هویت قومی و عزت نفس: بررسی نقش بافت اجتماعی

This study explored ethnic identity and self-esteem among 1062 Mexican-origin adolescents who were attending one of three schools, which varied in their ethnic composition (i.e., predominately Latino, predominately non-Latino, and balanced Latino/non-Latino). Significant relationships emerged between ethnic identity and self-esteem among adolescents in all school settings. Furthermore, controlling for generation and maternal education, adolescents attending the predominately non-Latino school reported significantly higher levels of ethnic identity than adolescents in the other schools. Consistent with ecological theory, these findings challenge researchers to design future studies in ways such that multiple layers of context and their influence on development can be examined.

In an increasingly ethnically diverse society, it is critical to gain an understanding of the role that individuals’ ethnic identities play in their lives. Previous research suggests that ethnic identity can have an important influence, as it has been related to outcome variables such as academic achievement (Arellano & Padilla, 1996), abilities to cope with racism and discrimination (Chavira & Phinney, 1991; Phinney & Chavira, 1995), and psychological well-being (for a review, see Umaña-Taylor, Diversi, & Fine, 2002). Although one's ethnic identity is one of many components that comprise an individual's global identity, it is interesting to note that this single component is consistently positively related to individuals’ self-esteem. To gain a clearer theoretical understanding of this relationship, scholars have drawn on social identity theory. According to social identity theory (Tajfel, 1981), individuals’ self-concepts are derived from knowing that they are members of particular social groups. Furthermore, if individuals evaluate the ethnic group they belong to favorably, their self-esteem may be enhanced via membership in that group (Lorenzo-Hernandez & Ouellette, 1998). In line with these ideas, researchers have consistently found a positive relationship between ethnic identity and self-esteem among ethnic minority adolescents (e.g., Asian, African American, Latino, Native American; Carlson, Uppal, & Prosser, 2000; Martinez & Dukes, 1997; Phillips Smith, Walker, Fields, Brookins, & Seay, 1999; Phinney, 1992; Phinney & Alipuria, 1990; Phinney, Cantu, & Kurtz, 1997; Phinney & Chavira, 1992; Phinney, Chavira, & Tate, 1993; Phinney, Dupont, Espinosa, Revill, & Sanders, 1994). Specifically among Latino adolescents, this relationship has been consistently positive (see Umaña-Taylor & Fine, in press, for a review) and effect sizes (i.e., r2) have ranged from 0.02 to 0.07 ( Carlson et al., 2000; Phinney, Dupont, Espinosa, Revill, & Sanders (1994) and Phinney, Cantu, & Kurtz (1997); Roberts et al., 1999). Thus, although the relationship has been consistent, the demonstrated effect has been small to moderate ( Cohen & Cohen, 1983). When studying ethnic majority adolescents (i.e., White), however, researchers have only found a significant positive relationship between these two constructs when White adolescents are in a minority or ethnically diverse context (i.e., Whites make up a small percentage of the population in their schools; e.g., Phinney, 1992; Phinney, Dupont, Espinosa, Revill, & Sanders (1994) and Phinney, Cantu, & Kurtz (1997); Roberts et al., 1999). As a result, it has been suggested that the relationship between ethnic identity and self-esteem holds only for adolescents for whom ethnicity is salient ( Phinney, 1991). In addition, scholars suggest that ethnicity and, in turn, ethnic identity is more salient for ethnic minority adolescents than for adolescents who are members of the ethnic majority (Martinez & Dukes, 1997; Phinney, 1992; Phinney & Alipuria, 1990). In fact, researchers who have compared the ethnic identity of White adolescents and ethnic minority adolescents have found support for this idea, as White adolescents consistently score significantly lower on measures of ethnic identity than their ethnic minority counterparts (e.g., Branch, Tayal, & Triplett, 2000; Phinney & Alipuria, 1990; Phinney et al., 1994). Researchers suggest that perhaps the lower scores are an indication of the lower salience that ethnic identity has for ethnic majority group members. Although significant, these differences have been moderate; for example, mean item differences between White adolescents and ethnic minority adolescents on ethnic identity have ranged from 0.29 to 0.52, for items rated on a 4-point Likert scale (Phinney et al., 1994). Nevertheless, they consistently demonstrate a possible lower salience of ethnicity. In terms of examining the social context and its influence on ethnicity, it is important to consider that individuals’ lives are embedded in multiple contexts and, as such, these contexts may work together to influence individuals’ experiences (Bronfenbrenner, 1989). As described above, previous studies on ethnic identity in which social context has been considered have explored the influence of divergent macro and micro contexts for White adolescents. That is, White adolescents have been studied in situations in which they are a numerical minority in their school context (i.e., micro) but a numerical majority in society (i.e., macro). However, we have not explored Latino adolescents in divergent contexts (e.g., majority in a school context and minority in societal context). It is possible that being a numerical minority in a broader (i.e., macro) context may overshadow the effects of being a numerical majority in a micro context. For example, adolescents who live in ethnic enclaves may continue to feel like a ‘minority’ because experiences with discrimination occur when they travel outside of their neighborhoods and, furthermore, discriminatory housing practices limit their ability to live in more integrated communities. However, it is also possible that the effects of being a numerical majority in a micro context can