Numerous studies indicate that experiences of racial exclusion such as discrimination and prejudice are quite common for urban ethnic minority youth across educational, institutional and peer settings (Fisher et al. 2000; Rosenbloom and Way 2004). Not only is discrimination a common experience for ethnic minority children (e.g., African American and Latinos) by adolescence, children are well aware of racial discrimination and prejudice as explanations for various social and power inequalities (Brown 2008). Children and adolescents associate low-status occupations with ethnic minority status, and associate affluence and greater economic and educational opportunities with ethnic majority status (Hughes and Bigler 2008); thus, status differences between ethnic majority and ethnic minority groups are something that youth are aware of by middle childhood. Moreover, research indicates that experiencing racial or ethnic discrimination and prejudice has a negative impact on a range of developmental outcomes (Sellers et al. 2006; Szalacha et al. 2003). Thus, investigating how low-income urban ethnic minority children and adolescents evaluate racial exclusion and the factors that are related to these types of judgments may contribute to our understanding of how to help these youth avoid such negative outcomes to which they may be particularly more vulnerable than their majority counterparts.

While a recent study with middle-income participants from different ethnic backgrounds found that perceptions and evaluations of interracial exclusion were significantly related to intergroup contact in everyday peer encounters (e.g., Crystal et al. 2008), little to no research has investigated the intergroup attitudes of urban ethnic minority youth regarding their evaluations of exclusion based on race and ethnicity and its connection to intergroup contact. In several recent studies with adolescents, intergroup contact has been defined as self-report information regarding cross-race friendships as well as contact in the school context with members of other ethnic and racial backgrounds (see also Edmonds and Killen 2009; Tropp and Prenovost 2008). Intergroup contact in the form of cross-race friendships aids in perspective-taking of other groups, and increases recognition of what it feels like to be excluded (Killen et al. 2007a; Tropp and Prenovost 2008). Thus, given that intergroup contact has been related to evaluations of exclusion based on ethnic majority and minority status (Crystal et al. 2008), and that this relationship has not been examined in a low-income urban ethnic minority sample, the present study was designed to fill this gap by investigating inner-city minority students’ perceptions regarding evaluations of exclusion and intergroup contact. In addition to measuring participants’ evaluations of exclusion, this study took a multi-measure approach assessing children’s and adolescents’ estimates of the frequency of exclusion, their recognition that individuals use stereotypes to justify exclusion, and their sense of obligation to intervene when witnessing exclusionary peer behavior.

In studies focused on children’ s awareness of discrimination, and status differences between racial ethnic groups, researchers have sought to understand how children and adolescents evaluate exclusion of an ethnic minority student by an ethnic majority student. For example, in a number of studies on evaluations of racial exclusion, the responses of ethnic majority (e.g., White) students were compared with those of ethnic minority students (combining African American, Latino, and Asian American students, Killen et al. 2002) in the US (e.g., Killen et al. 2002). Researchers in Spain (e.g., Enesco et al. 2008), Australia (e.g., Nesdale et al. 2005), and Germany (e.g., Feddes et al. 2009) have also conducted majority/minority comparisons to understand and investigate prejudicial attitudes. In these studies, few differences have been reported for how members of different ethnic minority groups evaluate the exclusion of an ethnic minority member by a White majority peer. These findings have been interpreted as reflecting the way that members of minority groups identify with exclusion by a majority “high status” individual.

Investigating urban ethnic minority youths’ reactions to racial exclusion provides an important and unique portrayal of how younger members of low-income minority groups, who, like their older counterparts, experience low status within the larger society, evaluate high status majority groups’ decisions, actions and judgments toward low status minority peers. Such research has been particularly important in the area of prejudice and exclusion where most studies with adults have focused only on the “White” ethnic majority perspective, with very few studies (until recently) including the target’s perspective (see Swim and Stangor 1998, for an exception). For these reasons, the present study adopted a majority/minority comparison design to investigate how an understudied group, low-income urban African American and Latino students, evaluate interracial peer exclusion, and how such evaluations relate to experiences of intergroup contact (see Tropp and Prenovost 2008).

