that has had real consequences in the United States. Although granted de facto White racial status with the United States conquest of much of Mexico in 1848 and having sometimes been deemed as White by the courts and censuses, Mexican Americans were rarely treated as White (Gomez, 2007; Haney-Lopez, 2006). Historically and legally, Mexicans have been treated as second-class citizens. Within a few short decades after their conquest in the mid-nineteenth century, Mexican Americans, although officially granted United States citizenship with full rights, lost much of their property and status and were relegated to low-status positions as laborers. Since then, Mexican immigration has continued to be of predominately low status. Throughout the twentieth century, Mexicans with low levels of education and from poor backgrounds immigrated to the United States to fill the lowest paid jobs (agriculture, domestic work, construction) with peaks during the Mexican Revolution in 1910 to 1929, during the agricultural guest worker program for Mexicans (Bracero program) from 1942 to 1964, and post the Immigration Act of 1965 which liberalized immigration from the Americas. Most of Mexican immigration has been to the southwestern United States, although Mexicans have begun to settle in nearly all regions of the United States since about 1990. This continuous immigration throughout the twentieth century has meant that the Mexican origin population in the United States includes many persons born in the United States, varying in generational status from first (immigrant) to fourth and even fifth generation. These later generations have continued to face educational and economic disadvantages as we documented in Generations of Exclusion (Telles & Ortiz, 2008).
Unfair and discriminatory treatment against Mexican Americans has extended beyond the economic realm. School segregation has been extensive, both historically and in contemporary periods. Throughout history, Mexican children were sent to separate and inferior schools (Alvarez, 1986; San Miguel, 1987; Sanchez, 1993). School segregation was repeatedly challenged in the courts. While they were treated as non-white by Whites, challenges to segregation were won by employing the racial designation of White under the law, meaning that Mexicans as Whites could not be segregated from other Whites (Martinez, 1997). Courts did allow the segregation of Mexicans due to language or migrant status. In the post civil rights era, Mexicans were used as the non-Blacks that integrated schools for Black children (Gross, 2003; Mechaca, 1995)). Eventually Mexicans moved from being considered White to brown, probably due to both legal and social changes although it is difficult to tell which of these occurred first (Gross, 2003). As Mexicans came to be defined as non-whites, they were better able to make claims of unfair treatment and seek legal remedy.7
Persuasive anti-immigrant sentiment and treatment has also worked against all Mexicans whether immigrant or born in the United States. Viewed as alien and low status, Mexican immigrants were (and continue to be) scapegoated and targeted for mistreatment. Even though immigrants were a minority of all Mexican Americans up to the 1980s, the perception of all Mexican Americans as low status immigrants has been pervasive (Massey, 2009; Vasquez, 2010). The immigration legislation of the 1980s has made legal entry to the United States by Mexicans almost impossible, yet immigration has continued. This forced the overwhelming majority of Mexican immigrants in the late twentieth century to enter the United States without proper documentation. This has served to further fuse anti-Mexican and anti-undocumented immigrant sentiment (Massey, 2009). This suggests that in the eyes of many White Americans, all Mexicans are “illegal” and all “illegals” are Mexican (Chacón & Davis, 2006; Chavez, 2008)
If Mexican Americans see themselves as part of a racial category and are treated largely as non-white, what implications does this have for their experiences? Racial experiences are varied and involve many aspects of a person's life. On the one hand, some experiences concern how members of racial/ethnic groups view themselves. They may perceive themselves to be members of an ethnic group, like Italian-American, in a largely symbolic manner (Waters, 1990). Or their identity may be a racial one, which implies a ranking along a racial hierarchy and which carries palpable social consequences. Secondly, they may encounter stereotypes which define how they should behave or who they should be; or they may encounter discrimination where they are treated differently due their group membership.
Racial appearance should factor into racial treatment, since we often define race as based on physical difference.8 For example, Mexican Americans who are darker and physically differ to a greater extent from Whites are more likely to be perceived as members of the group. Moreover, to the extent that the group is considered non-white and stigmatized, darker Mexican Americans would be subject to greater stereotyping and discrimination than their light skin counterparts. Another indicator of racial appearance could be that of having a non-Hispanic parent. The offspring of mixed marriages, in addition to a skin color advantage, might carry other characteristics of Whiteness such as non-Spanish names and cultural or social resources that make them more acceptable to Whites.
Additionally we expect education to be related to racial outcomes among Mexican Americans. On the one hand, Mexican Americans with less education may have stronger perceptions of themselves as members of the group than those with more education. Conceivably, this relationship could be reversed if those with more education have greater awareness of being part of the group. What of the relationship between education and racial treatment? Less educated Mexican Americans might experience more stereotyping and discrimination because of their disadvantaged educational status. Or the more educated might experience worse treatment because of greater interactions in mainstream institutions and with members outside of their group. Additionally being more educated might increase awareness that Mexican Americans are treated in a racial manner and that might explain part of the education effects; in other words educated Mexican Americans might perceive that discrimination exists to a greater extent and that might partially explain their reports of being discriminated against.