shelter individuals to a point where they do not experience being a ‘minority’ because all of their business is conducted in their neighborhoods, they have access to Spanish television, and they only expose themselves to media that portrays them as a majority. To date, our knowledge regarding Latino adolescents’ ethnic identity and self-esteem stems from studies in which they have been numerical minorities in their schools and in society. Furthermore, research in which the differential influence of minority/majority context has been examined has been conducted primarily with White adolescents. Despite recommendations indicating that it is critical to understand the influence of broader environmental contexts (e.g., schools, neighborhoods, societies) on adolescents’ developmental processes and outcomes (Garcia Coll et al., 1996), we know relatively little about the influence of social context on ethnic minority adolescents’ ethnic identity and self-esteem. As such, the current study explored whether ethnic identity, and its relationship with self-esteem, varied among three groups of Mexican-origin adolescents who were attending schools that varied in ethnic composition. Specifically, two hypotheses were examined: (a) The correlation between ethnic identity and self-esteem will be significant and positive for adolescents attending a predominantly non-Latino high school and a balanced Latino/non-Latino high school, while there will be no significant relationship between ethnic identity and self-esteem for adolescents attending a predominantly Latino high school; and (b) Adolescents attending a predominantly non-Latino high school will score significantly higher on ethnic identity than adolescents attending a predominantly Latino high school or a balanced Latino/non-Latino high school.

To examine the first hypothesis, correlations among ethnic identity and self-esteem were examined among adolescents in each of the three schools. A significant relationship emerged between ethnic identity and self-esteem for adolescents attending the predominately non-Latino school (r=0.15, r2=0.02, p<0.05), the balanced Latino/non-Latino school (r=0.19, r2=0.04, p<0.05), and the predominately Latino school (r=0.15, r2=0.02, p<0.001). Although effect sizes were small, findings suggest that the correlational relationship between ethnic identity and self-esteem may not be context specific among Mexican-origin youth. This is contrary to previous findings with White adolescents in which the relationship between ethnic identity and self-esteem emerges only in contexts where Whites are not a numerical majority. It is important to note that the effect sizes found in the current study were small, but within the range of those found in past studies (described above). As such, these results should be interpreted with caution. Prior to examining the second hypothesis, four one-way analyses of variance were conducted to explore whether adolescents from each of the three schools differed with respect to demographic variables (i.e., age, gender, generational status, and maternal education). Based on their school categorization, adolescents did not differ significantly with regard to age, F(2,1056)=1.35, or gender representation, F(2,1058)=0.09. However, significant differences emerged with regard to generational status, F(2,1048)=5.53, p<0.01, η2=0.01, and maternal level of education, F(2,981)=53.15, p<0.001, η2=0.10. In terms of generational status, Tukey's post hoc tests indicated that students attending the predominantly non-Latino school were more likely to have been born in the US than students attending the balanced Latino/non-Latino school (see Table 1 for means). With regard to maternal education, students attending the predominantly non-Latino school (i.e., 15% Latino) reported significantly higher levels of maternal education than students attending either of the other two schools. As a result, the second hypothesis was examined controlling for generational status and maternal education. Table 1. Means, standard deviations and percentage representation for demographic variables by school setting Predominately non-Latino (15%) Balanced Latino/non-Latino (45%) Predominately Latino (96%) M s.d. M s.d. M s.d. Age 15.75 1.25 15.70 1.26 15.59 1.34 (195) (137) (727) Maternal educationa 4.07 1.76 3.00 1.52 2.74 1.51 (189) (128) (667) % % % Female 48.5 50.4 49.5 (196) (136) (729) Born in the US 79.1 62 70.2 (196) (136) (719) a Note: 2, completed junior high school; 3, some high school; 4, completed high school; 5, some college. Table options A one-way analysis of covariance (ANCOVA) was used to examine whether adolescents’ ethnic identity scores varied by school context, while controlling for generational status and maternal education. Prior to examining the results of the ANCOVA, the assumption of homogeneity of variance was examined using Levene's Test of Equality of Error Variances. The variance of the dependent variable did not differ for adolescents in the three school settings, F(2,804)=1.12, ns. ANCOVA results indicated a significant main effect for school, F(2,802)=8.30, p<0.001, η2=0.02. Furthermore, planned simple contrasts indicated that adolescents attending the predominately non-Latino school reported significantly higher levels of ethnic identity (M=41.68; ) than adolescents attending the predominately Latino school (M=39.19; ; M difference=2.49, p<0.001) and adolescents attending the balanced Latino/non-Latino school (M=39.71; ; M difference=1.97, p<0.05). Similar to previous work with White adolescents, this finding could be interpreted to indicate that ethnic identity is more salient for adolescents who are in a minority context (i.e., 15% Latino) than their counterparts who are attending a school where their ethnic group represents the largest ethnic population in the school (i.e., 45% Latino) or is clearly in the majority (i.e., 96% Latino). Although significant, these findings should be interpreted cautiously due to small effect sizes. For example, although significant