Understanding Bias toward Latinos: Discrimination,

Dimensions of Difference, and Experience of Exclusion

John F. Dovidio∗ and Agata Gluszek

Yale University

Melissa-Sue John

University of Connecticut

Ruth Ditlmann and Paul Lagunes

Yale University

White – Black relations have historically been the defining form of intergroup rela-

tions in the study of prejudice and discrimination. The present article suggests that

there are limitations to applying this model to understanding bias toward other

groups and proposes that a comprehensive view of the dynamics of the Anglos’

bias toward Latinos requires consideration of the distinctive elements of this form

of intergroup relations. In four empirical studies, we experimentally document

discrimination against Latinos (Study 1), explore the potential dimensions that

underlie bias against Latinos (Study 2), and examine the effect of a particular so-

cial identity cue, accentedness, on perceptions of acceptance and belongingness

of Latinos and members of other groups (Studies 3 and 4). These studies consider

general processes of prejudice and identify how particular facets of bias against

Latinos can shape their experiences and, taken together, illustrate how under-

standing bias against Latinos can reciprocally inform contemporary theories of

prejudice.

White – Black relations have historically been the defining form of intergroup

relations in the United States socially, economically, and politically. Jones (1997)

explained that historically in the United States “blacks have been the largest single

∗ Correspondence concerning this article should be addressed to John F. Dovidio, Depart-

ment of Psychology, Yale University, P.O. Box 208205, New Haven, CT 06520 [e-mail: john.

dovidio@yale.edu].

Preparation of this article was supported by NSF grant BCS-0613218 awarded to the first author.

minority group, have been highly salient politically, and have been the focal point

of legislative, political, judicial, and economic systems. . . . the black experience is

a paradigm for social justice and policy beliefs in the United States” (pp. 18 – 19).

Perhaps because of this central role in the U.S. social experience, Black –

White relations have also been the traditional focus of psychological research

on prejudice and discrimination, while research on a group that has also ex-

perienced prejudice and discrimination—Latinos—has been surprisingly rare.

Ramirez (1988) found that only 7% of the social psychological work on prejudice

was conducted with Latino targets. Over 20 years later, a review of the articles on

prejudice and discrimination in three leading social psychology journals (Journal

of Experimental Social Psychology, Journal of Personality and Social Psychology,

and Personality and Social Psychology Bulletin) revealed that the vast majority

of articles in this area focused on Blacks as targets (61%). Similar to Ramirez

(1988), only 7% of articles focused on Latinos.

In addition to dominating the empirical study of prejudice and discrimination,

Black – White relations have shaped the foundation of theoretical understanding of

prejudice and bias. Current models of Anglos’ (ethnic majority group members’)

attitudes toward Latinos and Latino immigrants typically derive directly from

models of White – Black relations in the United States. Research has generally

emphasized the role of common underlying processes, such as intergroup threat,

that contribute to bias across groups (e.g., Z´

arate, Garcia, Garza, & Hitlan, 2003).

Although there is substantial value in emphasizing the common elements

of bias across groups and promoting general models of intergroup relations, a

comprehensive view of the dynamics of the attitudes and orientations of Anglos

toward Latinos requires consideration of the distinctive elements of this form of

intergroup relations. The current work thus examines the ways that attitudes of

Anglos toward Latinos and Latino immigrants are similar in nature and dynamics

to other forms of prejudice but also focuses on additional factors that may be

particularly relevant for understanding bias toward Latinos.

Directing more attention to bias toward Latinos is timely, practically as well as

theoretically. While Black – White relations continue to be of critical importance

in the United States, the Latino population in the United States has grown in size

and influence. Latinos have surpassed Blacks as the largest minority group in the

United States. Latinos currently constitute 15.4% of the U.S. population, whereas

non-Latino Blacks 12.2% (US Census Bureau, 2009). At the same time, leading

indicators of economic and social disparities (US Census Bureau, 2007) reveal that

Latinos, like Blacks, lag behind Whites in earnings (median income based on the

race of the householder: Blacks, $34,001, Latinos, $40,766, Whites, $55,096) and

educational attainment for those over 25 years old (BA or higher degree: Blacks,

17.3%; Latinos, 12.5%; Whites, 30.5%).