Lastly, social interactions with Whites and with other Mexican Americans might affect perceptions and treatment. Having a greater number and closer relationships with Mexican Americans should reinforce the connection with the group but it is uncertain how these will affect treatment. Conversely interactions with Whites might result in more negative treatment but will that affect perceptions of being Mexican.
Lastly, what kinds of discrimination experiences do Mexican Americans describe? In what settings do they experience discrimination? Responses to open-ended questions in our survey provide glimpses into these experiences.
In sum, this paper examines (1) racial characteristics like skin color, (2) education, and (3) social interaction variables like contact with other Mexicans and Whites as predictors of (1) racial identity as in choosing a racial identity as Mexican and being perceived as Mexicans (2) racial treatment as in experiences of stereotyping and discrimination.
This study draws on a unique data set of 758 Mexican American adults between the ages of 30 to early 50s who were interviewed between 1998 and 2002. These respondents are adult children of the original respondents in the Mexican American Study Project (MASP).
MASP is a unique data set based on two waves of data collection 35 years apart. The original data was based on a random sample of households with adult Mexican Americans in Los Angeles County and San Antonio City who were interviewed in 1965-66 and those findings were published in The Mexican American People (Grebler, Moore, & Guzman, 1970). We conducted a follow up survey with the original respondents from this earlier survey and their adult children in 1998-2002 and our findings were published in Generations of Exclusion (Telles & Ortiz, 2008).
The re-interviews with surviving original respondents produced a longitudinal sample (1965 and 2000) of 687 original respondents who were age 18 to 50 in 1965. Two adult children were interviewed producing an inter-generational sample of 758 child respondents who were age 30 to early 50s when interviewed in 2000. Moreover, both the original 1965 respondents and their children are almost evenly divided by generations-since-immigration with about one-third of the child sample being second generation, one-third of the third generation and another third in the fourth generation or more.
In 2000, original respondents and adult children were interviewed extensively about ethnic identity, behavior, and attitudes; education and socio-economic status; political attitudes and behavior; family attitudes and behaviors. In this paper, we analyze the child sample based on their responses as well as some information from their original respondent parents.
In our analysis, we study two sets of outcomes—(1) racial identity as in choosing a racial identity as Mexican and being perceived as Mexicans (2) racial treatment as in experiences of stereotyping and discrimination. We examine three key sets of predictors—(1) racial characteristics like skin color, (2) education, and (3) social interaction variables like contact with other Mexicans and Whites.
We examine four outcomes in this paper and we show the distribution for these variables in Table 1. Two of the outcomes measure racial identity. The first refers to how the respondents identify racially in response to “when forms or the census ask if you are White, Black, Asian, American Indian, or other, what do you answer?” To replicate the census question, we provided an “other” option for this question but purposefully did not include a Mexican American or Latino/Hispanic response category. Many respondents answered Mexican, Mexican American, or Mexican origin while others responded Hispanic or Latino. Responses to this question were grouped into Mexican, Mexican American, or Latino (coded as 1) compared to all other responses, such as White or Black (coded as 0). Two-thirds reported identifying as Mexican or Mexican American or Latino to the question about race (see Table 1). The second perception measure involves responses to “Do you think that when someone meets you for the first time that they think of you as Mexican? We coded their responses as perceived by others as Mexican (coded as 1) or not perceived or probably not perceived to be Mexican (coded as 0). Table 1 shows that 38 percent are definitely perceived as Mexican while 62 percent are not or probably not perceived as Mexican.
Distribution of Outcomes: Race Perceptions and Race Treatment
Our final two outcomes measure racial treatment. The first of these is whether respondents experience being stereotyped by others. We asked: “Sometimes people have ideas about what certain groups are like or what they are supposed to do. Do you ever find that other people expect you to be like or do things that they expect of Mexicans?” The responses are yes (coded as 1) or no (coded as 0). Table 1 shows that 58 percent perceive that they had experienced stereotyping. The second experience measure is whether respondents reported experiences of discrimination. We asked “Have you been treated unfairly because of your ethnic background?” The responses are yes (coded as 1) or no (coded as 0). Almost half (48 percent) reported experiences with discrimination. In contrast, Alba reports that approximately five percent of White ethnics experience discrimination (Alba, 1990).9
The first set of predictors captures physical markers of racial status, including one about actual skin color and another about parentage (means and ranges are presented on Table 2). Respondents' skin color was rated by interviewers rather than by respondents. We use the seven-point scale with seven being the darkest and one being the lightest.10 The average on the skin color scale was 4.3, which is about halfway on the seven-point scale and indicating a medium brown skin color. A non-Hispanic parentage variable codes respondents that have only one parent as non-Hispanic as one and those with two Hispanic parents are coded as zero. Approximately 8 percent of respondents in the sample have a non-Hispanic parent.11
Descriptive Statistics for Key Predictors
Education is the second key predictor in our analysis (see Table 2). We compare three categories: those with less than high school (around 16 percent and the reference category), high school graduates and those with some college (71 percent), and college graduates or more (13 percent)12.
