|Home | About | Journals | Submit | Contact Us | Français|
The present study examined ethnic, gender and age differences in perceived discrimination, and the association between perceived discrimination and psychological well-being in a nationally representative sample of Black adolescents. Data are from the National Survey of African Life (NSAL), which includes 810 African American and 360 Caribbean Black youth. Results indicate that the majority of Black youth perceived at least one discriminatory incident in the previous year. Adolescents at later stages of development perceived more discrimination than those at earlier stages, and African American and Caribbean Black males perceived more discrimination than their female counterparts. Perceptions of discrimination were positively linked to depressive symptoms and negatively linked to self-esteem and life satisfaction, regardless of ethnicity. However, Caribbean Black youth appear to be more vulnerable when they perceive high levels of discrimination.
Adolescence is a period of significant developmental changes for all youth. Specifically, improvements in reasoning and information processing are evidenced (Keating, 2004), pubertal changes are occurring (Sun et al., 2005) and identity development is prominent. Explorations and commitments with regard to racial or ethnic group membership require further efforts to crystallize an ethnic identity among many youth of color (Phinney, 1989). Thus, in addition to a heightened awareness of stress associated with the biological, cognitive, and social changes of adolescence, minority youth also contend with racial discrimination, which may negatively influence developmental paths. Theoretical work suggests that it is important to consider discrimination as a prominent environmental characteristic for minority youth (Garcia Coll et al., 1996) which may place them at risk for negative outcomes (Swanson et al., 2003).
In the present study, we rely on the Integrative Model by Garcia Coll and colleagues (1996) as the theoretical framework for examining perceptions of discrimination among Black adolescents. The Integrative Model proposed that American society stratifies individuals on the basis of social position variables, such as race, ethnicity, social class and gender (Garcia Coll et al., 1996). Further, social positions are influenced by racial discrimination, which is embedded within society and, therefore, believed to be a normative and frequent occurrence for children and adolescents of color (Garcia Coll et al., 1996). Developmental pathways are expected to be negatively influenced by racial discrimination for youth of color by placing them at higher risk for pessimistic outcomes and reducing the likelihood of positive outcomes (Garcia Coll et al., 1996). The present study examined whether the majority of Black youth perceived experiences of discrimination and whether these perceptions varied by ethnicity, gender and age. We also examined whether perceptions of discrimination were negatively associated with indicators of psychological well-being (i.e., depressive symptoms, self-esteem, and life satisfaction) and whether ethnicity moderated these associations.
Racial discrimination consists of dominant group members' actions, which are systematic and result in differential and negative effects on subordinate racial/ethnic groups (Williams, Neighbors & Jackson, 2003). Consistent with this proposition, previous research has indicated that 91% of pre-adolescent African Americans reported experiencing at least one discriminatory experience in their lifetime (Gibbons et al., 2004). Further, 52% of African American female adolescents indicated they experienced at least one discriminatory incident in the past year (Guthrie, Young, Williams, Boyd, & Kintner, 2002), and 77% of African American adolescents reported experiencing at least one discriminatory incident in the past three months (Prelow, Danoff-Burg, Swenson, & Pulgiano, 2004). Additionally, previous research has indicated that approximately 75% of Jamaican youth and 66% of Haitian youth perceived having been discriminated against in the past (Rumbaut, 1994). Despite the varying time periods used in prior research, the results suggest that the majority of African American and Caribbean Black youth perceived themselves to have been discriminated against.
