Introduction

Adolescence is a period when identity is focal, when young people explore who they are, what groups they belong to, and how they and others like them fit in the larger social order. For many, exploring ethnic identity is part of that process and there is a rich literature on this subject, including the role that discrimination plays in ethnic identity formation (see Hughes et al. 2006 for review). Less is known about how adolescents feel about their treatment as Americans, that is, whether they believe that the tenets of the American social contract of fair treatment and equal opportunity apply to all. This dearth of information exists, despite the fact that our collective identity as Americans is defined by the diversity of our ethnic roots. As the political theorist, Michael Walzer (1990), writes in his essay, “What does it mean to be an ‘American’?” in contrast to many other nations, the United States is a nation of immigrants from other homelands. Our ethnic identities fulfill “very American needs” (Waters 1990, p. xiii).

In this article, we argue that ethnic minorities have unique insights into the relationship between being both ethnic and American. By virtue of being members of ethnic minority groups they are attuned to ways that prejudice divides us and are more committed to the principles of tolerance that bind us (Walzer 1997). In the current study we consider the role of prejudice as it relates to one aspect of ethnic identity, i.e., ethnic awareness, for adolescents from four different ethnic backgrounds: African American, Arab American, Latino American, and European American. In addition, we explore the association between experiences of discrimination and parental admonitions about dealing with prejudice that the youth report, their beliefs about whether America and her institutions treat all groups fairly, and their civic commitments. We argue that experiences of prejudice play a particular role in the awareness that adolescents from ethnic minority backgrounds have of their ethnic group and of its status in American society and that such experiences are politicizing, that is, are associated both with the adolescents’ beliefs that American institutions fail to treat all groups equally and with their commitments to redress those inequities. Although group discrimination is a motivator of political action, we have no developmental studies linking youths’ experiences with prejudice, their ethnic awareness, and their civic commitments. The goal of this study is to begin to address this gap.

Although the majority of Americans have some ethnic roots outside of the United States, ethnic awareness tends to be more salient for members of ethnic minority groups. For many European American adolescents the term “ethnic” carries less personal meaning and some even assume that it applies only to ethnic minorities (Phinney 1989). The meaning of ethnic awareness is complicated by the diverse experiences of different ethnic groups (Phinney 1990). Nonetheless, we believe that a common element of the ethnic minority experience is a vexed relationship with the Eurocentric culture which dominates American economic and political life (Garcia and Hurtado 1995), i.e., a sense that full membership in American society is contested and that the principles of the social contract do not apply equally to all groups (West 1993). Personal experiences of prejudice are likely to be implicated in that awareness, as this quote from a 17-year old Latina suggests: “Before I knew anything about how the American government worked, I could tell Chicanos didn’t have much say in how things got done ‘cause of the way Anglo people would treat us” (Sanchez-Jankowski 1992, p. 84). In summary, for adolescents from ethnic minority backgrounds, personal experiences of discrimination may carry political overtones if those experiences are interpreted as reflective of the second class status of one’s ethnic group.

Adolescence is a period when youth are able to connect the dots (Brown and Bigler 2005). Compared to younger children, adolescents are more capable of seeing the links between personal experiences of prejudice, membership in a particular racial/ethnic group, and the status of that group in the larger society. Furthermore, it is in adolescence when the political implications of this awareness begin to crystallize (Ginwright et al. 2006; Watts and Flanagan 2007). Adolescents’ more mature understanding of race and ethnicity flows, in part, from their social cognitive capacities and ability to understand social groups and categories (Quintana 1998). It is also in adolescence when society reacts to ethnic minority youth with signals about their group membership and its status. As Tatum (1997, p. 55) summarizes in describing the experience of African American adolescents, “the environmental cues change and the world begins to reflect (one’s) blackness back to (her or him) more clearly.” Compared to younger children, even early adolescents have a greater understanding of and report more experience with racial discrimination (Cooper et al. 2008). The developmental imperatives of adolescence—to craft one’s identity, to interpret the personal meaning of membership in one’s ethnic group, and to deal with discrimination, combine to motivate discussions with parents about race and ethnicity (Hughes and Johnson 2001). There is ample evidence that parents are sensitive to such developmental imperatives, tailoring their practices, especially discussions about discrimination and preparation for bias, such that adolescents are more likely than younger children to hear them (Hughes et al. 2006; McHale et al. 2006).

Role of Prejudice in Ethnic and Political Awareness

Personal experiences of prejudice are likely to be implicated in the adolescent’s encounter with what it means to be a member of his or her ethnic group, which may lead to further exploration of identity (Cross 1991; Tatum 1997) as well as to questioning one’s allegiance to the dominant culture and its worldview (Cross 1991; Marshall 1995). Experiences of prejudice are a common denominator even for the most highly achieved minorities (West 1993). In fact, middle-class ethnic minority parents are more likely than their counterparts from lower SES backgrounds to prepare their children for bias (Hughes et al. 2006), perhaps because they more often move in mixed ethnic groups and have regular encounters with the majority culture. Since African Americans are more likely than European Americans to report experiences of discrimination (Fisher et al. 2000), it is not surprising that preparing their children for the bias they may encounter in life figures in the socialization practices of these families (Hughes et al. 2006; Peters 1985; Quintana and Vera 1999) and that awareness of discrimination and knowledge about one’s ethnic group are positively linked (Hughes et al. 2006).

Although cultural socialization is the most common way that ethnic minority parents describe the content of their ethnic socialization practices, when given specific lists of behaviors, the majority report that preparing their children to deal with prejudice is an integral part of their parenting (Hughes et al. 2006). Parents’ own encounters with prejudice are often a motivating factor for such discussions. African American parents are more likely to tell their children that they should be prepared to deal with bias if the parents themselves have experienced incidents of prejudice at work (Hughes and Chen 1997). At the same time, children’s experiences with prejudice may motivate discussions with their parents. For example, adolescents who report experiences of discrimination also report that their parents engage in more frequent ethnic socialization practices (Miller and MacIntosh 1999). In summary, parents’ discussions about discrimination convey information about what it means to be a member of an ethnic minority group, about the treatment that members of that group might expect in the larger society, and about ways to deal with those realities.