We include a measure indicating whether respondents perceive a lot of discrimination against Mexicans. We asked “how much discrimination do you think there is today again people of Mexican origin?” A significant percentage, 36 percent, reported that there is a great deal of discrimination against Mexicans. This is a measure of whether discrimination occurs generally and differs from the outcome measure of whether respondents report experiencing discrimination personally
The third set of predictors measure the extent of social interaction with racial/ethnic groups (see Table 2). The first involves the amount of contact with Whites, which was measured in our survey with four responses—not at all, a little, some, and a lot. Most respondents, 88 percent, reported having “a lot of contact” with Whites and thus we grouped this variable into two categories—not a lot of contact (coded as 0) and a lot of contact (coded as 1). The second social interaction variable gauges the frequency of friendships with Mexicans measured on a four-point scale ranging from none to few friendships (coded as 1), about half (coded as 2), most (coded as 3), and all friendships (coded as 4). The average is 2.3 or about halfway on the four-point scale.
We control for three sets of variables: individual characteristics, generational status and socio-economic background (means and ranges are presented on Table 3). The individual characteristics include gender, urban area, age and type of interview. Female (coded as 1) is compared to male (coded as 0), which divides at approximately 50 percent. The variable, San Antonio, refers to whether the original respondent parent was interviewed in San Antonio (coded as 1) or Los Angeles (coded as 0) with 36 percent of the sample from San Antonio and the remainder from Los Angeles. Age is a continuous variable ranging from 32 to 59 years old and averaging 42. Interviewed by phone (coded as 1) refers to respondents who were interviewed by phone, which is about 27 percent of the sample, rather than in person (coded as 0); we include this variable because some questions were not asked in the phone interviews and because they were somewhat more likely to have moved out of the area.
Descriptive Statistics for Control Variables
The second of the control variables is generational status (see Table 3). Generation 1.5 refers to respondents born in Mexico and raised in the United States, which is about 7 percent of the sample. Generation 2.0 refers to those born in the United States with two parents born in Mexico, which is 8 percent of respondents. Generation 2.5 are born in the United States with one parent born in Mexico and the other parent in the United States, which is 22 percent of the sample. Generation 3.0, the reference group, are born in the United States with both parents born in the United States and two to four grandparents born in Mexico, which is 40 percent of the sample. Generation 4.0 are born in the United States with both parents born in the United States and three of the four grandparents born in the United States, which comprises 23 percent of respondents.
The third set of control variables refers to socio-economic background (see Table 3). Much of this information is from the respondent's parent (the original respondent) and collected in either 1965 or 2000. Father's and mother's education range from 0 to 17 with an average of about 9 years of education. Income is family income in 1965. This was collected as a categorical variable, which we recoded to thousands. The average income is about $6,000 and the lowest category refers to income of less than $500 and the highest category refers to income of more than $20,000. Homeownership refers to whether the original respondent parent owned a home in 1965 (coded as 1); approximately 55 percent owned homes. The variable, number of siblings, ranges from 0 to 15 with an average of 5 siblings. Parent spoke Spanish to their children is coded as 1 for Spanish and 0 for not speaking Spanish (this was reported by the original respondent parent). About 85 percent of respondents had parents who spoke Spanish to them when they were children. The last variable is parents were married in 1965 while the respondent was growing up; 85 percent of parents were married at the time of the original survey.
Table 4 presents logistic regression analysis where racial identity as Mexican is the dependent variable. The numbers presented on this table, and the other tables with multivariate analyses, are odds ratios. An odds ratio equal to: one indicates no relationship, less than one indicates a negative relationship with numbers closer to zero indicating a stronger negative relationship, and greater than one indicates a positive relationship with larger numbers indicating a stronger positive relationship13.
Effect of Racial Appearance, Education, Social Interactions, and Controls on Identify Racially as Mexican among Mexican Americans
Racial Identity as Mexican
In the initial model on Table 4, we observe that skin color is marginally related to racial identity as Mexican, indicating that darker respondents are somewhat more likely to identify racially as Mexican.14 Having a non-Hispanic parent is unrelated to racially identifying as Mexican. More educated respondents were less likely than the low educated to identify racially as Mexican. Those who completed high school or have some college are significantly less likely to identify racially as Mexican—54 percent are less likely to do so. And the college educated is marginally less likely to identify racially as Mexican. In the second model, perceiving a lot of discrimination does not have a direct effect on identifying as Mexican or change the education effect. In the third model, contact with Whites and friendships with Mexicans are unrelated to identifying as Mexican.