A limitation of the Integrative Model is was the lack of articulation of expected ethnic, gender or age differences in perceived discrimination, even though historical evidence and empirical findings suggest differences may occur in some areas. Previous research indicated that African American youth perceived more racial discrimination than their Hispanic, Asian American or European American peers (Romero & Roberts, 1998; Fisher, Wallace & Fenton, 2000). Yet, most research has exclusively utilized African Americans to represent Black youth as Caribbean Black or other Black ethnic groups have rarely been studied in prior research (e.g., Rumbaut, 1994). Blacks of Caribbean ancestry are estimated to be the largest percentage of the Black immigrant population at 10% (Williams, Lavizzo-Mourey & Warren, 1994). The specific migration patterns, adaptive culture, and collective history of different ethnic groups have implications for adolescent development (Garcia Coll et al., 1996). Immigrant children are exposed to the belief that European Americans will respond more favorably to foreign-born Blacks (Waters, 1996). Thus, Caribbean Black adolescents may perceive less incidents of discrimination than their African American counterparts based on their cultural socialization. However, Caribbean Black youth who lack their parents' ethnic identity and labeled native-born Blacks (Rumbaut, 1994) may perceive equal or greater amounts of discrimination as their African American peers. A unique contribution of the current study is to identify differential risk in the areas of inquiry for African American and Caribbean Black youth.
Sidanius and Pratto (1999) argue that in a racial hierarchy, subordinate males as opposed to females will be the primary targets of discrimination. Specifically, the authors suggest that discrimination is practiced to reduce the competition for power among dominant and subordinate males. Thus, African American and Caribbean Black adolescent males may perceive more incidents of discrimination than their female counterparts because they may be perceived as more of a threat to the power base (i.e., white males) than African American or Caribbean Black females. Yet, previous research regarding gender differences in perceived discrimination among African American and Caribbean Black youth has been mixed. For example, two studies reported no gender differences in perceived racial discrimination among African American adolescents (Powell & Arriola, 2003; Scott, 2004). However, one study indicated that African American adolescent females perceived more racial discrimination than males (Gibbons et al., 2004), while another suggested that males perceived more racial discrimination than females (Matthews, Salomon, Kenyon & Zhou, 2005). Prior research among Caribbean Black youth suggests that adolescent males reported more racial harassment from European Americans and the police than adolescent females (Waters, 1996). One explanation for the conflicting findings includes the fact that some of the studies used urban samples (Powell & Arriola, 2003; Scott, 2004; Matthews et al., 2005; Waters, 1996), while one used a rural sample (Gibbons et al., 2004). Research among nationally representative samples of ethnic minority youth is necessary to enhance the external validity of findings with regard to gender differences in perceptions of discrimination among Black youth.
There are two developmental processes that occur during adolescence which could influence age differences in perceived discrimination: 1) formal reasoning and 2) racial or ethnic identity development. Formal reasoning, which is the ability to logically examine one's thoughts as well as the thoughts and intentions of others, develops during adolescence (Keating, 2004). Given the complexity of racial discrimination (Harrell, 2000), it is possible that youth at the later stage of adolescence who have attained formal reasoning, will perceive more incidents of discrimination compared to those at earlier stages of adolescence. Similarly, Phinney's (1989) model of racial/ethnic identity development proposes that early adolescents progress from an absence of exploration and commitment to an active stage of exploration and commitment. Consequently, adolescents engaged in exploration and commitment may have greater awareness of the importance of race (Phinney, 1989), resulting in increased perceptions of discrimination during late adolescence. Previous examinations of age differences in perceived racial discrimination have been mixed. For example, Powell and Arriola (2003) found no age differences in perceived discrimination among African American high school students youth while Romero and Robert (1998) found that older adolescents (aged 14-17) perceived more racial discrimination than younger adolescents (aged 10-13) in a study of ethnically diverse high school students youth. Additionally, prior research indicates that 12th graders reported more perceptions of racial discrimination than 9th or 11th graders among African American, Hispanic, Asian American and European American adolescents (Fisher et al., 2000). The current study provides an opportunity to examine age differences exclusively among Black youth, unlike the bulk of prior studies.