Dealing with Discrimination

Adolescents’ discussions with parents about race and coping with prejudice buffer the negative effects of being on the receiving end of bias (Fischer and Shaw 1999). In response to hypothetical dilemmas involving discrimination, adolescents whose parents have alerted them to bias are more likely to come up with proactive (and often collective) strategies such as seeking support and problem-solving to deal with it (Phinney and Chavira 1995; Scott 2003). It is noteworthy, however, that, although group discrimination historically has been a basis for political organizing and action (Bobo and Johnson 2000; Sanchez-Jankowski 2002), we have no developmental studies of how experiences of prejudice might affect adolescents’ political views and commitments. In the current study, we address this gap.

By political awareness we refer to adolescents’ views of the social contract, i.e., the ties that bind people together in a society, the reciprocity between committing as a stakeholder to the system and believing that the system is responsive to you. Theories of political socialization hold that the stability of a political system depends on diffuse support among the public, and especially in younger generations, for the principles that make the system work. That is, to remain stable, democratic societies depend on widely shared beliefs among the public that the principles of the political order are true. In the United States, this implies beliefs that America is an equal opportunity society where people, regardless of their background, can get ahead by dint of hard work and education (Flanagan et al. 2007). It also implies that the institutions of American society treat all groups equally. By civic commitments we refer to the personal importance that youth attach to goals of patriotism, of promoting inter group understanding, and of advocating for the rights of their ethnic group (Flanagan and Faison 2001; Flanagan et al. 2007). We ask whether an adolescent’s ethnic background, ethnic awareness, and experiences of discrimination are associated with his/her beliefs in the fundamental fairness of America and her institutions and with his/her goals for civic engagement.

Although adolescence is a time of growth in understanding social institutions and social groups, until recently there has been relatively little attention to the development of political views and civic commitments during this time of life. Even less attention has been paid to the correlates of political views among adolescents from ethnic minority backgrounds. Yet, according to longitudinal studies, affective ties to the political system, beliefs in the principles guiding the system, and the civic commitments that take shape in adolescence are highly predictive of civic participation later in adulthood (Jennings 2002; Smith 1997). Such affective ties and commitments to the polity are shaped by a youth’s sense of connection to community institutions and his/her perception that adult authorities in those institutions treat all groups fairly and with respect (Flanagan and Stout 2008). For example, for ethnic majority but especially for ethnic minority adolescents, both patriotism and the belief that America is fundamentally a fair society are significantly related to youths’ perceptions that their teachers create a civil climate at school where the diverse views and experiences of students are respected and where intolerant acts are not abided (Flanagan et al. 2007).

Although the literature on ethnic awareness has grown in the past several decades, few studies have looked simultaneously at ethnic awareness and beliefs in the tenets of the American promise. One exception is work by Phinney et al. (1994) who looked at the extent to which four ethnic groups of adolescents and young adults identified with their own ethnic group and with American ideals, and how identification with each was correlated with self-esteem. It is important for our study to note that the Identification with American Ideals scale used in the Phinney et al. study asks respondents the extent to which they believe that the ideals and opportunities that define American society apply to them and their ethnic group. Thus, in endorsing the extent to which they identified with American ideals, the respondents were agreeing with items such as: A democratic country in which the laws protect my interests; a land of economic opportunity for me; a society that is accepting and tolerant of my cultural background or ethnic group; a society that is concerned about the welfare of my cultural group (italics not in original). With this in mind, in the current article we assess whether ethnic group membership, ethnic awareness, and experiences of discrimination are related to differences in adolescents’ beliefs about the tenets of the American social contract, i.e., that the United States is a nation of equal opportunity and a level playing field.

Phinney et al. found that African Americans were more likely to identify with their ethnic group but were less likely to believe that the ideals that define American society applied to their group whereas European-Americans showed the opposite pattern. For Asian- and European-Americans, self esteem was positively correlated with beliefs that their ethnic group enjoyed the benefits of the American promise. Other scholars have shown that, although marginalized groups may believe in the promise of the American dream as a general principle (i.e., that it is an equal opportunity society where anyone can succeed), they may be less convinced that those tenets apply in concrete contexts or institutions to the average person or to people “like them” (Flanagan et al. 2003). With respect to education, Mickelson (1990) has shown that, while African American students endorse an abstract belief in the American ethos that hard work and education pay off in success, they also endorse a more concrete belief that education has not paid off for members of their own families, people “like them.” In summary, to understand youths’ beliefs about the social contract, it is important to tap both their general views as well as their views about how the principles of fair treatment play out in specific institutional contexts. With respect to the latter, we focus in this study on adolescents’ views about the government and the police.

National studies of adults point to lower levels of trust in government among ethnic minorities when compared to the majority (Smith 1997). Similarly, in national studies of high school students, Latino and African American adolescents, compared to their European peers, express less trust in government (Baldi et al. 2001) and are more skeptical about the amount of attention the government pays to the average person (Niemi and Junn 1998). Attitudes towards other institutions do not fare much better. Analyses of the Monitoring the Future study of high school students point to the persistently lower confidence that African American adolescents have compared to European Americans in the police, the Supreme Court, and the Justice System (Johnston et al. 1986). These beliefs have a basis in experience. One study of the juvenile justice system found that, even after controlling for the seriousness of the offense and social background factors, African American and Latino youth were more likely than their European American peers to be detained by the police or at other stages in the juvenile justice process (Wordes et al. 1994).

Might this vexed relationship with how American ideals apply to their group affect the civic goals of ethnic minorities? Minorities’ greater awareness of their exclusion might result in an increased commitment to redress inequities, particularly as they affect their group and others like them. According to social identity theory, one way to deal with discrimination is to strongly endorse one’s ethnicity (Tajfel 1981). Although awareness of a minority group’s exclusion may engender negative attitudes towards the majority group in power (Cross 1991), it may also motivate a desire to redress inequities and to promote the rights of the minority group. Historically, advocating for opportunity, power, and a voice for one’s group has been a means through which ethnic minorities have addressed issues that affect their group and have won a place in mainstream American politics (Bobo and Johnson 2000). Research with a wide range of immigrant groups in Florida indicates that youth respond to marginalization by the U.S. state by defending their cultural heritage and asserting their right to be different (Stepick and Stepick 2002). Discriminatory practices also can motivate feelings of group solidarity that overcome internal differences. For example, in studies of immigrant and non-immigrant Latino youth, Bedolla (2000) has shown that opposition to California’s Proposition 187 united Latinos, whether they were legal or illegal immigrants and regardless of their country of origin. These and other studies of immigrants have shown that the children of new immigrants are as likely as their native born counterparts to be civically engaged if engagement includes measures of assisting or advocating for their ethnic group (Lopez and Marcelo 2008; Stepick et al. 2008). Finally, there is a growing literature on political activism among ethnic minority and poor youth which indicates that, when young people see their individual marginalization in the context of their group’s marginalization and learn how to mobilize around this group identity, they can be effective in influencing adults and policymakers (Delgado and Staples 2007; Ginwright et al. 2006; Kirshner 2009; Watts et al. 2003). In summary, experiences of prejudice and ethnic awareness may be associated with perceptions that the system does not treat all groups equally and thus may motivate youths’ civic goals to redress such inequities.