The rest of Table 4 present control variables. Respondents from San Antonio and respondents who were interviewed by phone were significantly less likely to identify racially as Mexican. Gender and age are unrelated to identifying as Mexican. Generational status is unrelated to identifying as Mexican. The indicators of socio-economic status were also unrelated to identifying as Mexican (except for a marginal effect of income in the all three models).
Perceived as Mexican
Table 5 presents the findings for being perceived as Mexican. The first model indicates that having darker skin is significantly related to being perceived as Mexican. An odds ratio of 1.3 reveals that with every unit increase on the skin color scale, being perceived as Mexican increased by 30 percent. Those with a non-Hispanic parent are significantly less likely to be perceived as Mexican—about half as likely (odds ratio equals .44). More educated respondents were less likely than the less educated to be perceived as Mexican. Those who completed high school or have some college are 54 percent, and the college graduates are 31 percent, as likely to be perceived as Mexican. In the second model, perceiving a lot of discrimination does not have a direct effect on being perceived as Mexican or reduce the education effect.
Effect of Racial Appearance, Education, Social Interactions, and Controls on Perceived as Mexican among Mexican Americans
The third model indicates that those who have a lot of contact with Whites are half as likely to be perceived as Mexican (odds ratio equals .51). Persons with more Mexican friends are more likely to be perceived as Mexican—approximately one and a half times as likely (odds ratio equals 1.45). The effect of non-Hispanic parent, which was significant in the initial model (odds ratio equaled .44) and, is slightly reduced and is now marginally significant (odds ratio is .54). This is due to adding the extent of Mexican friends to the model. So persons with a non-Hispanic parent are less likely to be perceived as Mexican but some of this effect is through the friendships that Mexican Americans have with other Mexicans.15
The rest of Table 5 present control variables. Gender and being from San Antonio are unrelated to being perceived as Mexican. Older respondents are marginally less likely to being perceived as Mexican. Respondents interviewed by phone were less likely to being perceived as Mexican in the first and second model but is not significant in the third model when social interactions are added to the model. The fourth generation is less likely to be perceived as Mexican than the third generation. Generation two differs from generation three in the first and second model but is not significant in the third model when social interactions are added to the model. The indicators of socio-economic status are unrelated to being perceived as Mexican.
The first model presented on Table 6 indicates that skin color and having a non-Hispanic parent are unrelated to experiences of stereotyping. More educated respondents are more likely to have been stereotyped by others. The relationship between education and experiences of stereotyping is especially strong—those with a high school or college education are more than twice as likely to being stereotyped than those less than high school (odds ratio equal 2.1 and 2.4 respectively).
Effect of Racial Appearance, Education, Social Interactions, and Controls on Experience Stereotyping among Mexican Americans
The second model on Table 6 shows that perceiving a lot of discrimination against Mexicans has a direct and positive effect on personal experiences with stereotyping. The third model shows that contact with Whites also increases experiences with stereotyping—those with more contact with Whites are two and a half times more likely to experience stereotyping. The extent of Mexican friends, on the other hand, is unrelated to experiences of stereotyping.
One reason that education is thought to lead to more experiences with stereotyping is that education may sensitize minorities to experiences of racialization. In other words, through learning about the history of racism and discrimination against the group, Mexican Americans may become more aware of its existence in their own lives. Adding general perceptions of a lot of discrimination (in the second model) does not change the relationship between education and perceptions of having been stereotyped. That more educated Mexican Americans perceive having more experiences of being stereotyped remains strong and significant in the second model.
Another reason that education is thought to relate to experiences with stereotyping is that participation in educational institutions provides more contact with Whites and thus greater awareness of White attitudes. Consistent with this, we observe that the education effects are somewhat reduced when contact with Whites is introduced in the third model. For instance, the college graduate effect is 2.4 when contact with Whites is excluded from the second model and is 1.99 when contact with Whites is added in the third model. So greater contact with Whites probably explains some of the experiences of being stereotyped.
The rest of Table 6 presents the control variables. Gender, being from San Antonio, and age are unrelated to experiences with stereotyping. Respondents interviewed by phone were less likely to experience stereotyping. Generational status was unrelated to experiences with stereotyping (except for that marginal effect of generation 1.5 in the first model). The indicators of socio-economic status are unrelated to experiences of stereotyping (except for that marginal effect of homeownership in the first model).
As Table 7 shows, being darker also leads to reports of experiences with discrimination— an odds ratio of 1.2 indicates that with every unit increase on the skin color scale, reports of discrimination increased by 20 percent. Having a non-Hispanic parent is unrelated to reporting discrimination.
Effect of Racial Appearance, Education, Social Interactions, and Controls on Experience Discrimination among Mexican Americans
College educated respondents are more likely to report experiences of discrimination— twice as likely as those with less than a high school education (odds ratio equals 2.0). On the other hand, those with a high school education do not differ significantly from the least educated. In the second model, we observe that perceiving a lot of discrimination generally has a direct and positive effect on personal experiences with discrimination (odds ratio equals 2.3). In the third model, having more contact with Whites is shown to being more likely to be experience discrimination (odds ratio of 1.96). Having Mexican friends is unrelated to experiences of discrimination.