Recent research on the role of social experience and cross-group friendship in determining intergroup attitudes has drawn from both social psychology and developmental psychology in conceptualizing intergroup contact. Social psychological research on intergroup relations in adulthood suggests important short- and long-term benefits from positive associations with individuals from different racial and ethnic backgrounds (Pettigrew and Tropp 2005; Stephan 1999). According to intergroup contact theory (Allport 1954), an important condition of contact is cross-race friendship, which is essential for reducing racial segregation and prejudice. Most of the available research demonstrating the benefits of intergroup contact, however, has focused on majority, high status groups (such as European-American or Anglo-British), and mostly used adult participants (Brown and Hewstone 2005; Pettigrew and Tropp 2006).

Recently, developmental research outside the US has examined the role of intergroup contact and its relationship to social developmental outcomes. For example, several recent European studies have explored attitudes surrounding intergroup contact among ethnic majority and ethnic minority youth, such as German and Turkish children in Germany, Dutch and Muslim children in the Netherlands, and British and South Asian children in the United Kingdom (e.g., Binder et al. 2009; Feddes et al. 2009; Turner et al. 2007; Verkuyten 2008). Findings from these studies suggested that positive outgroup contact, in the form of cross-race friendships, improved intergroup attitudes among both majority and minority group children. The findings also indicated that intergroup contact was more effective for majority status children than for minority status children, supporting the assumption that intergroup contact may work differently for ethnic majority and minority youth (Tropp and Prenovost 2008).

The present study moves beyond a focus on attitudes alone and examines the extent to which opportunities for intergroup contact are related to evaluations of interracial peer exclusion in a low-income, inner city US ethnic minority sample of children and adolescents. Social Domain Theory (Turiel 2002) has provided the theoretical perspective for much of the recent developmental work on children’s and adolescents’ evaluations of interracial exclusion and has shown how young people’s judgments differ depending on the target of exclusion and the social context (for reviews, see: Killen et al. 2007a, c). As an illustration, Killen et al. (2007b) examined evaluations of exclusion in interracial contexts (e.g., friendship, sleepover, and dating) among majority and minority children attending racially and ethnically heterogeneous US middle-income schools. Findings indicated that minority children were more likely than their majority counterparts to report that racial exclusion occurs more often than non-race based exclusion for reasons such as lack of shared interests, or for the sake of group functioning. According to Killen et al. (2007b), these findings highlight the significance of racial/ethnic background, social experience, and context on individuals’ interpretations and expectations of exclusion.

A recent study examining students’ evaluations of racial exclusion in mixed-ethnicity schools, with varying opportunities for positive intergroup contact, also provides some useful insight for the current investigation. Crystal et al. (2008) found that both majority and minority students with high levels of intergroup contact were more likely to view race-based exclusion as wrong than were their counterparts with low levels of intergroup contact. In addition, students reporting higher levels of intergroup contact also provided lower estimates of the frequency of occurrence of race-based exclusion in mixed-raced settings than their peers with lower levels of such contact (Crystal et al. 2008). This finding suggests that racial exclusion is less likely to occur in contexts in which students have cross-race friendships. Thus, a measure of how often exclusion occurs, from the perspective of the participant, would be important to include in a study of interracial exclusion and intergroup contact.

More recently, researchers (Killen et al. 2010) examined the role of intergroup contact (in terms of a school’s level of ethnic diversity) on ethnic majority (European American) children’s and adolescents’ evaluations and judgments of racial exclusion and endorsement of racial stereotypes to explain interracial discomfort. Results suggested that European American children with little intergroup contact were less likely to view racial exclusion as wrong and more likely to affirm and endorse stereotypes as a valid explanation for racial discomfort, than participants with higher levels of intergroup contact. Whether this finding extends to low-income ethnic minority students is not yet known. Given the well-documented association between prejudice and stereotyping (Bigler and Liben 2006), an assessment of whether intergroup contact is related to the affirmation or rejection of stereotypes to justify exclusion would be an important measure to include in the present investigation.

Finally, relatively little empirical attention has been paid to the role of bystanders in dealing with racial prejudice and discrimination. In one of the few studies on the topic, Aboud and Fenwick (1999) examined the responses of college students toward racially offensive comments made by a fictitious partner in a dyadic problem-solving task. Findings revealed few differences between ethnic minority and ethnic majority participants in terms of how they responded to racial slurs directed towards an out-group member (Aboud and Fenwick 1999). Published work with younger participants has tended to focus on how elementary school children respond to bullying rather than to direct forms of racial exclusion or prejudice (see Aboud and Joong 2008). What has not been done is to ask children and adolescents whether individuals are obligated to respond when racial exclusion occurs. In the present study, we predicted that there would be a positive relationship with age for bystanders’ response to racial prejudice in a sample of urban ethnic minority children and adolescents due to their frequent role as a victim in exclusion contexts.