Gender, Age, and Social Class

Although not central to our hypotheses, the literature on ethnic socialization points to factors associated with the likelihood and the content of family practices. For example, parents are more likely to discuss discrimination with older versus younger children. Parental education and income are positively associated with the likelihood of preparing one’s offspring to deal with discrimination (Hughes et al. 2006; McHale et al. 2006). With respect to gender differences, the findings are somewhat mixed but, based on their review of the literature, Hughes and her colleagues conclude that, in African American families, males may hear more messages about racial barriers (and experience more discrimination) and females more about racial pride (Hughes et al. 2006). Besides ethnic awareness, gender is a relevant variable for the socialization of civic commitments. Adolescent girls are more likely than their male peers to endorse civic commitments as personal goals (Flanagan et al. 2007) and both parents and children report a greater emphasis on prosocial behavior in the socialization of daughters when compared to that of sons (Eisenberg and Morris 2004). Likewise, other work has documented an inverse relationship between age and adolescents’ pro social and civic commitments (Flanagan et al. 2004). In light of this literature, we have included gender, age, and parental education as predictors in our models.

Hypotheses

Based on this integrative review of relevant literature, we tested the following hypotheses. First, there is consistent evidence that ethnic identity is more salient for minority youth. Thus, we expect that adolescents from ethnic minority backgrounds will report greater ethnic awareness when compared to that reported by their peers from European American backgrounds. Second, youth from ethnic minority backgrounds will be more likely to report that they or someone close to them has faced prejudice and reports of prejudice will be related to greater ethnic awareness. The former hypothesis is based on a significant body of work on young people and adults indicating that experiences of discrimination are more common for people of color, even when social class differences between groups are controlled. The latter is based on literature indicating that ethnic discrimination and knowledge about one’s ethnic group are positively linked. Since ethnic minorities are more likely to experience discrimination and to be aware of their ethnic identity, we hypothesize that the two may be related, i.e., that ethnic awareness may be correlated with experiences of discrimination.

Our third set of hypotheses focuses on parental socialization practices concerning prejudicial behavior. According to the literature on racial socialization, preparation for dealing with discrimination is a common practice reported by ethnic minority parents, especially by those parents who have themselves experienced discrimination. Thus, in discussions of the barriers that prejudice may pose to them, youth are very likely processing information about how their membership in an ethnic minority group and discrimination are linked. Based on our reading of the racial socialization and ethnic identity literature, we expect greater endorsement of this socialization practice (parental admonitions that prejudice may pose a barrier to them) among youth from ethnic minority backgrounds, among youth who report an instance of prejudice, and among youth who report that they are very aware of their ethnicity.

With respect to parental admonitions against prejudicial behavior toward others, we posit that there will be no differences based on ethnic background, ethnic awareness, or on experiences of discrimination. Insofar as tolerance is a widely shared American value, we expect that, regardless of these factors, adolescents will be equally likely to report that their parents admonish them that prejudice against others, i.e., judging people before getting to know them, is wrong. Finally, based on research pointing to gender differences in the socialization of values, we expect that, on average, females will be more likely than their male peers to hear messages from families about prejudice being immoral. At the same time, racial socialization studies suggest that parents may be more likely to warn their sons than their daughters about the potential barriers that prejudice may pose for them. Thus, we hypothesize that boys will be more likely than girls to report that their parents admonish them that prejudice may pose a personal barrier.

Our fourth set of hypotheses concerns adolescents’ beliefs about the fundamental fairness of the principles of the American social contract. Specifically, we posit that, regardless of ethnic background, youth will endorse the belief that America is an equal opportunity society but that youth who report experiences of discrimination will be less likely to believe that America is an equal opportunity society. Research indicates that beliefs in equal opportunity and hard work are widely shared across groups in the United States and that mediating institutions such as schools reinforce those beliefs. At the same time, studies have shown that, when presented with the concrete practices of specific institutions, minority groups are less likely than the majority to contend that all groups share equally in the American dream. Specifically, when the tenets of the American social contract (equal protection under the law, equal opportunity, tolerance and welfare for all) are framed as applied to oneself or one’s group, ethnic minorities are less likely to endorse such beliefs. Thus, in the current study, we expect that ethnic minorities will be less likely than European-Americans to contend that the government is responsive to the average person and that the police treat all groups with the same standards of fairness. Experiences of discrimination should sensitize youth to inequity. Thus, we expect that reports of prejudice will be related to lower endorsements of the fundamental fairness of the system (that America is an equal opportunity society) and its institutions (that the government and police treat all groups fairly).

Our final set of hypotheses concern relationships between ethnicity, discrimination, and three types of civic commitments. According to developmental research, service in the local community and commitment to the country are highly linked, regardless of ethnicity. In light of this body of work, we hypothesize that there will be no ethnic differences in adolescents’ patriotism (commitments to serve country and community). However, research on the civic commitments and activism of immigrant, ethnic minority, and poor youth points to the need to define civic commitments broadly to include measures of redressing inequality and of advocating for one’s ethnic group. Furthermore, when individuals appreciate the connection between their own experiences of discrimination and the status of their ethnic group in society, that awareness can be a motivator of civic action, especially action on behalf of their group. Thus, we expect that ethnic minority status, greater ethnic awareness, and experiences of discrimination will be positively related to civic goals of promoting racial understanding between groups and to advocating for one’s own group. Finally, several studies have found that females are more likely than males to feel socially responsible and to engage in community work. In light of this work, we expect that females will be more likely than males to endorse the three types of civic commitments.