differences emerged, the largest mean difference was 2.49; furthermore, only 2% of the variance in ethnic identity was explained by school. Clearly, there are additional variables to consider in understanding the factors that influence adolescents’ ethnic identity. Nevertheless, these findings contribute to our understanding by providing information on the influence of school ethnic composition. In addition, it is important to note that adolescents in the three schools differed significantly from one another on an index of SES (i.e., maternal education). This is important to consider because it may be difficult to assess school ethnic composition as a variable without controlling for SES, as the two may be highly interrelated. In other words, the ethnic composition of a school may be largely dependent on the SES of families living in the area. In the current study, SES was introduced as a control variable in order to account for this possible confound; ideally, future researchers should explore schools in which ethnic composition varies but SES does not. It is possible that this could be achieved by targeting schools that serve families living in ethnic enclaves, which are located in suburbs of large metropolitan areas. Although moderate, these findings suggest that ethnic identity appears to be more salient for Mexican-origin adolescents when they are in a minority context than in a majority context, as adolescents attending the school with the largest Latino population reported the lowest ethnic identity scores. These findings are in line with previous studies conducted among White adolescents, and also resonate with previous work among Cuban adults, which found that contact with outside groups was related to heightened ethnic awareness (Portes, 1984). Contrary to previous work with White adolescents, however, the relationship between ethnic identity and self-esteem remained significant for Mexican-origin adolescents, regardless of whether they were a numerical minority or a majority in their schools. This finding suggests that regardless of the ethnic composition of school, ethnic identity remained somewhat salient for Mexican-origin adolescents. Perhaps this differential finding can be explained by the fact that ethnic minority adolescents (Mexican-origin adolescents, in this case) are still a minority in the broader context in which they live (i.e., the US). Thus, while they may not be in the “minority” when they are attending school, they will still be a “minority” when they watch TV or come across magazine advertisements, for example. It is possible that because of these macro context experiences, which White adolescents do not have, their ethnic identity will be salient to them. It is important to note, however, that the strength of the correlation between the two variables was the same for adolescents attending the predominately non-Latino school and the predominately Latino school (i.e., r2=0.02). This finding suggests that perhaps the strength of the relationship between ethnic identity and self-esteem may not change based on social context, although the salience that ethnic identity has for an individual may be more context specific. In other words, while adolescents attending a predominately Latino school may score significantly lower on ethnic identity than adolescents attending a predominately non-Latino school, the relationship between ethnic identity and self-esteem may not vary across the school contexts. It should also be noted that the findings indicate that although ethnicity is salient for all groups of Mexican-origin adolescents in this study (as evidenced by the significant correlation with self-esteem), ethnicity may not be as salient for adolescents who are an ethnic majority in their school contexts, compared to their counterparts who are attending schools where their ethnic group is a distinctive minority. Thus, it appears that both the micro and macro contexts have an effect on salience of ethnicity for Mexican-origin adolescents. As such, these findings are line with ecological theory ( Bronfenbrenner, 1989), which posits that individuals’ lives are embedded within multiple settings and that these settings interact with each other as well as with the individual to influence developmental outcomes. Although this study provides insight into the role of social context among Mexican-origin adolescents’ ethnic identity, there are a number of limitations to consider. First, the results demonstrated low effect sizes. As such, findings should be interpreted with caution. Second, adolescents attending each of the three schools differed significantly with regard to generational status and maternal education; it is possible that the findings may be a result of a third, unknown, variable, which is associated with either generational status or maternal education. Future studies should attempt to recruit schools that have similar demographic characteristics (e.g., SES). Finally, this study only focused on Mexican-origin adolescents living in the Southwest, it will be important to explore whether adolescents’ experiences are similar in other geographical locations that do not have a predominant Mexican Latino population (e.g., Florida). In closing, the findings of this study underscore the need to consider social context and its influence on adolescents’ ethnic identity. Although these findings are limited with respect to the observed effect sizes, they replicate existing work and provide support for the existing literature, which demonstrates that a minimal correlation between ethnic identity and self-esteem exists and that the salience of ethnic identity varies according to social context. Furthermore, consistent with ecological theory, these findings challenge researchers to design future studies in ways such that multiple layers of context (and their influence on development) can be examined. Finally, it will be important for future studies explore these relationships with other ethnic minority groups in minority and majority contexts.