One of the issues we have tracked in the multivariate analysis is what happens to the education effects as we introduce perceptions of discrimination and contact with other groups. First, we observe that the education effect does not change between the first and second models when general perceptions about discrimination are entered into the model. Second, we observe that the college graduate effect is somewhat reduced between the second and third models when contact with Whites is introduced into the model. The college effect is 2.0 and statistically significant in the second model but when social interactions are introduced in the third model, the effect is reduced to 1.79 and is only marginally significant. Having more contact with Whites may partially explain why college graduates report more experiences of discrimination.
One of the control variables, gender, indicates that women are less likely to report discrimination (in contrast, gender was unrelated to any of the other racial outcomes). Given the strong relationship of gender and discrimination, we did some exploratory analysis separately by gender and we found that skin color has a much stronger effect for men than women. This is shown as the interaction effect between skin color and gender (Darker Color X Female) in the fourth model on Table 7. This significant interaction effect indicates that darker men are much more likely to experience discrimination while darker and lighter women do not differ in their experiences of discrimination.
The rest of Table 7 presents the control variables. Being from San Antonio and age are unrelated to perceptions of experiences with discrimination. Respondents interviewed by phone were less likely to experience discrimination. Generation 1.5 reports more experiences with discrimination while the other generational groups do not differ from third generation.
The last group of control variables refers to socio-economic status. Parental education has an unusual pattern of relationships to discrimination experiences in that the greater the father's education, the more likely they are to report experiences of discrimination while the greater the mother's education, the less likely they are to report experiences of discrimination. Number of siblings is significantly related to experiences of discrimination in the third model when social interactions are entered in the model. The other controls—income, homeownership, speaking Spanish, married parents—are unrelated to experiences with discrimination.
Summary by Predictor Variables
Overall, skin color, education, and contact with Whites have the strongest relationships with the racial outcomes. Mexican Americans who are darker, more educated, and have more contact with Whites are more likely to be perceived as Mexicans, experience more stereotyping, and experience more discrimination. Additionally, skin color has a different relationship with discrimination experiences for men and women such that darker men report more discrimination.
Describing Incidents of Discrimination
The question in our survey on discrimination, which we used in this paper, was “Have you been treated unfairly because of your ethnic background?” Respondents who indicated that had been treated unfairly were also asked to describe these experiences. Respondents provided examples from various aspects of their lives including work, school, police, public life, and peers and their responses illustrate their real life experiences, beyond the yes/no responses to close-ended questions that we examined through quantitative analysis above.
The largest number of comments—over 90—was about employment incidents. The most prevalent among these were reports about being denied promotions. One respondent reported that “they were hiring for assistant foreman, and I had seniority and better qualifications and I was overpassed for the position.” Other respondents reported that they were not hired for jobs based on their racial appearance. Respondents also reported negative or hostile interactions with supervisors—one respondent reported filing a federal discrimination complaint against a supervisor. Still other respondents reported difficulties with customers or clients where their help was rejected. There were also a small number of reports about being denied an apartment.
Another large number of reports—45 reports—referred to school-based incidents. These included encountering teachers that had low expectations of them as in indicating surprise at the respondents' academic ability, or assuming that respondents did not speak English or had recently arrived from Mexico. Some respondents reported getting into trouble for speaking Spanish. Other reported derogatory names, such as “wetback,” by teachers or other school officials.
Respondents also reported problems with peers. Some of these incidents were physical as in the respondent who reported that other children tried to cut out his brown eyes in first grade. Other respondents reported being beat up by peers in junior high or high school. While these were events that took place in school setting, other similar events happened in neighborhoods and with peers while growing up. Incidents where respondents were called derogatory names or rejected by friends of friends or by parents of friends.
The other very large category of incidents involved being denied service in restaurants. Some of incidents involved direct comments like “we do not serve people like you.” Other incidents involved being ignored or receiving very slow service. Respondents also reported receiving poor service in stores by being ignored or followed.
There were about 20 reports involving police. These involved being stopped and harassed by the police. Some of these events led to arrests or searches. Some respondents reported being called derogatory names like “wetback.” There were other incidents involving government officials, like being scrutinized by the border patrol as they crossed the border
Although all of our respondents came of age in the post civil rights era, they reported fairly extensive experiences of discrimination and being stereotyped. This is evident in their responses to standard survey questions (with close-ended responses) as well as in their accounts of specific instances discrimination (presented in the previous section). These experiences were prevalent in institutional settings like the work place and school as well as in public places like restaurants and retail stores. These experiences, although almost certainly fewer in frequency and lesser in intensity than that documented historically (Montejano, 1987), are indicative of pervasive racism and discrimination continuing in the post civil rights era.