Methods

Participants

These data were collected in social studies classes before September 11, 2001 and during a period of relative calm in the United States. The project was presented to school superintendents and principals and then discussed with the social studies teachers. The project was described to students and in letters sent home to their parents as a study of adolescents’ opinions about their schools, communities, and about life in America. Written consent was sought from parents as well as students prior to their participation. These active consent procedures resulted in a response rate of 68%. Surveys were administered by research assistants during regularly scheduled social studies classes. With one exception, all students were literate in English. They completed the surveys on their own and were able to ask research assistants questions for clarification during the survey administration. Survey items were read to the one student who was not fluent in English. (For additional details, see Flanagan et al. 2007).

Data were collected from 1,096 adolescents from three communities in the Midwest and Northeastern United States representing African (n = 115), Arab (n = 115), Latino (n = 127), and European (n = 749) backgrounds. Table 1 presents a summary of the sample characteristics. Not surprisingly, there were more recent immigrants in some ethnic groups than others. Whereas, 95% of the European American and 88% of the African American youth reported that their families had been in the United States for three or more generations, only 44% of the Latino- and 31% of the Arab-American youths’ families had resided in the United States that long. The age of participants ranged from 11 to 18, with an average of 15 years. However, there were age differences between groups: adolescents from Arab backgrounds were the oldest followed by European, African, and Latino students. Overall the sample was 53% female but there were more males in the African American and more females in the Arab American sample. Parental education was calculated by averaging participants’ reports of the highest level of education attained by their mother and father. There were significant differences between groups in mean parental education: Arab- and Latino-American adolescents reported significantly lower parental education than their peers in the other two groups. On average, most of the Arab- and Latino-American parents had only finished high school while African- and European-American parents were more likely to have undergone some post-high school training. As a result of these demographic differences by ethnic group, age, gender, and parental education were all entered as covariates in the analyses.

Table 1 Demographic characteristics of the participants
Full size table

Measures

All of the measures used in the present study are based on students’ self-reports, where appropriate, separate alphas were calculated for each ethnic group. The measures of adolescents’ ethnic awareness and experiences of prejudice allowed participants to provide both a quantitative (i.e., binary) response as well as the option to provide (in writing) a more rich qualitative description.

Ethnic Self-identification

Adolescents were asked the following: “People in the United States come from different ancestries. Some of these are listed below. Check what best describes your family’s ancestry.” Choices included the following: African/Black, Arab, Asian, European/White, Mexican/Latin American, Native American/Indian, Puerto Rican, other Caribbean ancestry. There were too few Asians and Native Americans in our sample to form meaningful groups. The few adolescents (n = 27) who identified as Mexican were combined with the larger Puerto Rican group to form the Latino ethnic category.

Ethnic Awareness

To assess participants’ ethnic awareness, they were asked to respond to the following statement: “Some people are very aware of their racial, ethnic, or cultural identity. Others are not. How about you?” Adolescents responded by indicating they were either very aware or not very aware. This was followed by the open-ended item, “Describe some times when you felt like a member of your ethnic/cultural group.” Participants’ responses were captured using 11 different codes. Details are provided in the discussion of open ended responses later in the paper.

Experiences of Prejudice

To measure experiences of prejudice, adolescents were asked the following: “Have you or someone close to you ever faced prejudice?” This dichotomous item (0 = No; 1 = Yes) was followed by an open-ended one, “If yes, can you describe some of these times?” Responses were coded for; (a) identity of the person who experienced prejudice (e.g., self, other, unclear), (b) the context in which the event occurred, (c) the basis of the prejudice message (e.g., personal attributes, race), and (d) description of the abuse (e.g., physical, non-verbal).

Parental Admonitions About Prejudice

Adolescents were also asked two sets of questions about the extent to which parents discussed prejudice with them. Reports that parents taught them that prejudice was wrong were measured with four items such as: “My family tells me it is wrong to judge people before you get to know them” and “My family teaches me to treat everybody equally.” Together these four items make up the Prejudice is Unjust scale (alphas for the four ethnic groups ranged from .75 to .83). A second three-item scale, Prejudice as a Barrier, tapped adolescents’ reports of the degree to which parents warned them that prejudice may pose a barrier to them or others like them (alphas ranged from .66 to .75). Items included reports of parents telling their children about family members who were treated unfairly because of prejudice and warnings that similar biases might impede their opportunities. Participants responded to the items in both of these scales using a 5-point Likert scale ranging from Strongly Disagree (1) to Strongly Agree (5).

Beliefs in the Tenets of the American Promise

Items were developed for this study to gauge adolescents’ beliefs that America is an equal opportunity society, that the government responds to the average citizen, and that the police mete out justice fairly. The Equal Opportunity scale (alphas ranged from .62 to .78) included three items about equal opportunity in America, i.e., that, regardless of their race or social background, all people are given a fair chance and an equal opportunity to get ahead. The Responsiveness of American Government (alphas ranged from .67 to .83) was measured with four items assessing adolescents’ beliefs that the government was responsive to the needs and opinions of average people like themselves. High scores reflect adolescents’ faith that the American government treats all people equally. The Police Mete Out Justice Fairly scale (alphas ranged from .67 to .80) included two items tapping perceptions that the police in their community treat all groups equally. A 5-point response scale was used for all items in these constructs: Strongly Disagree (1) to Strongly Agree (5).

Civic Commitments

Kasser and Ryan’s (1993) method of indexing aspirations based on the personal importance individuals attached to a set of future goals was used in the survey. Items were similar to those used to measure civic goals in other studies of American youth (Johnston et al. 1986). We focused specifically on three civic goals: patriotism, improving race relations, and advocating for one’s ethnic group. Participants responded to all items using a 5-point Likert scale ranging from Not at All Important (1) to Very Important (5). Patriotism (alphas ranged from .75 to .87) was measured with three items which assess the importance adolescents attached to serving their country or helping society. Improving Race Relations, a two-item scale, measured how important an adolescent said it was for him/her to improve understanding between racial and ethnic groups, (alphas ranged from .72 to .80 for three groups and .52 for Latino-Americans). Group Advocacy (alphas ranged from .72 to .83) was a three-item scale, reflecting the importance of personally standing up for or supporting one’s ethnic group. (Information on civic commitments also can be found in Flanagan et al. 2007).