Taken together, these findings highlight the importance of racial/ethnic background, context, and social experience in terms of the level of intergroup contact on individuals’ interpretations and expectations concerning racial exclusion. Surprisingly, however, there have been few systematic investigations of the effects of intergroup contact on the racial attitudes and judgments about racial exclusion of low-income ethnic minority inner-city children and adolescents in the US. These youth are more likely to live in environments characterized not only by factors such as poverty, unemployment, and limited resources but also by both subtle and obvious prejudice, discrimination, and racism (Garcia Coll et al. 1996; Simons et al. 2002). In addition, for inner-city ethnic minority students, especially African American and Latino students, racially and ethnically diverse schools are becoming the exception rather than the rule as more US public schools become increasingly segregated (Frankenberg and Orfield 2007; Orfield 2001; Orfield and Eaton 1996). Thus, examining how low-income urban ethnic minority students attending predominantly ethnic minority schools evaluate racial exclusion provides an important comparison to existing research on this topic which, until now, has primarily focused on middle-income suburban ethnic minority youth attending integrated schools (Crystal et al. 2008).

Current Study

In the current study, we investigated how low-income urban ethnic minority children and adolescents attending schools where they are the numerical majority and hence with fewer opportunities for positive intergroup contact, judge interracial peer exclusion and the use of stereotypes. Using a multi-measure approach to prejudice development, we examined the impact of intergroup contact on how minority children and adolescents evaluate exclusion, affirm or reject stereotypes to justify exclusion, and assess a same-aged peer bystander’s responsibility to respond to race-based peer exclusion.

A number of research predictions were generated in the present study. We expected that, in general, minority participants would judge interracial exclusion to be wrong. Prior research, has shown this to be the case with a sample of suburban ethnic minority students (Killen et al. 2002). Due to the robust findings regarding intergroup contact reported in previous investigations (Crystal et al. 2008; Pettigrew and Tropp 2006; Tropp and Prenovost 2008), we expected that participants with higher levels of intergroup contact would be more likely to view race-based exclusion as wrong, provide higher estimates of the frequency of exclusion, show greater awareness of majority students’ use of stereotypes, and be more likely to know how to respond to situations involving racial exclusion. We also predicted that, independent of the level of intergroup contact, with increasing age minority children would be more likely to indicate that majority group members would use stereotypes to justify racial exclusion, given minority adolescent’s awareness of racial discrimination and prejudice (Brown 2008; Fisher et al. 2000; Szalacha et al. 2003). This prediction was consistent with available research and theory on age-related advances in children’s social-cognitive development concerning stereotyped knowledge (Brown 2008; Brown and Bigler 2005; McKown 2004; McKown and Weinstein 2003). In view of the dearth of literature on this topic, we did not make explicit predictions between intergroup contact and minority students’ evaluations of bystanders’ responses. However, we did predict a positive relationship between age and bystander intervention, such that older children would be more likely than younger children to suggest that a bystander should actively intervene.

Few prior studies have found significant gender effects on racial attitudes. In addition, oftentimes, gender has not been included as a variable due to the focus on ethnicity and age (see Crystal et al. 2008). Yet, research on evaluations of exclusion in solely ethnic majority samples has reported gender differences (Killen and Stangor 2001). Therefore, we expected that females would view racial exclusion as more wrong than would males based on previous research.

Method

Participants

Participants were one hundred and twenty-nine children in 4th, 7th, and 10th grade attending four urban predominately minority public schools in the New York City metropolitan area. Fifty-nine 4th grade students (30 girls and 29 boys; M = 9.22 years, SD = .46), 40 7th grade students (26 boys and 14 boys; M = 12.35 years. SD = .48), and 30 10th grade students (22 girls and 8 boys; M = 15.47 years, SD = .82). The overall ethnic composition of the schools was 55% African American, 31% Latino, 8% European American, and 6% Asian. The ethnic breakdown of students taking part in the study was 49% African American, 42% Latino, and 9% Biracial (African American/European American or African American/Latino). The racial ethnic composition of the sample reflected the racial ethnic composition of the schools taking part in the study, with the exception that European-American students were excluded from the study due to the focus on ethnic minority youth. Overall, 70% of the children attending these schools were eligible for free or reduced-cost lunches. Hence, the majority of African American and Latino children taking part in the study were either from poor or low-income working families.