Skin color is important in our findings in that darker Mexicans are more likely to be perceived as Mexican and experience discrimination. These are strong relationships controlling for the many other factors in our analysis. To outsiders, skin color is a key marker of group membership, consequently darker Mexican Americans are treated as stereotypically Mexican. Additionally, darker men report more experiences of discrimination than lighter men and women in general. This is consistent with prior research showing that minority men are especially likely to face obstacles in education, the labor market, and criminal justice system (Harrison, Reynolds-Dobbs, & Thomas, 2008; Hersch, 2008; Reimers, 1983). Some respondents indicate this in their reports of incidents with police officers. On the other hand, having a non-Hispanic parent has a relatively weak effect. Although being the child of inter-marriage is considered one mechanism by which Mexican Americans can move away from being Mexican to being honorary White (Alba, 2009), we do not find this to be the case. Children of intermarriage do not differ in most ways from those with two Mexican parents.16
Education is important. Mexican Americans with more education report that they were less likely to racially identify as Mexican and be considered Mexican by outsiders, and more likely to be stereotyped and face discrimination. Among the strongest relationships we identified was that of education with being stereotyped. It appears that educated Mexican Americans go against the notions that outsiders have about the group. This may be partly due to the low levels of education among Mexican immigrants and that Mexican Americans even in later generations have relatively less education. Outsiders expect Mexican Americans to be less educated and treat them accordingly. But to apply that expectation to all Mexican Americans creates a stereotype.
Having more education means living and working in environments with more Whites and participating in institutions of higher education also involving more interactions with Whites. Greater contact with Whites partly explains the relationship between education and experiences of stereotyping and discrimination consistent with prior research by Feagin and his colleagues (e.g., Feagin & Sikes, 1994). Moreover, contact with Whites has a significant and independent effect on experiences of stereotyping and discrimination (which is not diminished by controlling for other factors). The specific examples of discrimination that respondents shared illustrate how this operates. In work places with employers and co-workers, Mexican Americans are likely to come in contact with Whites that treat them in discriminatory ways; for instance they report being passed over promotions or not getting hired. In education settings, teachers and other school staff make derogatory remarks or convey the message that Mexican Americans are less worthy. Mexican Americans also reported unfair treatment in public spaces, like restaurants and stores. It is in these interactions beyond the family and ethnic neighborhood, that Mexicans Americans face unequal treatment.
The most significant limitation of this study is probably identifying the extent of discrimination and stereotyping. Collecting precise information on discrimination and stereotyping in survey research is challenging. Individuals may not always have information about how or whether they are systematically being treated differently from others or they overestimate whether they are being treated unfairly due to their group membership. While measuring racial treatment in a precise manner is unlikely, several factors make us confident that we are capturing a real phenomenon. One is that experimental studies show strong evidence that discrimination exists17 (Pager & Hana, 2008) supporting our respondents' reports of discrimination. Two is the legal history showing the ways in which Mexican Americans have been and continue to be treated in discriminatory ways (Gross, 2003; Haney-Lopez, 2006). Three is the much greater reports of experiencing stereotyping and discrimination by Mexican Americans than other ethnic groups, such as Italians (Alba, 1990). Lastly, the pervasive patterns of racial inequality as we describe here and in our prior work (Telles and Ortiz 2008) support the view of substantial racialization of Mexican Americans over several generations.
We acknowledge comments by Gary Koeske.
Correlations among Outcome Variables
|Identify Racially as Mexican||Perceived to be Mexican||Experience Stereotyping|
|Identify racially as Mexican||1.000|
|Perceived to be Mexican||0.090||1.000|
View it in a separate window
11Enumeration forms and questionnaires found in the IPUM website: http://usa.ipums.org/usa/voliii/tEnumForm.shtml
2Enumeration forms and questionnaires found in the IPUM website: http://usa.ipums.org/usa/voliii/tEnumForm.shtml
3Enumeration forms and questionnaires found in the IPUM website: http://usa.ipums.org/usa/voliii/tEnumForm.shtml
4Up to 1950, the Mexican origin population comprised the vast majority Hispanics in the United States. Puerto Ricans were only beginning to migrate to New York City and the Cuban immigration had not begun in earnest (Bean and Tienda 1987). Mexican Americans continue to be the majority at about 66 percent.
5By 1980, while Mexicans continue to be the largest Hispanic group, there are significant numbers of Puerto Ricans and Cubans residing in the United States. By 2000, there are significant of Dominicans and Central Americans (Guzman 2001).
6Also unacknowledged by census officials is that the race question itself is confusing since it includes seven Asian categories (Asian Indian Chinese, Filipino, Japanese, Korean, Vietnamese, other Asian), four Pacific Islander categories (Native Hawaiian, Guamanian or Chamorro, Samoan, other Pacific Islander), and only four race categories (White, Black, American Indian, some other race). The census question on race includes many more country of origin or ethnic categories than racial categories.