Analytic Strategy

Our analytic strategy involved several steps. In the first step we performed a confirmatory factor analysis (CFA), using the statistical software package Mplus (Muthén and Muthén 1998–2005), in order to show that our measurement model was an appropriate fit for the overall sample. Each of the latent constructs representing the conceptual variables for civic commitments, beliefs about America and American institutions and parental admonitions about prejudice were simultaneously estimated in the measurement model. The results of this CFA are presented in Table 2. All items loaded on their hypothesized construct (all standardized factor loadings >.50) and the model had a good fit to the data considering sample size (χ2(239, N = 1,194) = 680.75, p < .01, CFI = .95, TLI = .93, RMSEA = .039). The latent constructs had moderate to low intercorrelations (<.55), suggesting they were related but independent constructs.

Table 2 Unstandardized and standardized factor loadings for latent constructs for the full sample
Full size table

Statistical Analyses

First, chi-square tests were used to test for differences between ethnic groups in their ethnic awareness and experiences of prejudice. Second, chi-square tests were run to examine, by ethnic group, the relationship between adolescents’ ethnic awareness and whether or not they had experienced prejudice. Third, ANCOVAS were performed to test for relationships of ethnic group, ethnic awareness, experiences of prejudice and adolescents’ (a) reports of parental admonitions about prejudice, (b) beliefs about America and American institutions. To adjust for correlations among our measures of civic commitment, we performed MANCOVAS on the final set of dependent variables. Correlations among the civic commitment dependent variables ranged between .35 and .42 whereas the other sets of dependent variables were correlated less than .32. Gender, parental education, and age were included as covariates in each of these analyses. Differences between groups were identified using Bonferroni multiple comparison tests.

Textual Analyses

A textual analysis was conducted on adolescents’ written open-ended responses to the two qualitative questions: ethnic awareness and experiences of prejudice. A three-stage process of analysis was utilized. First, four independent researchers made a complete pass through the responses to each question to note emerging themes. After discussion of initial impressions, the researchers re-read the responses to derive categories and compared notes to agree on a general coding scheme for each question. This stage of analysis was an interpretive process, meaning the coding scheme was driven entirely by the themes that emerged in adolescents’ responses. Multiple codes were assigned to the ethnic awareness (up to 2) and experience of prejudice (up to 4) responses. All descriptions of adolescents’ reports of prejudice were first coded to identify the target of the prejudice (self, other, unclear) and were then coded to capture the setting, basis of the prejudice message, and description of the experience. Second, two researchers coded a sub sample of responses (n = 50 per question) to check inter-rater reliability. Cohen’s kappa was .84 for the ethnic awareness code and .91 for the experiences of prejudice code. Disagreements were reconciled through discussion. Third, the researchers each independently coded a quarter of the responses. The researchers collectively agreed on how to code any unclear responses. Sixty-one percent of participants (n = 671) responded to the ethnic awareness item, while 85% (n = 467) of participants who indicated they or someone close to them had experienced prejudice (n = 547) provided additional written description of the experience. Discussion of the ethnic awareness and experiences of prejudice codes are limited in this paper, serving primarily to illuminate the close-ended responses.

Ethnic Awareness

Chi-square tests revealed a significant relationship between adolescents’ ethnic background and their ethnic awareness (χ2(3, N = 988) = 71.11). Whereas 69, 83, and 64% of African-, Arab- and Latino-American adolescents, respectively, said that they were very aware of their ethnicity, only 45% of European American adolescents reported the same level of ethnic awareness. Adolescents’ responses to the open-ended item, “Describe some times when you felt like a member of your ethnic/cultural group” indicated that compared to 12, 9, and 18% of African-, Arab-, and Latino-Americans who wrote “always,” only 6% of European Americans did. In contrast, 19% of European- and 16% of African Americans wrote “never” but only 3% of Arab- and 7% of Latino-Americans mentioned “never” feeling like a member of their ethnic group. Other coding categories, presented in Table 3, revealed the contexts or times when adolescents felt like members of their ethnic group. In general, adolescents were most aware of their cultural/ethnic roots when they were participating in cultural customs and activities (e.g., performing folk dances, eating ethnic food) or when interacting with people who share their ethnic background (e.g., “when I’m around my people”). African American adolescents were the most likely (14%) followed by European American youth (11%; compared to 2% of Arabs and 3.2% of Latinos) to describe incidents of exclusion as rousing ethnic awareness. In contrast, it was participation in religious activities that evoked feelings of ethnic group membership in Arab-American adolescents (20%) compared to their African- (3%), Latino- (4%), and European-American (6%) peers.

Table 3 Response frequencies for adolescents’ descriptions of times they felt like a member of their ethnic/cultural group codes
Full size table

Experiences of Prejudice

Our second hypothesis, i.e., ethnic minorities would be more likely to report instances of prejudice, received only partial support. Compared to 66% of both the African- and Arab- American adolescent groups who said that they or someone they knew had experienced prejudice, slightly less than half (48 and 47%, respectively) of Latino- and European- Americans answered “yes” to this item (χ2(3, N = 1,071) = 24.46, p < .001). Analyses of the open-ended responses of adolescents who responded “yes” are summarized in Table 4. European American adolescents were more likely than their peers of other ethnic origins to mention that someone they knew (rather than themselves), or were in the company of, had experienced prejudice. Specifically, 38% of the incidents reported by European American adolescents reflected personal experiences of prejudice compared to 59, 69, and 74% of Latino-, African-, and Arab-American responses. Of the adolescents who provided descriptions of their (personal or indirect) experiences of prejudice, we found that the majority of these discriminatory acts occurred in schools and informal settings such as during sporting events or while hanging out with friends. Restaurants, stores, and banks (i.e., service businesses) were also likely contexts for experiencing prejudice for Latino-, Arab-, and African-American adolescents more so than for European American adolescents.