In the current article, in four empirical studies, we examined contempo-

rary discrimination against Latinos and the experience of Latinos in the United States. Study 1 investigated the operation of bias against Latinos relative to

Anglos (Whites) in a field setting. Study 2 explored the potential dimensions

that can underlie bias against Latinos. Studies 3 and 4 examined the effect of a

particular social identity cue, accentedness, on perceptions of acceptance and be-

longingness of Latinos and members of other groups. Taken together, these studies

consider how understanding the general processes of prejudice and bias enhances

recognition of discrimination against Latinos, as well as how appreciating some

of the particular facets of bias against Latinos can shape their experiences and

reciprocally inform contemporary theories of bias.

Direct empirical evidence of discrimination against Latinos is relatively

sparse, despite the fact that Latinos represent the largest racial/ethnic minority

group in the United States, exhibit substantial social and economic disadvantage,

and report experiences of differential treatment comparable to those reported by

Blacks (Perez, Fortuna, & Alegria, 2008). We thus first explore evidence of con-

temporary discrimination again Latinos (Study 1), and then we consider the factors

that contribute specifically to this type of bias in subsequent studies (Studies 2 –

4).

Study 1: Bias toward Latinos

While there is clear evidence of negative attitudes (prejudice) and stereotypes

toward Latinos (Deaux & Ethier, 1998), documentation of discriminatory behavior

is rare. Archival evidence often implicates discrimination against Latinos, for ex-

ample in earnings of Latino-owned firms (Carvajal, 2004) or in capital sentencing

(Lee, 2007). Experimental research has revealed discrimination against Latinos in

simulated jury decision-making contexts (Bottoms, Davis, & Epstein, 2004) and

when applying for low-wage jobs (Pager, 2007). However, other studies reveal

that Latinos are evaluated comparably to Anglos for equivalent performance (e.g.,

Romero & Garza, 1986).

One explanation for the inconsistent evidence of bias against Latinos in lab-

oratory studies is that participants are often concerned about revealing their true

underlying prejudice (Dovidio & Gaertner, 2004). Consistent with this reasoning,

discrimination against Latinos is more likely to be observed when participants

are unaware that their social biases are being examined and/or the context offers

justifications other than ethnicity for responding negatively. In Study 1, we thus

examined potential discrimination against Latinos in a field experiment (Ditlmann

& Lagunes, 2009).

In particular, Study 1 investigated differential treatment of Latino shoppers.

We recorded the responses of cashiers to requests by one of three Anglo and three

Latino male confederates to purchase a gift certificate.

Method

Participants. The participants were cashiers in 217 stores in four different

locations in Connecticut. Latino confederates visited 111 stores and Anglo con-

federates visited 106. The stores all indicated over the telephone that they sold

$10 gift certificates and accepted personal checks as a valid form of payment.

Procedure. Three male Latino and three male Anglo confederates, matched

on five key characteristics (age, weight, attractiveness, extroversion, and mas-

culinity), were selected from a pool of 40 Latino and Anglo men on the basis

of three raters’ agreement that they would be readily perceived as Latino or

Anglo. These confederates received extensive training that involved memorizing

and practicing the script through role-play and at actual stores, and standardizing

confederates’ appearances and verbal and nonverbal behaviors. To keep confed-

erates unaware of the purpose of the study, confederates were separated during

training and the execution of the study. None suspected the true purpose of the

research. In addition, 34 of the 217 interactions in the stores were monitored by a

trained observer without the knowledge of the confederates. The observers found

that the confederate followed the protocol in all cases.

The confederates, who were randomly assigned to stores, first requested to

purchase a $10 gift certificate. If the salesperson responded that the minimum

dollar amount was larger than $10, the confederate was instructed to accept this

higher limit. The confederate then attempted to pay with a personal check. The

main measures of bias against Latinos were (a) the minimum dollar amount

confederates were quoted for the gift certificate, and (b) the frequency with which

Latino confederates, relative to Anglo confederates, were asked for identification

when they paid with a personal check.

Results

Overall, Latino confederates were treated differently than Anglo confederates

on both measures of bias. Despite the fact that all of the stores included in the study

offered $10 gift certificates, logistical regression analysis revealed that salespeople

quoted a minimum dollar amount higher than $10 more frequently for Latino than

for White confederates, 9% versus 2% of the time, b

= 1.64, SE = .78, p < .05.