The Integrative Model proposes that racial discrimination reduces the capacity for minority youth to mature optimally because it increases the likelihood of negative developmental outcomes (Garcia Coll et al., 1996). Previous research has indicated that perceived racial discrimination was negatively related to African American adolescents' achievement motivation, grade point average and self-esteem (Powell & Arriola, 2003; Fisher et al., 2000) and positively associated with anger, depressive symptoms, and problem behaviors (Wong, Eccles & Sameroff, 2003; Prelow et al., 2004). Guthrie et al. (2002) found that increased perceptions of racial discrimination were associated with a 37% greater likelihood of smoking among African American adolescent females. The results are consistent for Caribbean Black youth in that previous reports of racial discrimination were linked to elevated depressive symptoms (Rumbaut, 1994). Consistent with cross-sectional studies, longitudinal research supports the association between perceptions of racial discrimination, psychological well-being and risky behaviors among African American youth. For example, a study of African American pre-adolescents indicated that perceived racial discrimination predicted subsequent psychological distress and substance use while the reverse relations were non-significant (Gibbons et al., 2004). The burgeoning research suggests that perceptions of racial discrimination were associated with diminished psychological well-being among African American and Caribbean Black youth and increased risky behaviors among African American youth.
The current study examines the prevalence of discrimination among African American and Caribbean Black youth in an effort to test and expand aspects of the Integrative Model related to the pervasiveness of discrimination in the lives of Black youth and the detrimental consequences of discrimination to their psychological well-being. In addition, we are expanding prior theoretical formulations of the Integrative Model (Garcia Coll et al., 1996) with support from empirical research (Gibbons et al., 2004; Prelow et al., 2004; Rumbaut, 1994) to suggest differential expectations for discrimination based on ethnicity, gender, and age. First, we are anticipating that the majority (at least 75%) of Black youth will report at least one discriminatory incident within the past year.
Next we are examining whether African American and Caribbean Black youth differ in their perceptions of discrimination. We hypothesize that African American and Caribbean Black youth will not differ in their perceptions of discrimination based on limited previous studies suggesting high levels of discrimination for both groups. Although the findings have been mixed, (Powell & Arriola, 2003; Powell & Arriola, 2003) we rely on the majority of studies to expect that Black males will perceive more discrimination that Black females, regardless of ethnicity. Age will be categorized as early (13-14), middle (15-16), and late (17) to reflect different developmental periods within adolescence, and we are anticipating that youth at late adolescence will perceive more discrimination than youth at early or middle adolescence (Fisher et al., 2000). Lastly, we examine differences in perceived discrimination based on the interactions between ethnicity, gender and age. These analyses are exploratory because previous research has not been conducted to provide sufficient guidance.
Finally, we will examine the relation between perceived discrimination and psychological well-being among African American and Caribbean Black youth while controlling for demographic factors. Previous research has indicated a link between perceived discrimination and various indicators of psychological well-being among African American youth (Wong et al., 2003; Fisher et al., 2000) and depressive symptoms among Caribbean Black youth (Rumbaut, 1994). We anticipate that perceived discrimination will be associated with increased depressive symptoms, lower self-esteem and lower life satisfaction among Black youth, regardless of ethnicity. However, in an effort to assess the influence of ethnicity, we also examine whether the relation between perceived discrimination and psychological well-being is moderated by ethnicity.
The participants were African American and Caribbean Black youth who participated in the National Survey of American Life (NSAL). The study was part of the NIMH Collaborative Epidemiology Survey (CPES) initiative that included three national representative surveys – the NSAL, the National Comorbidity Survey Replication (NCS-R), and the National Latino and Asian American Study (NLAAS) (Colpe, Merkiangas, Cuthbert & Bourdon, 2004). The NSAL, which consists of a nationwide survey of the African American, Caribbean Black and non-Hispanic White adult population, is based on a stratified, multi-stage area probability sample of the non-institutionalized civilian population in the 48 contiguous states (Jackson et al., 2004). The NSAL provides a comprehensive study of Black Americans with an emphasis on mental disorders, stressors, and risk/resilient factors (Jackson et al., 2004). The study received Institutional Review Board approval from the University of Michigan. It was administered by the Survey Research Center staff at the University of Michigan from February 2001 to June 2003.