Table 4 Response frequencies for prejudice experiences codes
Full size table

Experiences of prejudice based on race were the most common across all four ethnic American groups although more African- (69%) and Latino-Americans (74%) described this type of prejudice than Arab- (46%) and European-Americans (50%). A 12-year old Latino American female offered the following example: “I went to New York and an American man was walking down the street and I said hi and he said get away from me nigger.” Prejudice stemming from group stereotypes (e.g., Arabs as terrorists, blonds as unintelligent, Puerto Ricans as poor) were also reported fairly equally across groups. Interestingly, Arab- and European-Americans were the most likely to describe discriminatory experiences related to personal attributes: 16 and 17%, respectively, of experiences compared to less than 4% for the other two ethnic groups. While for most Arab American adolescents theses messages involved references to the scarf, or veil, traditionally worn by Muslim women, European American adolescents’ experiences focused more on personal choices (e.g., music) or disposition (e.g., sexual orientation, being a loner). A 16-year old Arab-American explained, “I am a Muslim woman wearing a veil, and many times in stores or restaurants people will ridicule the way I look.” (A more detailed discussion of the Arab-American adolescents’ reports of prejudice is provided in Wray-Lake et al. 2008).

While the contexts and bases of discrimination experienced by Latino-, Arab-, and African- American participants were similar, European Americans were more likely to refer to instances of intrafamilial prejudice (i.e., within family division such as a grandfather who would not speak to an aunt because she had married someone from a different racial background). The discriminatory behavior most likely to be cited by all groups was abusive language, such as racial slurs and derogatory remarks. Comments by some adolescents suggested this type of prejudice had become so commonplace in their lives that, as one Arab American adolescent said, “I get use to it & try to ignore it,” whereas others, such as one African American adolescent, described how they retaliated by “beating up” the person who made the racially charged remarks.

Ethnic Awareness and Experiences of Prejudice

Next we ran separate chi-square tests for each ethnic group to test for a relationship between ethnic awareness and experiences of prejudice. The results are presented in Table 5. As the table shows, for African- and Arab American adolescents, experiences of prejudice were unrelated to ethnic awareness whereas for Latino- and European- American adolescents, there was a link between the two. Among Latinos, those who reported an instance of prejudice were more likely to say they were ethnically aware whereas those who did not report an instance of prejudice were not aware. Among European Americans, those who reported an instance of prejudice were equally likely to say they were or were not aware of their ethnicity but those who did not report an instance of prejudice were more likely to also say they were not aware of their ethnicity.

Table 5 Relationship between experiences of prejudice and ethnic awareness for four ethnic American groups
Full size table

Parental Admonitions About Prejudice

As expected, there were no significant group differences in the parental admonitions against discrimination that adolescents reported. However, there were differences based on gender and age: females (M = 4.23) were more likely than males (M = 4.01) and younger more than older adolescents to report that they heard these admonitions in their family. As expected, there were significant differences in adolescents’ reports that their families warned them that prejudice might pose a personal barrier for them. Post hoc tests indicated that African- and Latino-Americans were more likely to hear about the personal barriers of prejudice than were Arab- and European-American youth. Ethnically aware youth were slightly more likely to say that their families warned them that prejudice might pose a barrier. In addition, adolescents who reported experiences of discrimination were significantly more likely (M = 3.42) than their peers who did not (M = 2.80) to say that their parents warned them that discrimination might pose a personal barrier. Further, contrary to prediction, females (M = 3.01) were more likely than males (M = 2.74) to report that they heard this warning in their families (Table 6).

Table 6 Parental admonitions about prejudice by ethnic American group
Full size table

Beliefs About America and American Institutions

As the results in Table 7 show, regardless of their ethnic background, adolescents were equally likely to endorse a belief in Equal Opportunity, i.e., that America is basically a fair society where anyone willing to work hard can get ahead. There were, however, significant differences by reports of prejudice with those who reported an instance less likely to believe that America was an equal opportunity society (M = 2.86) compared to peers who reported no instance of prejudice (M = 3.23). With respect to the more specific question of whether the American government is equally responsive to all groups, including people “like them,” there were significant differences between groups with African Americans less likely than Arab- and European-Americans to endorse this belief, although it should be noted that these data were collected prior to September 11, 2001. Finally, contrary to our prediction, neither experiences of prejudice nor ethnic awareness predicted adolescents’ beliefs that the police meted out justice fairly. In fact, the means for all groups were below the mid point on the scale. That said, African American youth were least likely, and significantly less likely than European Americans to believe that the police mete out justice fairly.

Table 7 Beliefs about America and American institutions by ethnic American group
Full size table

Civic Commitments

As expected, adolescents’ patriotic commitments did not vary by ethnicity, by ethnic awareness, or by experiences of discrimination. Age was a significant covariate with younger more likely than older adolescents to endorse patriotic goals. Group membership and ethnic awareness were related to adolescents’ goals of improving understanding between racial/ethnic groups. More ethnically aware youth were slightly more likely to endorse this goal (M = 4.10 vs. M = 3.86 for the unaware group) and African Americans more likely than European Americans to endorse this goal. In addition, gender and age were significant covariates. Females (M = 4.06) were significantly more likely than males (M = 3.66) to want to improve race relations as were younger when compared to older adolescents. Lastly, the goal of standing up for or supporting one’s ethnic group differed by ethnic awareness, ethnic group, and age. Ethnically aware youth were significantly more likely (M = 4.05) than their less aware peers (M = 3.59) to endorse this goal, younger more likely than older, and European American youth were less likely than all other groups to endorse this goal (Table 8).

Table 8 Civic commitments by ethnic American group
Full size table

Discussion

The United States of America is largely a nation of immigrants whose collective identity is captured in the diversity of her ethnic roots (Walzer 1990) and that diversity has been increasing in recent years (Portes and Rumbaut 2006; U.S. Census Bureau News 2006). Furthermore, tolerance of difference and protection of the civil liberties of minority groups from the “tyranny of the majority” is a fundamental founding principle that is supposed to unite us as Americans (Walzer 1997). To understand what the future of democracy in the United States may look like, it seems imperative that we know more about how youth from diverse ethnic backgrounds experience and interpret the principles and responsibilities of the social contract that binds us as Americans. This study sheds light on how the ethnic diversity that describes and the principles of tolerance that bind us as Americans play out in the experiences and views of adolescents from different ethnic backgrounds. What did we learn?