Procedure and Measures

Written parental informed consent (mean response rate of approximately 65%) and child assent was obtained for all participants taking part in the study. Trained ethnic minority interviewers and coders, matching the participants’ race/ethnicity for the majority of the children, administered the semi-structured Social Reasoning about Exclusion Interview and coded the data. After the interview, participants were asked to complete the Developmental Intergroup Contact Survey.

Social Reasoning about Exclusion

The Social Reasoning about Exclusion interview consisted of three short scenarios each depicting a context in which interracial exclusion might occur (Killen et al. 2007c). Extensive pilot testing was conducted on the interview scenarios in order to ensure that participants were familiar with the situations, and that the language was developmentally appropriate. The scenarios were presented in a pre-established order (see Crystal et al. 2008), with the last two contexts representing more intimate situations than those employed in earlier research (e.g., Killen et al. 2002; Killen and Stangor 2001). The scenarios were: excluding a cross-race peer from lunch at school (where a majority child does not want to invite a minority child to have lunch with him/her and another majority friend), excluding a cross-race peer from a sleepover party (where a majority child does not want to invite a minority child to a sleepover), and excluding a cross-race peer from a high school dance (where a majority student does not want to invite a minority student from another school to a school dance as a date). In order to capture the historical dimension of racial exclusion in the United States, all three scenarios involved a European American child excluding an African American child. Brown (2008) reported that in a racially and ethnically diverse sample, US children often gave examples of discrimination directed towards African American children, when presented with open-ended questions about discrimination. In addition, she noted that these same children also gave examples of racial discrimination from a historical perspective, providing further support for our use of situations involving a European-American child excluding an African American child.

After the presentation of each of the three scenarios participants responded to a series of assessments (1) Wrongfulness of race-based exclusion (How good or bad is it to exclude someone on the basis of their race?); (2) Stereotype assessment (What is it about race that makes people uncomfortable?); (3) Estimations of the frequency of race-based exclusion among peers (How often do you think children your age might not invite someone to lunch because they are a different race?); and (4) Bystander Response (What do you think a bystander child should do about race-based exclusion?).

Responses to the wrongfulness ratings ranged from 1 (“very, very good”) to 8 (“very, very bad”). Responses to the assessment about stereotypes were coded as (a) Affirming stereotypes (affirming or endorsing stereotypes); (b) Stereotype recognition (recognizing people use stereotypes); and (c) Social contexts of stereotypes (awareness of social factors and historical circumstances contributing to the manifestation of stereotypes); and (d) Don’t know or no response. Interrater agreement calculated as Cohen’s kappa coefficient = .90. Responses to frequency estimations ranged from 1 (“never”) to 5 (“always). Participants’ responses to the question of how a bystander should deal with race-based exclusion were coded as: (a) Focus on the excluder (e.g., “he should tell them that just because kids are different races doesn’t mean they have different interests”); (b) Focus on the excluded (e.g., “she should tell the Black girl that she will be her friend even if the other girl won’t”); (c) Focus on both (e.g., “I would talk to both of them and see if I could bring them together”); and (d) Not know what to do (e.g., “I wouldn’t know what to do”). Interrater agreement by two independent raters calculated as Cohen’s kappa coefficient = .99.

Developmental Intergroup Contact

At the end of the interview, participants completed the Developmental Intergroup Contact Survey (Crystal et al. 2008) originally adapted from a 10-item Diversity Attitudes Questionnaire (DAQ) (Kurlaender and Yun 2001). Following Crystal et al. (2008) the 10 DAQ items were subjected to a principal axis factor analysis with varimax rotation (Kaiser normalization), which yielded a primary factor, explaining 34% of the variance and consisting of six items. Those six items comprised our Intergroup Contact Scale and included: (1) How many students in your school are from racial or ethnic groups different from you own? (2) How often do you work on school projects and/or study with students from other racial/ethnic groups? (3) At school, how many friends do you have who are from a different racial or ethnic group than you? (4) Outside of school, how many friends do you have who are from a different racial or ethnic group than you? (5) In the neighborhood where you live, do you have neighbors from other racial or ethnic groups? and (6) How many of your friends from your neighborhood are from a different racial or ethnic group than you? Responses to these items ranged from one (“none”) to four (“many”), and were summed and then averaged to form the Intergroup Contact Scale, with a Cronbach’s α of .72. In order to include the Intergroup Contact Scale in more complex multivariate analyses, the scale was dichotomously split along the mean into groups of “low” and “high” intergroup contact.