The adolescent sample was drawn only from African American and Caribbean Black households. The original adolescent sample consisted of 1193 cases but 23 were dropped for analyses because they were 18 or older at the time of the interview. Thus, the resulting sample is 1170 African American (n=810) and Caribbean Black (n=360) youth ranging in age from 13 to 17 who were attached to the adult households. The overall sample was equally composed of males (N=563 unweighted, 48% weighted) and females (N=605 unweighted, 52% weighted), and there was an equal gender distribution for African American and Caribbean Black youth. The mean age was 15 (SD = 1.42), and the age groups were categorized as follows: early (age 13-14; N=477 unweighted, 40% weighted), middle (age 15-16; N=441 unweighted, 41% weighted), and late (age 17; N=252 unweighted, 19% weighted). Approximately 96% of the sample was still enrolled in high school and 9th grade was the average. The median family income was $28,000 (approximately $26,000 for African Americans and approximately $32,250 for Caribbean Blacks).
A national probability sample of households was drawn based on adult population estimates and power calculations for detecting differences among the adult samples. The African American sample served as the primary core sampling base for the entire study (Heeringa et al., 2004). The first stage of selection consisted of 21 self-representing metropolitan statistical areas (MSA), 27 non self-representing MSAs, and 16 non self-representing non-MSAs. MSAs are the primary stage units in national probability samples, and consist of geographically defined metropolitan counties defined by the U.S. Office of Management and Budget for federal, state and research purposes. There are approximately 350 MSAs, which consist of either metropolitan counties or combinations of counties that form distinct economic and labor areas (see Heeringa et al., 2004).
African American ancestry was defined as persons who self-identified as black but did not identify ancestral ties to the Caribbean. If an individual identified as Black and other races but did not identify ancestral ties to the Caribbean, they were classified as African American for purposes of the study. Size measures are household counts used to determine sample selection probabilities in multi-stage designs, and it was more optimal to use counts of African American households, as opposed to total U.S. household counts, for creating a nationally representative sample of African Americans. Size measures based on African American housing units were used for sample selection of 64 primary sampling units (PSU's), and 56 of these primary areas overlap substantially with existing Survey Research Center national sample primary areas. The remaining eight primary areas were chosen from the Southern region to represent African Americans proportionally according to their national distribution. From these 64 PSU's, 456 segments (blocks or block groups) were selected with probabilities proportional to the size of the African American population (Jackson et al., 2004). Caribbean ancestry was defined as persons who identified as black and indicated one or all of the following: 1) they were of West Indian or Caribbean descent, 2) they were from a country included on a list of Caribbean countries presented by the interviewer, and/or 3) their parents or grandparents were born in a Caribbean country. The Caribbean Black households were drawn from two sub-samples. The first consisted of households in the 456 African American segments. In order to achieve the expected sample size for Caribbean Black households, an additional eight primary sampling areas were selected in five states and the District of Columbia, which contain more than 80% of the Caribbean Black population in the United States. From these eight PSUs, 86 segments with the highest proportion of Caribbean Black households were selected.
Every African American and Caribbean Black household that included an adult participant was screened for an eligible adolescent living in the household, and adolescents were selected using a random selection procedure. If more than one adolescent was eligible for the study, two adolescents were selected based on the gender of the first selected adolescent, which resulted in non-independence for some households. As such, the adolescent supplement was weighted to adjust for non-independence in selection probabilities within households, as well as non-response rates across households and individuals. The weighted data were post-stratified to approximate the national population distributions for gender (males and females) and age (13, 14, 15, 16, and 17) subgroups among African American and Caribbean Black youth.
Prior to conducting the interview, informed consent was obtained from the adolescent's legal guardian and assent was obtained from the adolescent. Most of the adolescent interviews were conducted face-to-face using a computer-assisted instrument in their homes, but about 18% were conducted either entirely or partially by telephone. The interviews averaged 1 hour and 40 minutes for African American adolescents and 1 hour and 50 minutes for Caribbean Black adolescents. Respondents were paid $50 for their participation in the study. The overall response rate was 80.6% (80.4% for African Americans and 83.5% for Caribbean Blacks).