Salespeople also asked for identification more frequently when the confederate

was Latino than Anglo, 85% versus 73%, b

0.69, SE .34, p < .05.

Discussion

Taken together these findings demonstrated that salespeople treated Latino

confederates differently and less positively than they treated Anglo confederates.

Thus, paralleling the patterns of disparities in major social indicators (e.g., in-

come) and perceptions of experiencing discrimination (Perez et al., 2008), Latinos

may experience discrimination today much as Blacks do. The nature of these

contemporary biases is relatively subtle; Latino confederates were not openly re-

fused service nor were they the target of blatantly racist remarks by salespeople.

Nevertheless, with respect to racial discrimination, Dovidio and Gaertner (2004)

noted that subtle expressions of bias can have negative consequences as insidious

as blatant discrimination. Moreover, microaggressions of this type can be particu-

larly cognitively and emotionally taxing for members of stigmatized groups (Sue

et al., 2007; see also Quintana, Herrera, & Nelson, this issue). Thus, even slight

but consistent disparities in treatment can have a cumulative negative effect for

Latinos in terms of psychological well-being (e.g., Finch, Kolody, & Vega, 2000),

social adjustment (e.g., Greene, Way, & Pahl, 2006), and aspirations (e.g., Devos

& Torres, 2007) over time.

While Study 1 suggests parallels in the discriminatory treatment of Blacks and

Latinos and its consequences, we note that the underlying dynamics of discrimi-

nation that lead to actual and perceived exclusion may differ somewhat between

the groups. The nature of Whites’ attitudes toward Blacks and their intergroup

relations have been shaped by the legacy of slavery and the historical relationships

between Blacks and Whites throughout the history of the United States (see Jones,

1997). Black Americans generally share common language, history, and political

destiny with White Americans. By contrast, the Latino population has shown dra-

matic growth in the United States relatively recently, and it is culturally diverse,

reflecting different histories than that of Anglo Americans.

In addition, recent concerns about illegal immigration can play a critical role

in the attitudes of Anglos toward Latinos in general (see also Z´

arate & Shaw,

this issue). Latinos constitute 80% of the estimated 11.6 million undocumented

immigrants in the United States who have been the focus of considerable de-

bate and criticism (Deaux, 2004; Hoefer, Rytina, & Baker, 2009). In 2007, the

issue of illegal immigration continued to be of “great importance” or “very im-

portant” for 62% of respondents in a Gallup survey (Carroll, 2007), and 43% of

respondents believed that the United States had lost ground with regard to ille-

gal immigration (Saad, 2007). Substantial portions of the U.S. population blame

undocumented immigrants for taking jobs away from Americans, suppressing

wages, contributing to higher unemployment, and imposing a fiscal burden on

other tax payers (Deaux, 2006). It is likely that negative attitudes toward un-

documented immigrants extend to Latino immigrants in general because of how

closely associated these two groups are. Thus, bias toward Latinos may involve

an additional dimension, one of foreignness, compared to bias against Blacks.

Our next study represented our initial exploration of the different dimensions

that may underlie prejudice and discrimination against different racial and ethnic

groups.

Study 2: Dimensions of Difference

The United States is a pluralist society composed of many identifiable ethnic

groups. However, despite demographic trends producing unprecedented racial and

ethnic diversity, the standard for being American remains basically singular—

being White (Devos & Banaji, 2005; Sidanius, Feshbach, Levin, & Pratto, 1997).

However, the distinction between which groups are seen as American is not a

simple in-group – out-group perception. Various minority groups in the United

States may differ systematically in how “American” they are perceived to be by

Whites. Devos and Banaji (2005), for example, found that people were more likely

to pair White with American than either Asian or Black with American on implicit

tasks, thus demonstrating that White is the prototypical image of American. Using

a direct measure of similarity of various groups of citizens to Americans (1

disagree strongly to 7

= agree strongly) in a preliminary study, we found that

White Americans were judged most similar to—virtually synonymous with—

Americans (M

= 6.69). Black Americans were rated next most similar (M =

6.17) but significantly lower than White Americans; they were followed by Latino

Americans (M

= 5.88), Asian Americans (M = 5.79), Native Americans (M =

5.67), and Arab Americans (M

= 4.94). Study 2 therefore examined the different

“shades” of being perceived as American for different racial/ethnic groups and

explored the dimensionsthatdistinguishthesegroups. Despite increasing pluralism in the United States, as we observed earlier,

models of intergroup relations have focused primarily on Black – White relations.