7School segregation of Mexican American children was common in California and Texas. In 1928, 64 schools in eight Southern California counties had Mexican student enrollments of 90percent or more (cited in Sanchez 1993:258). Another study reported that 80 percent of school districts with substantial Mexican American enrollment practiced segregation (cited in Sanchez 1993:258). Alvarez (1986) lists 64 schools in eight California counties, including several in Los Angeles, that were 90 to100 percent Mexican. Well-known California judicial cases of segregation included the school districts of Lemon Grove (in the 1930s) and Westminster (in the1940s). Although segregation was documented in both California and Texas, egregious discrimination may not have endured into the 1960s in California to the same extent as it did in Texas.
8A number of studies have examined the relationship between skin color and socio-economic outcomes. These studies have shown that darker Mexican Americans have lower education, earnings, and other economic outcomes (Arce, Murguía and Frisbie 1987; Murguía and Telles 1996; Telles and Murguía 1990).
9Appendix A presents correlations among the outcome variables. The interrelationships are low(less than .1) except for the relationship between perceptions of being stereotyping and reports of being discrimination (which equals .3).
10The skin color chart used in our survey had 9 categories. There were few respondents in the two lightest categories and few in the two darkest categories so these categories were combined so that we ended up with seven categories.
11The original MASP may have under-sampled Mexican origin women married to non-Mexicanmen because Spanish surname was used for sampling in more integrated neighborhoods.
12Our sample appears more educated than current data on the Mexican origin. This is due to the fact that our sample is U.S. born with most being third and fourth generation. A sample of Mexican origin drawn today would be overwhelming immigrant and second generation. This results in low education levels among Mexicans today since immigrants have low levels of education.
13We use robust standard errors in the logistic regression to adjust sibling clustering.
14We did additional analysis in an attempt to explain the marginally relationship between skin color and identify racially as Mexican. Specifically we examined whether omitting education and non-Hispanic parent changed this relationship—making it stronger or weaker—but did not find that this marginal relationship changed in any meaningful way.
15One indicator of having a non-Hispanic parent is having a non-Spanish name. We substituted having a Spanish name for having a non-Hispanic parent in the analysis to see if this effect was different. We found similar results as we did with non-Hispanic parent. Also since respondents with a white parent tend to be lighter skin than those with two Mexican parents, having a non-Hispanic parent is correlated with skin color. Thus it is possible that excluding skin color from the analysis might change and make significant the effect of non-Hispanic parent. We did additional analysis where we excluded skin color and doing so did not change the effect of non-Hispanic parent.
16As reported in footnote 16, the effect of non-Hispanic parent is due to a relationship with skin color since excluding skin color from the analysis does not change the effect of non-Hispanic parent.
17Experimental studies are the gold standard of research, including research on discrimination. As an example of this kind of study, profiles of “applicants,” actually fictional individuals with equal credentials, are submitted to employers; researchers examine whether employers call back or offer a job to “applicants”; and treating “applicants” differently based on race is evidence of discrimination because other differences among “applicants” are held constant. What an experimental design does not provide is an understanding of how discriminatory treatment affects victims in real life.
Vilma Ortiz, Department of Sociology, University of California, Los Angeles.
Edward Telles, Department of Sociology, Princeton University.
- Acunña R. Occupied America: The Chicano's Struggle Toward Liberation. Francisco, CA: Canfield Press; 1972.
- Alba R. Ethnic Identity: The Transformation of White America. New Haven, CT: Yale University Press; 1990.
- Alba R. Blurring the Color Line: The New Chance for a More Integrated America. Cambridge: Harvard University Press; 2009.
- Alba R, Islam T. The Case of the Disappearing Mexican Americans: An Ethnic-Identity Mystery. Population Research Policy Review. 2009;28:109–121.
- Alba R, Nee V. Remaking the American Mainstream: Assimilation and Contemporary Immigration. Cambridge, MA: Harvard University Press; 2003.
- Almaguer T. Racial Fault Lines: The Historical Origins of White Supremacy in California. Berkeley, CA: University of California Press; 1994.
- Alvarez RR., Jr The Lemon Grove Incident: The Nation's First Successful Desegregation Case. The Journal of San Diego History. 1986;32(2):116–135.
- Arce C, Murguía E, Frisbie P. Phenotype and Life Changes among Chicanos. 1987. 1987;9:19–32.
- Barrera M. Race and Class in the Southwest: A Theory of Racial Inequality. Notre Dame, IN: University of Notre Dame Press; 1979.
- Bean F, Gillian S. America's Newcomers and the Dynamics of Diversity. NY: Russell Sage Foundation; 2003.
- Bean F, Tienda M. The Hispanic Population of the United States. NY: Russell Sage Foundation; 1987.
- Bonilla-Silva E. Racism Without Racists: Color-Blind Racism and the Persistence of Racial Inequality in the United States. Lanham, MD: Rowman & Littlefield; 2003.
- Chacón J, Davis M. No One is Illegal: Fighting Racism and State Violence on the U S-Mexico Border. Chicago: Haymarket Books; 2006.