Adolescent gender, age and ethnicity were assessed with standard questions as part of the randomized respondent selection process used in the household sampling procedure for the study. Adolescents were asked to provide general demographic information such as grade level during the interview. Interviewers also recorded the gender of the adolescent respondent. Imputed family income was calculated based on information provided by the adult respondent for the household in which the adolescent lived for the year prior to the adult interview.
The Everyday Discrimination Scale assesses chronic, routine, and less overt experiences of discrimination that have occurred in the prior year (Williams, Yu, Jackson, & Anderson, 1997). The Everyday Discrimination Scale is a subtle measure of perceived discrimination that does not prime the respondent to think about race, which eliminates cues to prejudice prior to responding to the questions (Deitch et al., 2003). Previous studies that have used this approach find that African Americans report significantly more unfair treatment when compared to Whites (Deitch et al., 2003), which suggests that racial discrimination exists. This measure has been found to be correlated with measures of institutional racial discrimination and interpersonal prejudice (Hughes, 2003; Krieger et al., 2005). Thus, it appears to effectively tap into racialized discriminatory experiences that are often part of the life experiences of people of color. The original measure included ten items, but three items were added to reflect perceptions of teacher discrimination, resulting in a 13-item scale. It was developed and normed among adult samples; therefore, we conducted factor analyses to examine its' psychometric properties with an adolescent sample. Exploratory and confirmatory factor analyses were conducted on random half samples and the results indicated a one-factor structure, which is consistent with results from adult samples (Krieger, Smith, Naishadham, Hartman & Barbeau, 2005; Williams et al., 1997). The stem question is: “In your day-to-day life, how often have any of the following things happened to you?” Sample items include: “People act as if they think you are dishonest” and “You are followed around in stores.” The Likert response scale for frequencies ranged from 1 (“never”) to 6 (“almost everyday”), (α = .86). The reliability scores reflect the current sample for this particular measure and all others. The reports of experiencing each event were counted to capture the number of discriminatory events that occurred within the previous year. Specifically, the responses were dichotomously coded to indicate whether an event occurred versus an event never occurring. Higher scores indicate a greater number of events that occurred in the previous year, regardless of the frequency for each event.
The shortened version of the CES-D assesses the frequency of depressive symptoms experienced within the past week (Radloff, 1977). Although originally developed as a 20-item measure, the CES-D has been shortened from 20 items to 12 items and has been used with adolescent populations in previous research (Roberts et al., 1999). The Likert scale (α = .68) consists of responses ranging from 0 (“rarely”) to 3 (“most or all of the time”). Sample items include “I did not feel like eating, my appetite was poor” and “My sleep was restless”. Higher scores were indicative of higher depressive symptoms.
The Rosenberg Self-esteem scale is an assessment of self-acceptance (Rosenberg, 1965). The 10-item Likert scale (α = .72) consists of rating items with responses ranging from 1 (“strongly disagree”) to 4 (“strongly agree”). Negative items were reversed so higher scores represent higher levels of self-esteem. Sample items include “I feel that I have a number of good qualities” and “On the whole, I am satisfied with myself.”
One question was used to assess adolescent perceptions of general life satisfaction (Campbell, 1976). The item read “How satisfied with your life as a whole would you say you are these days?” The responses ranged from 1 (“very satisfied”) to 4 (“very dissatisfied”). The rating items were reversed so a high score represents a high level of life satisfaction.