Results

Univariate analysis of variance (ANOVA) was used to provide descriptive analyses of the Intergroup Contact Scale. Repeated measure ANOVAs were used to assess main, interaction and contextual effects on participants’ judgments of the wrongfulness and frequency estimations of race-based exclusion. Follow-up tests included univariate ANOVAs to determine between-subject effects and interactions. In those cases where sphericity was not met, corrections were made using the Huynh–Feldt method. To determine whether there were significant differences between the responses of African Americans and Latinos, repeated measures ANOVAs comparing the two groups on all major dependent variables of interest for each story were conducted. Since results of the ANOVAs revealed few significant differences between the two groups, African American and Latino students were combined for the purposes of further analyses in the present paper. In the discussion, we consider the rationale, complexities, and limitations of such an approach.

Grade and Gender as Predictors of Intergroup Contact

A univariate ANOVA was performed on the Intergroup Contact Scale with grade and gender serving as the independent factors. Means of intergroup contact, broken down by gender and grade are displayed in Table 1. Significant main effects were found for grade F(2, 123) = 5.06, p < .05, η 2 p  = .08. Tenth graders (M = .27, SD = .08) reported the lowest levels of intergroup contact followed by 4th graders (M = .50, SD = .07) and 7th graders (M = .62, SD = 0.08). Post-hoc tests indicated that 10th graders reported significantly lower levels of intergroup contact than 7th graders, p < .05.

Table 1 Level of intergroup contact by grade and gender
Full size table

Wrongfulness Ratings of Race-Based Exclusion

A 3 (grade: 4th, 7th, 10th) × 2 (gender: female, male) × 3 (scenario: lunch, sleepover, dance) ANOVA with repeated measures on the last factor was conducted on students’ wrongfulness ratings for race-based exclusion. There were no significant effects for grade, gender or by type scenario. As predicted, the majority of participants rated race-based exclusion as “very wrong” (M = 7.21, SD = 1.00).

Intergroup Contact as a Predictor for Wrongfulness Ratings of Race-Based Exclusion

To examine the influence of intergroup contact on participants’ ratings of the wrongfulness of race-based exclusion a 3 (grade: 4th, 7th, 10th) × 2 (gender: female, male) × 2 (intergroup contact: low, high) × 3 (scenario: lunch, sleepover, dance) ANOVA with repeated measures on the last factor was conducted on evaluations of the wrongfulness of race-based exclusion. Results of the repeated measures ANOVA indicated a significant main effect for intergroup contact, F(1,114) = 8.19, p < .01, η 2 p  = .07. Specifically, as hypothesized, students with high intergroup contact (M = 7.48, SD = 1.07) were significantly more likely to rate race-based exclusion as wrong than were students with low intergroup contact (M = 7.01, SD = 1.11).

In addition, a significant three-way interaction for intergroup contact, gender, and scenario was found, F(1, 126) = 5.08, p < .05, η 2 p  = .04. Follow-up analyses showed that, in the lunch scenario, in line with predictions, male students with high levels of intergroup contact (M = 7.42, SD = .77) were more likely to view race-based exclusion as wrong than were males with low levels of intergroup contact (M = 6.65, SD = 1.14). No significant differences in ratings of the wrongfulness of race-based exclusion were found between females with high levels of intergroup contact (M = 7.29, SD = .96) and their low contact counterparts (M = 7.28, SD = .95). Thus, intergroup contact was significantly related to evaluations of race-based exclusion for males but not for females in the lunch scenario. No significant effects were found in the other two scenarios.