First, consistent with research documenting the greater salience of ethnic identity for ethnic minorities, the three minority groups in this study were more likely than their European-American peers to report that they were very aware of their ethnic identity. However, our hypothesis that experiences of prejudice would be related to greater ethnic awareness received only partial support. African- and Arab- American adolescents were significantly more likely to report instances of prejudice than were Latino- and European- Americans but such experiences were not associated with greater ethnic awareness in the former two groups. In part, this may be due to the fact that there were relatively few African- and Arab-American adolescents who reported that they had not experienced instances of prejudice. In addition, as documented in the literature on ethic socialization, ethnic awareness is fueled not only by prejudice but by many factors with cultural socialization the most common practice mentioned by parents (Hughes et al. 2006).

The ecology of adolescents’ everyday lives also is implicated in their ethnic awareness. Undoubtedly, the fact that the Arab Americans in our sample reside in one of the largest Arab American communities in the United States where their ethnic roots are reinforced in local businesses, language, religion, and media plays an important role in these adolescents’ ethnic awareness (Wray-Lake et al. 2008). Chi-square tests did reveal that, among Latinos and European-Americans, ethnic awareness and reported experiences of prejudice were related. Among Latinos who reported instances of prejudice, 86% also reported that they were very aware of their ethnic identity. Among European Americans who did not report an instance of prejudice, 59% said that they were not very aware of their ethnicity. In summary, ethnic awareness was higher for youth from minority backgrounds than it was for their peers from the majority group but experiences of prejudice were not consistently associated with ethnic awareness.

Second, the open-ended data provided insights into the phenomenon of prejudice as experienced by adolescents from different ethnic backgrounds. When asked to describe an experience of prejudice, youth from the three ethnic minority groups more often reported on their own experiences whereas their peers from the majority (White) group tended to refer to an incident directed at someone they knew or were with when it occurred. The settings most often mentioned by all groups were schools or informal settings with peers and, for ethnic minorities, incidents in businesses or stores also were noted. Compared to the ethnic minority youth, the European-Americans were more likely to refer to instances of prejudice within their nuclear or extended families. Across all four groups, race was the most common theme as well as prejudice stemming from group stereotypes.

What do parents convey about prejudice to their adolescent children? According to adolescents, parents from all ethnic backgrounds admonished their children that prejudice was immoral, i.e., that they should not judge people before getting to know them. Mean levels were quite high for all groups (above 4.0 on a 5-point scale). Consistent with gender socialization research (Eisenberg and Morris 2004), females were more likely than their male peers to say that their parents emphasized this value. And consistent with age differences in pro-social behavior (Flanagan et al. 2004), younger adolescents were more likely than older to say that they heard this value at home.

At the same time, the extent to which parents discussed the personal barriers posed by prejudice varied by ethnicity: Arab- and European- American adolescents were less likely than African- and Latino-Americans to report such discussions. Preparation for the bias they may encounter in life has been noted in other studies as an important component of socialization by ethnic minority parents (Hughes and Chen 1999; Quintana and Vera 1999; Hughes et al. 2006). Yet discussions about how to deal with racial bias are reported more often by African American parents when compared to those from Mexican or Japanese heritage (Phinney and Chavira 1995) or from Dominican or Puerto Rican heritage (Hughes 2003). In our study, there were no differences between African- and Latino-Americans, although the former had higher means, which may reflect a broad set of beliefs and child rearing practices among African Americans that tap into the historical legacy of oppression borne by this group (Sanchez-Jankowski 2002; Ward 1991). These results also have to be placed in historical context. Were this study to be conducted today, we suspect that Arab-American adolescents would be more likely to report that their families discuss ways that discrimination might pose a barrier to them.

Contrary to our hypothesis, females were more likely than males to report that their parents discussed the barriers that prejudice could pose. Although this is a main effect, we did not find any interactions of gender with ethnic group. This finding is inconsistent with gender differences in racial socialization research in which boys are more likely to hear about barriers due to bias (Hughes et al. 2006). Yet, as Stevenson and Davis (2004) argue, gender differences in ethnic socialization are likely to vary according to the contexts in which adolescents experience discrimination. Finally, consistent with other studies (Miller and MacIntosh 1999), our analyses revealed a highly significant relationship between adolescents’ reports of discrimination and of parental discussions about prejudice posing a personal barrier. The fact that these were cross-sectional data limits our ability to determine causality. It may be that adolescents are more attuned to instances of prejudice in families that discuss its barriers or that an experience of prejudice motivates discussions with parents about bias, or both (Cross 1991; Hughes et al. 2006). Whatever the direction, it is clear that prejudice carries a more personal, and we contend, political, meaning for many youth from ethnic minority backgrounds. Besides the highly significant relationship of prejudice with family discussions of the barriers it poses, such discussions also were positively associated with youths’ ethnic awareness, i.e., youth who felt very aware of their ethnicity were more likely to report family discussions of the barriers that prejudice might pose to them, suggesting that discussions of discriminatory barriers and ethnic awareness are linked. In summary, whereas tolerance, a value that unites us as Americans, appears to be emphasized by families regardless of their ethnic background, discussions about the personal implications of prejudice are more likely to occur in families of ethnically aware youth, with youth who experience an instance of prejudice, and in Latino- and African-American families.

With respect to beliefs in the fundamental tenet of the American promise of equal opportunity, there were no ethnic group differences. However, youth who reported instances of prejudice were less likely to feel that opportunities were equal for anyone willing to work hard. With respect to the government’s responsiveness to the average person, African-Americans were less likely than European- and Arab-Americans to endorse this belief. African Americans were also less likely than their European American counterparts to feel that the police meted out justice fairly. It is important to note that, at the time these data were collected, there were instances reported in the news of police brutality against African Americans in two of the communities where we collected data.

These results are consistent with Phinney et al.’s (1994) finding that African American youth were less likely than other ethnic minorities to believe that the rights and opportunities of the American polity applied to them. We hasten to add that this result should not be interpreted as a sign of mistrust towards members of other ethnic groups. Although mistrust and vigilance in inter group relations does come up in research with ethnic minority parents, it is not a prominent theme in parents’ narratives (Hughes et al. 2006). Furthermore, as Hughes and Johnson (2001) note, when ethnic minority parents caution their children about trusting people from the ethnic majority group, these warnings are correlated with perceptions that children have been unfairly treated by those “others”—typically peers—because of their race. Rather than bias from individuals, our dependent measures tap adolescents’ perceptions that the principles and rules of fair and equal treatment implied in the American social contract are implemented in the practices of two institutions (the government and the police). The low mean for the African-American youth suggests that this group is less likely to believe that the government cares and works for the average person or that the police are fair in their dealings with everyone. This is consistent with results from a national survey of 15–25 year olds which found that African-Americans were less likely than Latino or White youth to feel that the government is responsive to the needs of the public (Marcelo et al. 2007). Likewise, the lower confidence in the police expressed by African-American youth in our study is consistent with national data documenting the lower confidence in the police of this group when compared to that of White youth (Johnston et al. 1986).