STATA 9.1 was used to calculate the complex design-based estimates of variance. Actual numbers are reported for sample sizes, while weighted data are used in the analyses. The initial analyses consisted of a frequency count of the perceived discrimination variable separately by ethnic, gender and age groups. Poisson regression analyses were conducted to study main and interaction effects of the demographic correlates (i.e., ethnicity, gender and age) on perceived discrimination. Poisson regression analyses were chosen because perceived discrimination is a count variable and does not assume a normal distribution. Age was categorized into early, middle and late to examine age differences. Linear regression analyses were conducted to assess if perceived discrimination was associated with depressive symptoms, self-esteem and life satisfaction while controlling for gender, age (continuous variable to maximize variance), ethnicity and family income. An ethnicity × discrimination interaction term was included in the analyses to examine if the relation between perceived discrimination and the psychological well-being indicators varied across the two ethnic groups. The analytical techniques adjusted the standard errors to account for the complex sample design of the NSAL, which involved multiple stages, clustering and stratification. Standard errors adjusted for complex design effects are usually larger than non-adjusted standard errors. In this study, the standard errors for Caribbean Blacks were typically higher than those for African Americans because the Caribbean Black sample is significantly more clustered than the African American sample. Consequently, ethnic differences that appeared to be large were not necessarily statistically significant.
Descriptive information regarding the study variables was presented in Table 1. On average, Black youth perceived five incidents of discrimination out of a possible thirteen. The prevalence of discrimination by ethnic, gender and age groups was presented in Tables 2 and and3.3. Eighty-seven percent of African American youth and 90% of Caribbean Black youth indicated that they had experienced at least one discriminatory incident in the past year. The most common incident across the three demographic groups was “People act as if they are better than you are,” and the least common incident was “Your teachers act as if they are afraid of you.”
Separate analyses using Poisson Regression were conducted to determine main and interaction effects for ethnicity, gender and age on perceived discrimination, while controlling for family income. There were no differences in perceptions of discrimination among African American (M = 5.11, SE = .21) and Caribbean Black (M = 5.22, SE = .54) youth. There was a main effect for gender (B = -.13, p < .05), with boys (M = 5.49, SE = .27) reporting more discriminatory incidents than girls (M = 4.80, SE = .20). There was also a main effect for age (B = .24, p < .01) with late adolescents (M = 5.97, SE = .33) reporting more discrimination than early (M = 4.64, SE = .25) or middle age adolescents (M = 5.19, SE = .28). The gender × ethnicity interaction (B = -.47, p < .05) was significant and the results indicated that African American males (M = 5.40, SE = .28) perceived more discrimination than African American females (M = 4.83, SE = .22), and that Caribbean Black males (M = 6.11, SE = .86) perceived more discrimination than Caribbean Black females (M = 4.51, SE = .34). There was a trend (p < .10) for mean differences between African American males and Caribbean Black females, but there were no differences in perceived discrimination among African American and Caribbean Black males.
Hierarchical regression analyses were conducted to assess if perceived discrimination was a risk factor for the psychological well-being indicators (depressive symptoms, self-esteem and life satisfaction) among Black youth. A Bonferroni correction was used such that an alpha of .02 (.05/3) was the minimum significance level. Demographic characteristics (i.e., ethnicity, gender, age and familial income) were controlled before regressing depressive symptoms, self-esteem and life satisfaction on perceived discrimination. Tables 4--66 present the results of these analyses. None of the demographic variables predicted depressive symptoms or self-esteem; however, gender and age were related to life satisfaction. Specifically, boys (B = -.12, p < .01) and late adolescents had lower life satisfaction levels (B = -.05, p < .01) (see Model 1, Table 6). As expected, perceived discrimination was associated with increased depressive symptoms (B = .17, p < .01), decreased self-esteem (B = -.02, p < .01) and decreased life satisfaction (B = -.04, p < .01) among Black youth (see Model 1, Tables 4--66).
An interaction term (ethnicity × perceived discrimination) was included in the analyses to assess if the relation between perceived discrimination and psychological well-being was moderated by ethnicity (Model 2, Tables 4 – 6). The interaction term (ethnicity × perceived discrimination) was significant for depressive symptoms (B = -.28, p < .02) and self-esteem (B = .01, p < .02). The Aiken and West procedure was used to graph the significant interactions. Figure 1 indicates that at high levels of perceived discrimination, both African American and Caribbean Black youth report high levels of depressive symptoms. With regard to self-esteem, Caribbean Black youth report lower levels of self-esteem than African American youth at high levels of perceived discrimination (see Figure 2).