Grade and Gender as Predictors of Estimates of the Frequency of Race-Based Exclusion

To analyze participants’ estimations of the frequency of race-based exclusion across scenarios, we conducted a 3 (grade: 4th, 7th, 10th) × 2 (gender: female, male) × 2 (intergroup contact: low, high) × 3 (scenario: lunch, sleepover, dance) ANOVA with repeated measures on the last factor. The results indicated a significant within-subjects interaction among context, gender, and grade F(2,115) = 3.85, p < .05, ή2 = .06. Follow-up analyses revealed two significant findings for excluding a cross-race peer, specifically, from the lunch scenario. First, female 10th graders (M = 3.14, SD = 1.25) reported that race-based exclusion occurred more often than did their female 4th (M = 2.0, SD = 1.17) and female 7th grade (M = 2.04, SD = 1.15) counterparts, F(2,77) = 7.028, p < .005, η 2 p  = .16. Second, male 4th graders (M = 2.39, SD = 1.26) provided higher estimates of the frequency of race-based exclusion than did their male 7th grade (M = 1.50, SD = .52) counterparts, F(2,49) = 3.9, p < .05, η 2 p  = .14. There were no significant findings for intergroup contact.

To summarize, the oldest ethnic minority females were more likely than their younger counterparts to provide higher estimates of the frequency of race-based exclusion. In contrast, the youngest ethnic minority males provided higher estimates of the frequency of racial expression than older (i.e., 7th grade) ethnic minority males.

Intergroup Contact as a predictor of Stereotypic Explanations for Race-Based Exclusion

To examine participants’ responses to the question, “What is it about race that makes people uncomfortable?” (i.e., use of stereotypic explanations), we evaluated participants’ explanations of discomfort due to race across 3 scenarios. A 3 (grade: 4th, 7th, 10th) × 2 (gender: female, male) × 2 (intergroup contact: low, high) × 3 (scenario: lunch, sleepover, dance) ANOVA with repeated measures on the last factor was conducted for each of three response categories: (1) stereotype affirmation; (2) stereotype recognition and awareness; and (3) social context of stereotypes.

Due to the fact that less than 10% of the sample affirmed stereotypes to explain racial discomfort, no analyses were conducted for this category. Significant differences were found for participants’ recognition and awareness of stereotypes as a reason to explain majority group members’ discomfort, however. Specifically, across contexts, there was a significant between-subjects interaction effect for intergroup contact and gender, F(2,110) = 3.12, p < .05, η 2 p  = .05. Follow- up analyses indicated that, in accord with hypotheses, males with high intergroup contact (M = .18, SD = .63) were more likely than males with low intergroup contact (M = .11, SD = .33) to recognize that majority individuals would use stereotypes to justify exclusion. There were no significant differences between females with high (M = .34, SD = .50) and low (M = .31, SD = .44) levels of intergroup contact in terms of their expectations that majority group members would use racial stereotypes as an explanation for racial exclusion. Overall, minority females (M = .36) were more likely than minority males (M = .14) to expect majority individuals to use stereotypes to explain exclusion, F (1, 127) = 8.03, p < .005, η 2p  = .059.

For participants who reported that they did not know why race would make people uncomfortable, a significant main effect was found for grade F(1,55) = 3.22, p < .05, η 2 p  = .105. Post-hoc comparisons showed that 4th (M = .58, SD = .50) and 7th (M = .68, SD = .48) graders were more likely to report that they did not know why race would make people uncomfortable than were 10th (M = .18, SD = .39) graders.

To summarize, consonant with predictions, males with high levels of intergroup contact were more likely than males with low levels of such contact to be aware that majority children evoke stereotypes to explain racial exclusion. In terms of gender differences, ethnic minority females were more likely than ethnic minority males to suggest that majority individuals would use stereotypes to justify racial exclusion. The oldest participants were more likely than the younger ones to know why race would make some people uncomfortable.

Intergroup Contact as a Predictor of Bystander Responses

To examine participants’ responses to the bystander assessment, a 3 (grade: 4th, 7th, 10th) × 2 (gender: female, male) × 2 (intergroup contact: low, high) × 3 (scenario: lunch, sleepover, dance) ANOVA with repeated measures on the last factor was conducted. A significant within-subjects interaction effect was found among context, gender, and grade for participants who suggested focusing on the excluder when responding as a bystander, F(4,114) = 2.97, p < .05, η 2 p  = .05. Follow-up analyses revealed that, in the lunch scenario, male 7th graders (M = .78, SD = .43) were significantly more likely than male 4th graders (M = .36, SD = .49) to indicate that they would focus on the excluder, F(2,49) = 4.01, p < .05, η 2 p  = .15.