The fact that reports of discrimination were significantly and inversely related to beliefs in the American promise points to the political implications of these experiences insofar as they undermine adolescents’ general beliefs that the United States is a land of equal opportunity. If youth do not believe that the system is fair, why would they want to be stakeholders in that system? In fact, theorists of political socialization accorded great importance to the younger generation’s diffuse support for the fundamental fairness of the system as a foundation whereby political stability was maintained across generations (Easton 1975; Easton and Dennis 1967). In other words, the theory holds that the stability of our democratic system depends on widely shared beliefs in the younger generation that the system and its guiding principles are fair, a belief that is reinforced in schools and community based settings where a civic ethic is practiced (Flanagan et al. 2007). In contrast, the current study suggests that, when youth encounter intolerance, prejudice, or social exclusion, instances when a civic ethic is not practiced, those experiences may undermine their beliefs in the fundamental fairness of the system. Thus, we infer that youths’ experiences with prejudice have both personal and political implications.

Finally, with respect to civic goals, there were no ethnic group differences in commitments to patriotism (serving the country and helping the community and society). Nor were ethnic awareness or reports of prejudice associated with patriotism. However, as expected, ethnicity (in terms of group membership and awareness) did have an impact on the other two civic goals. As a group, the majority (of European descent) were less likely than each of the minority groups to endorse the goal of advocating for their ethnic group, which is hardly surprising given the lower salience of ethnic awareness for this group. However, they also had the lowest means on building tolerance and improving race relations, significantly lower than their peers from African American backgrounds. Although this group was equally likely as their peers from minority groups to say that tolerance was a value emphasized by their parents, they were less likely than the African-American youth to rate improving race relations as highly as a civic goal. However, we hasten to add that the European-Americans were not different from the Latino- and Arab-Americans on this civic goal. Furthermore, their endorsement of this civic goal (improving race relations) was actually higher than their endorsement of patriotism. The very strong endorsement that African-American youth gave to improving race relations also deserves some discussion. Clearly, improving race relations and ending prejudice in society have strong civic and personal resonance for these youth. Although they did not differ significantly from the Latino- and Arab-Americans on this goal, the mean for this group on this civic goal was higher than any other group mean for any of the civic goals.

Ethnic awareness also was positively related to commitments to improving race relations and especially to the civic goal of advocating for one’s ethnic group. Notably, in contrast to the role of prejudice in family discussions of the personal barriers it poses and to the role of prejudice in undermining youth’s beliefs in the American promise, reports of prejudice were unrelated to any of the three civic commitments. The fact that ethnically aware youth were more motivated to improve race relations and to advocate for their group is consistent with research on immigrant youth which shows that civic engagement often takes the form of assistance and advocacy for one’s ethnic group (Lopez and Marcelo 2008; Stepick et al. 2008).

The civic commitments endorsed by the ethnic minorities in our study suggest that serving the broader society and advocating for one’s ethnic group are not at odds. Indeed, advocating for the rights of one’s ethnic group has historically been the means by which new immigrant groups have carved a space for themselves as Americans. The results also suggest that ethnic awareness may be implicated in the importance that adolescents attach to doing something to improve race relations in society. Unfortunately, such indicators of civic engagement are not typically measured in national surveys and thus we have little information about the role that such goals might play in the civic identity development of younger generations. Given the rapidly changing demographic profile of the U.S. population, we need more studies of how personal experiences intersect with political development for different ethnic groups of young people and we need to define concepts such as civic and political engagement broadly such that they capture the breadth of youth engagement. For example, one recent study revealed that cynicism about electoral politics does not automatically result in a lack of political efficacy. Marcelo et al. (2007) found that, although African American youth were more likely than Latino and White youth to view politics as a way for the powerful to keep power, they also were the most likely of all groups to believe that they could make a difference in solving problems in their community.

The limitations of our study should be noted. Besides the cross-sectional nature of the data, we relied on self-reports from adolescents alone and our results may be inflated by this single reporter bias. Most of our scales were comprised of relatively few items and, in some cases, the alphas for some ethnic groups were less than ideal. Additional items may have improved the internal consistency of those measures. In addition, our study examined only one aspect of ethnic identity, i.e., ethnic awareness. Both awareness and identity change over time and are responsive to experiences in different contexts. Since our data were collected at only one point in time, they miss the dynamic nature of adolescents’ experiences of being both ethnic and American. The Arab Americans in our sample may be the best group to illustrate this point. These youth reside in a community where they are surrounded by reminders of their cultural heritage. As noted in the results, they were more likely than other groups to mention religion as a context where they felt very aware of their ethnic group membership. In the schools they attend, a large segment of the student body celebrates Ramadan, fasting during the school day; many female students wear a hijab; and each year several students have family members doing a Hajj. In short, the density and coherence of ethnic messages from the institutions and fellow members of their local community reinforce their cultural values (Phinney et al. 2000; Wray-Lake et al. 2008). But these data were collected before September 11, 2001. If similar data were collected today we can only surmise that their feelings about being Arab American might be colored by current conditions and reveal a more vexed relationship.

In asking, “What does it mean to be an ‘American’?,” Walzer (1990) notes that, although we have appropriated the adjective, there is no country called America. Our sense of ourselves is not captured by the fact of our union. Rather, we are bound by the commitment to tolerance codified in our Constitution. What makes us Americans is a commitment to respect others who are different from us and with whom we may ardently disagree. With each new generation we renegotiate the principles that established our nation and that define our collective identity. The results of this study suggest that in negotiating their ethnic and American identities and in dealing with prejudice, adolescents from diverse ethnic backgrounds provide unique insights into the principles that bind us as Americans and to the future of our democracy.