Nevertheless, the underlying causes of prejudice and discrimination may differ.

Social groups have different histories and varying levels of contact, power, and

status; these variables play an important role in how people are categorized and

stereotyped. Our research thus focused on determinants of how people respond to

members within an inclusive ingroup category, Americans. Studying who is con-

sidered American and how that judgment is determined can enhance understanding

of both the unique and common influences that shape bias toward members of

various groups, including Latinos.

Much of the previous research on the extent to which members of different

groups are perceived to be American has focused on a single dimension of differ-

ence (Devos & Banaji, 2005; Sidanius et al., 1997), but recent research suggests

that more than one dimension may be involved in national definitions. In partic-

ular, Pehrson, Vignoles, and Brown (2009) identified two dimensions of national

inclusion, ethnic nationalism and civic nationalism. Ethnic nationalism defines

national inclusion in terms of shared ancestry, often reflected by physical appear-

ance and language as well as by formal heritage. From this perspective, group

membership has a strong essentialist quality, in which ethnic as well as racial

groups are often differentiated along a “biological” continuum (Jost & Hamilton,

2005). For example, until recently Germany relied on a model of citizenship that

Bias toward Latinos 65

was based primarily on jus sanguinis (right of blood), in which citizenship was

restricted to the natural offspring of citizens. Civic nationalism, in contrast, defines

national group membership in terms of commitment to the ideals and standards

perceived to define a nation. Ideology rather than biology represents the defining

basis for national inclusion.

Although these dimensions are sometimes perceived as oppositional, more

recently they have been conceptualized as separate, independent dimensions un-

derlying nationalistic expressions (Pehrson et al., 2009). Indeed, Devos and Banaji

(2005, Study 1) found that items related to ethnic identity/nativity, such as “being

born in America” and “having lived in America most of one’s life,” and items

associated with civic orientation, such as “respect America’s political institutions

and laws” and “be patriotic,” loaded on different dimensions in a factor analysis

of items defining American identity.

Thus, within the context of relations between two groups, such as Blacks

and Whites, only the single dimension of “distance” may be relevant. However,

when multiple group relations are considered more dimensions can emerge. The

goal of Study 2 was to explore different dimensions that determine how various

groups deviate from the American prototype. This study used college students as

participants for comparability to previous work (e.g., Devos & Banaji, 2005) and

to assess a newer generation’s intergroup perceptions. These data were collected

as part of a larger project on intergroup perceptions (John, 2009). We hypothesized

that these two dimensions, ethnic and civic, define not only how much different

groups are seen as American but also the fundamental ways in which these groups

differ from the American prototype. That is, we propose that the distance from the

American prototype varies in two-dimensional space reflecting ethnic and civic

(ideological) dimensions. For instance, Black Americans are likely to differ more

from White Americans (who are expected to most closely represent the prototype

of Americans) than would Latino Americans on the ethnic dimension. However,

because of their long history in the United States with Whites, Black Americans

would be expected to diverge less than Latinos on the civic dimension.

Method

Participants. Sixty White college students (31 men and 29 women) between

the ages of 18 and 22 participated in the study for course credit.

Procedure. Participants completed a survey in which they were asked to

indicate the similarity of 22 social groups with each other, on a 1 (disagree strongly)

to 7 (agree strongly) scale. Groups of primary interest included American, White

American, Black American, Latino American, Asian American, Native American,

and Arab American.

Results

Multidimensional scaling, ageometricdatamappingtechniquefordatathat

expressthedistancesamongtheobjectsofaset,wasusedtoanalyzethesimilarity

ratings. Evaluationofthescreeplotindicatedthatthereweretwoprimaryunder-

lyingdimensions. Figure1showsthesevengroupsof primaryinterest mapped

ontothistwo-dimensional space. Thehorizontal axisappearstoreflect acivic

dimension, whereastheverticalaxisrepresentsanethnicdimension.