- Chavez L. The Latino Threat: Constructing Immigrants, Citizens, and the Nation. Sanford CA: Stanford University Press; 2008.
- Delgado R. Locating Latinos in the Field of Civil Rights: Assessing the Neoliberal Case for Radical Exclusion. Texas Law Review. 2004;83:489–524.
- Feagin J, Sikes MP. Living with Racism: The Black Middle-Class Experience. Beacon; 1994.
- Foley N. The White Scourge: Mexicans, Blacks and Poor Whites in Texas Cotton Culture. Berkeley, CA: University of California Press; 1997.
- Gibson C, Jung K. Population Division Working Paper No 76. U.S. Census Bureau: 2005. Historical Census Statistics On Population Totals By Race, 1790 to 1990, and By Hispanic Origin, 1970 to 1990, For Large Cities And Other Urban Places In The United States.
- Gomez L. The Birth of the “Hispanic” Generation: Attitudes of Mexican-American Political Elites toward the Hispanic Label. Latin American Perspectives. 1992;19(4):45–58.
- Gomez L. Manifest Destinies: The Making of the Mexican American Race. NY: New York University Press; 2007.
- Grebler L, Moore JW, Guzman R. The Mexican American People: The Nation's Second Largest Minority. NY: Free Press; 1970.
- Gross A. Texas Mexicans and the Politics of Whiteness. Law and History Review. 2003;21(1):195–205.
- Guzman B. The Hispanic Population. Washington DC: U.S. Dept. of Commerce, Economics and Statistics Administration, U.S. Census Bureau.; 2001.
- Haney-Lopez I. White Latinos. Harvard Latino Law Review. 2003;6:1–7.
- Haney-Lopez I. White By Law: The Legal Construction of Race. Second. NY: New York University Press; 2006.
- Harrison MS, Reynolds-Dobbs W, Thomas KM. Skin Color Bias in the Workplace: The Media's Role and Implications Toward Preference. In: Hall R, editor. Racism in the 21st Century. Springer; 2008.
- Hersch J. Profiling the New Immigrant Worker: The Effects of Skin Color and Height. Journal of Labor Economics. 2008;26(2):345–386.
- Hochschild J, Powell BM. Racial Reorganization and the United States Census 1850–1930: Mulattoes, Half-Breeds, Mixed Parentage, Hindoos, and the Mexican Race. Studies in American Political Development. 2008 Spring;22:59–96.
- Jiménez T. Immigrant Replenishment: Mexican Americans, Immigration, and Identity. Berkeley: University of California Press; 2010.
- Martinez GA. The Legal Construction of Race: Mexican-Americans and Whiteness. Harvard Latino Law Review. 1997 Fall;2:321–347.
- Massey DS. Racial Formation in Theory and Practice: The Case of Mexicans in the United States. Race and Social Problems. 2009;1:12–26.[PMC free article][PubMed]
- Mechaca M. The Mexican Outsiders: A Community History of Marginalization and Discrimination in California. Austin: University of Texas Press; 1995.
- Montejano D. Anglos and Mexicans in the Making of Texas: 1836-1986. Austin, TX: University of Texas Press; 1987.
- Munoz C. Youth, Identity, Power: The Chicano Movement. London: Verso; 1989.
- Murguía E, Telles EE. Phenotype and Schooling Among Mexican Americans. Sociology of Education. 1996 Oct;69:276–289.
- Ngai ME. Impossible Subjects: Illegal Aliens and the Making of Modern America. Princeton, NJ: Princeton University Press; 2004.
- Pager D, Hana S. The Sociology of Discrimination. Annual Review of Sociology 3. 2008;4:181–209.[PMC free article][PubMed]
- Perlmann J. Italians Then, Mexicans Now: Immigrant Origins and Second-Generation Progress 1890 to 2000. NY: Russell Sage Foundation; 2005.
- Reimers C. Labor Market Discrimination Against Hispanic and Black Men. The Review of Economics and Statistics. 1983;65(4):570–579.
- Rodriguez C. Changing Race: Latinos, The Census, and The History of Ethnicity in the United States. NY: New York University Press; 2000.
- San Miguel G. Let All of Them Take Heed: Mexican Americans and the Campaign for Educational Equality in Texas 1910-1981. Austin, TX: University of Texas Press; 1987.
- Sanchez GJ. Becoming Mexican American: Ethnicity, Culture, and Identity in Chicano Los Angeles 1900-1945. NY: Oxford University Press; 1993.
- Telles E, Murguía E. Phenotypic Discrimination and Income Differences Among Mexican Americans. Social Science Quarterly. 1990;71(4):682–696.
- Telles E, Ortiz V. Generations of Exclusion: Mexican Americans, Assimilation, and Race. New York: Russel Sage Foundation; 2008.
- Vasquez J. Blurred Borders for Some but not “Others”: Racialization, “Flexible Ethnicity,” Gender, and Third-Generation Mexican American Identity. Sociological Perspectives. 2010;