In regard to participants who “did not know” what to do about race-based exclusion, a significant between-subjects interaction was found for intergroup contact and how to respond to racial exclusion, F(4,114) = 2.55, p < .05, η 2 p  = .04. Follow up analyses revealed that, students with low intergroup contact (M = .04, SD = .18) were more likely to indicate that they did not know how to respond to racial exclusion than participants with high intergroup contact (M = .02, SD = .15). Additionally, a significant between-subjects interaction was found for grade and gender, F(4,114) = 4.78, p < .001, η 2 p  = .08. Further analyses indicated that male 4th graders (M = .05, SD = .17) were more likely to say that they did not know what to do about race-based exclusion than female 4th graders (M = .03, SD = .14). In addition, male 7th graders (M = .09, SD = .23) were more likely to indicate that they did not know what to do about race-based exclusion than female 7th graders (M = .03, SD = .20). Interestingly, all of the participants in the 10th grade indicated that they knew what to do in situations involving race-based exclusion.

To summarize, students reporting low intergroup contact were less likely to know how to respond to racial exclusion than were students reporting high intergroup contact. Additionally, male 7th graders were more likely than their 4th grade counterparts to focus on the excluder in the lunch scenario. Finally, male 4th graders and 7th graders were less likely to know what to do about racial exclusion than were their female peers.

Discussion

Overwhelmingly, the vast majority of inner city, ethnic minority children and adolescents evaluated racial exclusion as “very wrong”. Central to the hypotheses, intergroup contact, in concert with grade, gender, and context was found to be a significant predictor of responses related to situations of exclusion. These results demonstrate that benefits of intergroup associations on various measures of prejudice and reactions to situations of interracial exclusion can be generalized to low-income urban African American and Latino youth, participants not typically included in studies on intergroup contact (see Tropp and Prenovost 2008).

Thus, the current research, with its focus on a unique sample—low-income, urban ethnic minority 4th, 7th, and 10th graders—and its use of a variety of self-report measures, including, most importantly, evaluations of racial exclusion, contributes to the extant developmental literature on intergroup contact and prejudice.

In the present investigation, in accord with prior research (e.g., Crystal et al. 2008; Killen et al. 2007b), ethnic minority students reporting high levels of intergroup contact were more likely to rate race-based exclusion as wrong than were students with low levels of intergroup contact. However, this general finding was further qualified by interactions with gender and context. More specifically, male students with high levels of intergroup contact were more likely to view race-based exclusion as wrong than were males with low levels of intergroup contact. This was particularly evident in the lunch scenario. No significant differences in ratings of wrongfulness were found between females with low or high levels of intergroup contact. Intergroup contact has been shown to be more effective for ethnic majority groups than ethnic minority groups (Tropp 2006; Tropp and Pettigrew 2005; Tropp and Prenovost 2008); that there were gender differences within a low-income ethnic minority sample was surprising and calls for further research to understand this finding.

In addition, gender differences also emerged with regard to participants’ awareness of the use of stereotypes to justify exclusion. Minority males with high intergroup contact were more likely than minority males with low intergroup contact to recognize that majority individuals would use stereotypes to justify exclusion. Again, intergroup contact was not significantly related to ethnic minority female responses. Intergroup contact appears to be a more salient factor for ethnic minority males when evaluating exclusion; why this is so remains to be more fully investigated. When compared to males, females were more likely to recognize that ethnic majority individuals use stereotypes to justify exclusion. Age-related differences also emerged in terms of bystander response. Seventh-grade males were more likely than 4th grade males to view that the bystander should intervene. For the group of children who did not know what the bystander should do, there were gender differences. The youngest two groups of males were more likely than their female counterparts to suggest that they did not know what to do about peer racial exclusion. In part, this finding of differences between males and females supports previous work suggesting that girls are more sensitive to exclusion than boys due to their own experiences of being excluded from gendered specific activities such as sports (e.g., Killen et al. 2002; Killen and Stangor 2001). Moreover, the gender findings call attention to the perspective of low-income ethnic minority male participants who were less inclined to view exclusion as wrong when they lacked cross-race friends. The opportunity for cross-race friendships appears to be particularly important for ethnic minority male adolescents, and this issue warrants further examination.

We also found that students with low intergroup contact were more likely to indicate that they would not know what to do. Hence, they would not intervene, as a bystander of interracial exclusion than were participants with high intergroup contact. The fact that students with high contact were more likely to take a pro-active stance toward racial exclusion compared to their low contact counterparts adds to the list of social-cognitive benefits of intergroup association (Killen et al. 2007a). Also, recent work has suggested that bystander interventions may provide one useful way for young people to address peer exclusion and discrimination in school settings (Aboud and Joong 2008). Studies exploring the mechanisms by which high levels of intergroup contact may lead minority (as well as majority) children and adolescents to develop an “activist attitude” toward addressing inter-racial peer exclusion are warranted.