Ashypothesized,andconsistentwiththefindingsofDevosandBanaji(2005)

andSidaniusetal. (1997), WhiteAmericansappearmostprototypicalofAmeri-

cans;theyarethegroupclosesttoAmericaninthistwo-dimensionalspace.Asian

AmericansandBlackAmericansbothdifferedsubstantially(but indifferent di-

rections)fromWhiteAmericansontheethnicitydimension,butBlackAmericans

werecloseronthecivicdimension. Ofallofthesegroups, ArabAmericanswere

furthest awayfromAmericansandWhiteAmericansonthecivicdimension. Of

particularrelevancetothefocusofthepresentarticle,LatinoAmericansdeviated

modestlyfromWhiteAmericans(andBlackAmericans)onthecivicdimension

butsubstantiallyfromWhiteAmericansontheethnicdimension.Thus,whilethe

absolutedistancesnicelyreplicatedpreviousresearchonthedifferent degreesto

which various groups reflect “Americanness,” this study extended the previous

work by identifying two fundamental dimensions that determine how much and

in what ways different groups deviate from the American prototype.

Discussion

We believe that these findings have important implications for understanding

the ways that different groups may be perceived and treated, as well for how

different minority groups may experience stigma. With respect to differential

treatment, the nature of bias expressed toward Blacks, Latinos, Asians, Arabs,

and Native Americans may vary systematically as a function of how these groups

deviate from Whites on the ethnic and civic dimensions. When groups differ from

Whites on dimensions related to ethnicity, Whites may be particularly likely to

attribute the negative actions of members of these groups to negative dispositional

characteristics, reflecting the essentialist quality of this dimension (e.g., Hewstone,

1990). Because people make stronger dispositional attributions for more novel

and distinctive stimuli, our analysis further suggests that this type of intergroup

bias may be more pronounced for groups perceived to deviate more from White

Americans on the ethnic dimension (e.g., more for Blacks than for Latinos).

Distances from White Americans on the civic dimension may also system-

atically affect the degree and nature of intergroup bias (see also Ryan, Casas,

& Thompson, this issue). With respect to the degree of bias, when Whites per-

ceive that an action violates conventional standards (e.g., a crime is committed),

Whites may be more sensitive to detecting these violations and may respond more

negatively to members of groups more distant from White Americans on this di-

mension; these groups are already perceived to deviate more from American moral

standards. Thus, Whites may punish members of different groups more severely

for moral violations as a function of the group’s distance from White Americans

or the American prototype on this dimension. Future research might consider

whether individual differences in the extent to which Anglos associate Latinos

with illegal immigration increase their perception of the distance of Latinos from

White Americans on the civic dimension, and how this distance relates to the

treatment of individual Latinos and support for policies that benefit the group as

a whole.

Moreover, people may respond in qualitatively different ways to violations

of standards by members of groups viewed as close to or distant from White

Americans on the civic dimension. Specifically, whereas punishment may be

direct and corrective for close groups (e.g., Blacks), people may instead choose to

isolate and ostracize members of other groups who are perceived to deviate from

the civic standards of prototypical Americans. For instance, people may be more

likely to recommend deportation of a Latino American than a Black American for

committing a crime, even though both are American citizens and thus cannot be

legally deported. Thus, a two-dimensional perspective on group relations suggests

some novel directions for future research on discrimination.

Members of minority groups are also vigilant to differential treatment by

majority group members and fear rejection associated with stigmatization of their

group (Mendoza-Denton, Downey, Purdie, Davis, & Pietrzak, 2002). Thus, minor-

ity group members may be particularly sensitive to behaving in ways that increase

the salience of how their group might differ from the American prototype on

the civic or ethnic dimension. With respect to the civic dimension, for example,

minority group members who feel socially vulnerable in a majority group con-

text may affirm conventional, majority-group standards even more emphatically

than majority group members. Cues of deviance along the ethnic dimension, such

as comments related to physical difference, may similarly lead minority group

members to emphasize the ways that they are similar to the American prototypic

standards (Cheryan & Monin, 2005). In the next study, we investigated one impli-

cation of this analysis. Specifically, we examined how self-perceptions of accents,

which represent an indicator of difference or foreignness, relate to feelings of

belongingness and social acceptance among members of different groups.

Whereas Study 2 suggested a broader framework for studying intergroup re-

lations beyond the traditional Black – White paradigm, Studies 3 and 4 focused on

one aspect of that framework. Ethnic difference defined one of the two fundamen-

tal dimensions determining where different groups were situated relative to the

American prototype. Studies 3 and 4 thus considered the relationship of a com-

mon cue of ethnic difference, possessing a nonnative accent, with how members of

various groups, including Latinos, feel included or excluded in the United States.