The results of the present study indicate that the majority of African American and Caribbean Black youth report at least one discriminatory incident within the past year. On average, African American and Caribbean Black youth report experiencing five of the 13 assessed discriminatory incidents. These findings are consistent with prior empirical research (Guthrie et al., 2002; Prelow et al., 2004), and theoretical formulations. Specifically, the results provide support for the Integrative Model's hypothesis that racial discrimination is pervasive and normative for children and adolescents of color (Garcia Coll et al., 1996).
The results of the present study suggest that African American and Caribbean Black youth perceive equal amounts of discriminatory incidents. One explanation for this finding is that because Caribbean Black youth lack their parents' immigrant characteristics, they may perceive that they are subject to similar treatment as African Americans. Thus, race may be as ubiquitous for Caribbean Black youth as it is for African American youth despite their varying backgrounds. Waters (1996) reported that almost half of Caribbean Black youth downplayed their ethnic backgrounds as distinct from African Americans. Similarly, they considered racial discrimination to be pervasive and did not believe they were protected from racial discrimination because of their immigrant origins or their parents' identifying characteristics (Waters, 1996). Consequently, Caribbean Black youth may discover that race is a more salient construct than ethnicity, resulting in their equivalent perceptions of discrimination with African American youth.
Gender differences are evident in that males perceive more discriminatory incidents than females, which is consistent with prior research (Matthews et al., 2005; Waters, 1996). The gender × ethnicity interaction also suggests that this gender difference occurs within the two ethnic groups. African American and Caribbean Black adolescent males perceive more discrimination than their female counterparts. One explanation may be that discrimination is gendered such that all members of subordinate groups are not equally targeted (Sidanius & Pratto, 1999). Social Dominance Theory suggests that in racially based hierarchical systems, subordinate men will be the primary targets of discrimination as opposed to women (Sidanius & Pratto, 1999), and males should perceive more discrimination than females. Gender differences in discrimination also may depend on the type of discrimination examined (Krieger & Sidney, 1996). For example, no gender differences were apparent among African American adults in school, employment and medical care settings, but males perceived more discriminatory incidents with police and in the judicial system (Krieger & Sidney, 1996). Yet, prior research examining objective experiences of discrimination indicates that African American females were more likely to be discriminated against in housing markets than African American males (Fischer & Massey, 2004). Future research should assess the congruence between subjective perceptions of discrimination and objective experiences of discrimination among Black youth. For example, are Black males more likely to be targeted for specific types of discrimination such as racial profiling, and is this consistent with their subjective perceptions? An additional explanation for the main effect of gender concerns the parental racial socialization messages provided to adolescent males and females. Prior work indicates that African American parents are likely to discuss discrimination with their adolescents (Hughes, 2003), and that African American girls are more likely to receive racial pride messages whereas boys are more likely to receive racial barrier messages (Thomas & Speight, 1999). Given the emphatic difference, adolescent males may be primed to perceive more discriminatory incidents than their female counterparts.
The results of the present study also indicate that adolescents at later stages of development perceive more incidents of discrimination than early and middle stage adolescents, which is also consistent with prior research (Romero & Roberts, 1998; Fisher et al., 2000). One explanation may be that late adolescence is the period when individuals solidify their identities as it relates to race/ethnicity (Phinney, 1989). Two facets of identity development among Black youth may influence their perceptions of being discriminated against. From a developmental perspective, early adolescents may not have explored what it means to be a member of their racial/ethnic group, whereas older adolescents have which might account for differences in perceived discrimination. Previous research indicates that being in the exploration phase of racial identity development positively predicts perceptions of racial discrimination among Caribbean Black adults (Hall & Carter, 2006). Similarly, racial identity content may influence the degree to which discrimination is perceived. African American adults who consider race to be an important aspect of their self-concept perceive more racially discriminatory incidents than their counterparts (Sellers, Caldwell, Schmeelk-Cone, & Zimmerman, 2003). It is also possible that older youth, particularly Black males, are more likely to be targeted for discriminatory treatment compared to younger adolescents. Future research should examine racial identity changes in development and content with perceptions of discrimination among African American and Caribbean Black youth.