Our prediction that with increasing age participants would be more likely to indicate that majority group members would use stereotypes to justify racial exclusion was not supported.

However, grade-related findings emerged with regard to children’s and adolescents’ knowing why race would make people uncomfortable. Specifically, younger children were more likely than older children to report that they did not know what it was about race that would make majority group members uncomfortable. The finding that the oldest participants were able to provide a response as to why race would lead to interracial discomfort is in accord with prior research showing that older children experience more racism and discrimination than younger peers (Lerner 2004; Szalacha et al. 2003).

There were several limitations in the present study that need to be addressed. First, due to the fact that we found significant effects of both age and gender, a larger sample would have been helpful for testing interactions between these factors. With multiple contributing variables, we were unable to fully test all possible predictions regarding the interactions for gender by age to the extent that we would have liked. Nevertheless, a number of significant gender by age interactions emerged that could be fruitfully explored in a follow-up study with a larger sample, which have not been documented with middle-income ethnic minority samples. Moreover, we matched the gender of the participant with the gender of the characters in the interview protocols. Given that gender was a significant predictor of students’ responses, it would be of interest to vary the gender of the characters in the scenarios. This method would allow us to assess whether male and female participants differentially evaluate both male and female peer dyadic exclusion encounters.

Second, this study focused on a low-income inner-city ethnic minority sample attending ethnic minority schools with few majority students. In the future, it would be helpful to examine the research questions in this study with a sample of low-income youth attending heterogeneous multi-ethnic educational settings. This would permit parceling out school environment and socioeconomic status in relation to intergroup contact and judgments of exclusion.

Third, to streamline this study, only one pairing of the scenario characters’ ethnic background was included (White/Black). This design was based on previous work using similar situations with middle-income African American, Asian American and Latino students (see Killen et al. 2002). In that study, participants across racial ethnic groups appeared to identify with an excluded African American child, as reflected in the lack of significant minority intergroup differences in evaluations of interracial exclusion. However, given that African Americans and Latinos have distinct cultural backgrounds and unique historical experiences, there may be other contexts in which differences in evaluations of exclusion are quite salient (Fisher et al. 1998; Pahl and Way 2006). Future research, therefore, would benefit by creating scenarios with protagonists whose race and ethnicity were presented in more varied pairings.

Finally, the current study revealed a number of interesting age-differences in terms of intergroup contact, estimates of the frequency of exclusion, awareness of stereotypes, and bystander response. However, the cross-sectional nature of the study prevents any definitive conclusions regarding developmental change or causation. Future work employing a longitudinal design would allow us to better understand age-related changes in intergroup contact and urban minority students’ judgments of racial exclusion.

Notwithstanding these limitations, our study provides an important examination of the relationship between intergroup contact and evaluations of race-based exclusion in a sample of US low-income urban ethnic minority youth. In effect, this investigation answers questions as to the applicability of the intergroup contact hypothesis (Allport 1954) to an understudied population. The findings from the present study contribute to the literature on intergroup contact and children’s understanding of discrimination. They also illustrate the importance of positive intergroup contact in predominately ethnic minority schools where there are limited opportunities for cross-race friendships with majority students. Our results suggest that research considering opportunities for intergroup contact, rather than focusing exclusively on attitudes, can reveal important benefits for minority children, concerning their broader views about interracial peer exclusion. In addition to age and limited effects of context, the present study also revealed interesting gender differences not previously reported in the developmental work on intergroup contact and students’ evaluations’ of interracial exclusion. A promising and hopeful finding from this investigation was that low-income ethnic minority children and adolescents did not resort to stereotypic explanations for racial discomfort. Instead, these youth focused on their recognition and awareness of stereotypes, and the wrongfulness of racial exclusion. Future research needs to examine the types of personal experiences (such as interactions with authority figures), and societal expectations that may lead to differences in how low-income urban ethnic minority children and adolescents think about interracial interactions and contact across a variety of settings. This line of research will enable educators to design effective intervention programs to reduce prejudice, and to promote positive intergroup environments for all youth.