Specifically, the goal of Studies 3 and 4 was to examine the association between

participants’ perceptions of their own accents and social belonging (Study 3) and

perceived discrimination (Study 4). Study 3 explored the responses of people with

nonnative accents generally; Study 4 included enough participants with nonnative

accents to examine the responses of Latinos, Asians, and Europeans separately.

The data for these two studies represent a subset of data that are being collected

for a larger project (Gluszek & Dovidio, in press).

stereotypes of blacks, mexicans, asians, gays, and blondes

Word Search

Ethnic group: A social category of people who share a common culture, for example, a common language or dialect; a common nationality; a common religion; and common norms, practices, customs and history.

Race: A group treated as distinct in society based on certain characteristics, some of which are biological, that have been assigned social importance.

Prejudice: Evaluation of a social group and the individuals within it based on conceptions about the social group held despite facts that disprove them; the beliefs involve both prejudgment and misjudgment.

Discrimination: Overt negative and unequal treatment of the members of some social group or stratum solely because of their membership in that group or stratum.

Racism: The perception and treatment of a racial or ethnic group, or member of that group, as intellectually, socially, and culturally inferior to one’s own group.

Anti-Semitism: The hatred of Jewish people.

Segregation: The spatial and social separation of racial and ethnic groups.

Assimilation: Process by which a minority becomes socially, economically, and culturally absorbed within the dominant society.

Pluralism: The maintenance and persistence of one’s culture, language, mannerisms, practices, art and so on.

Affirmative Action: A heavily contested program for change, is a race-specific policy for reducing job and educational inequality that has had some limited success.

RACISM: Song lyrics

I can't believe what they did what I saw on T.V. in the 20th century They beat a man like a dog to the ground, yeah, they made him bleed what does it mean

I just want to tell the world to stop  all the hating wars it's only tearing us apart If we're going to live together on this earth We need to learn to stop Racism

On the news everyday someone blows someone away It's all because of hate People persecute and fear, discriminate, what is our fate

Iran, Iraq,Bosnia, South Africa the IRA to name a few today So much to hate,worth killing for Your colored skin, your God and more what is this for

Tim Lamberson 1994

Chapter 10: Race and Ethnicity

I chose to make this post on chapter ten because I feel that the prejudices and stereotypes towards minority, racial and ethnic groups in the United States need to come to an end. I also feel that one of the best ways to combat this is by educating people on the matter. This is why I decided to talk about racism, racial stereotypes, and prejudice and discrimination.

The picture at the beginning of the post shows children from all over the world standing together. Each of them come from very different backgrounds and are all part of different ethnic groups. The picture symbolizes the idea that it doesn't matter what ethnic or racial group you are a part of; we are all equal. The article just beneath the picture focusses on the discrimination against latinos in the United States. I specifically chose this article because chapter ten discusses Latino Americans and how they came to immigrate to the United States and I wanted to give readers a better understanding of some of the ways they are often discriminated against. The article also includes the results of experiments that have been conducted on behalf of Latinos and how they are often treated in certain situations. The next thing I included in my post are the two links to websites that are related to the topics found in chapter ten. The first one is a website regarding the different ethnic groups in the United States. I chose to include this website because it does a wonderful job in providing information on the history of the groups and provides links to other sites that go into even greater detail on the discrimination they often face. I think people often forget that prejudice and discrimination towards specific ethnic and racial groups happens all over the world in all different countries, not just in the U.S. That is why I included this website. It gives examples of the prejudices that occur in different countries. The fourth thing I included in my post is a video that gives examples of racial stereotypes. I found this video interesting because it showed people of different races and beliefs talking about how they are often stereotyped. The way the video was set up allowed the viewer to see how real and painful these stereotypes are to them. Underneath the video is the word-search filled with words that represent the ideas found in chapter ten the most accurately. Lastly, I chose to incorporate the song "Racism" because it represents the feelings that a lot of people have towards racism. There are a lot of people out there who want racism and the discrimination of others to come to an end.

Sociology 101

Sunday, April 3, 2011

Global Concerns and Predictions

Chapter 10: Race and Ethnicity

stereotypes of blacks, mexicans, asians, gays, and blondes