The present findings are consistent with theoretical formulations suggesting that discrimination increases the likelihood of negative outcomes for minority youth (Swanson et al., 2003; Garcia Coll et al., 1996) and with empirical findings that perceptions of discrimination are linked to diminished psychological well-being among African American and Caribbean Black youth (Prelow et al., 2004; Rumbaut, 1994). Given that cross-sectional and longitudinal research has consistently shown that perceived discrimination is negatively linked to a variety of developmental outcomes, a central question concerns whether some adolescent populations are more vulnerable to discriminatory experiences than others. The present results suggest that perceived discrimination may be more harmful for Caribbean Black youth's psychological well-being than African American youth. One explanation may be that immigrant parents reinforce the notion that hard work and educational attainment are the primary determinants for success in this country (Waters, 1996). Yet, experiences of racial discrimination are inconsistent with these beliefs and Caribbean Black youth may have a difficult time reconciling their parents' philosophy with racially discriminatory experiences. Additionally, Caribbean Black youth may lack adequate coping skills for discriminatory incidents when they are perceived, and may develop an adversarial stance toward the dominant society contradicting their immigrant parents' philosophy (Rumbaut, 1994). The discrepancy between Caribbean Black youth's socialization and their experiences with discriminatory treatment may result in a vulnerability to discriminatory experiences. Future research should address the question: Which adolescent subgroups are most vulnerable to experiences of discrimination? For example, given the gender and age differences in perceptions of discrimination, are males and older adolescents more affected by perceptions of discrimination than their female and younger counterparts?
Future research should also consider the mechanisms by which perceived discrimination is linked to psychological well-being for adolescents of color. The stress paradigm (Lazarus & Folkman, 1984) posits that mediators (i.e., coping strategies) and moderators (i.e., person characteristics) influence and alter the relation between stressors and health. Are there distinct mediators and moderators for perceived discrimination among African American and Caribbean Black youth? Previous research indicates that discrimination distress results in increased internalized and externalized coping strategies among African American youth (Scott, 2004). Are perceptions of discrimination linked to increased depressive symptoms through internalized coping strategies as opposed to externalized coping strategies for Black youth? Racial identity has been shown to moderate the relation between perceptions of discrimination and psychological well-being among African American youth (Sellers et al., 2006). Yet, is this relation true for Caribbean Black youth? Lastly, there is a dearth of research examining specificity models, or simultaneous analyses of mediators and moderators for stressors and physical/mental health (Grant et al., 2003). For example, is perceived discrimination linked to psychological well-being via a specific mediating process (i.e., coping strategies) in the context of a particular moderating variable (i.e., racial identity) for Black adolescents? As future research attempts to understand which adolescent subgroups are more likely to be affected by perceptions of discrimination, the specific mechanisms by which discrimination is linked to psychological well-being will also be understood.
There are several limitations in the present study that need to be considered. First, the cross-sectional nature of the study prevents causality from being inferred such that we cannot articulate that perceptions of discrimination predict subsequent psychological well-being. Yet, the utilization of nationally representative samples of African American and Caribbean Black youth affords generalizations to youth of similar backgrounds. An additional limitation concerns the measurement of perceived discrimination, which is believed to be a critical area in discrimination research (Williams et al., 2003). Though the instrument in the present study utilizes an annual time period, smaller time periods and experiential sampling procedures might capture discriminatory experiences more precisely. As such, the use of annual periods may not adequately capture the phenomenon of discrimination among an adolescent sample. Another limitation concerns the notion that the measure makes no inherent racial attribution assumptions in the stem or with individual questions; discrimination may be considered a developmental as opposed to a racial phenomenon for adolescents. Yet, prior empirical research indicates that the predominant reason for perceived discrimination was race/ethnicity among African American youth, and age for European American youth (Matthews, Salomon, Kenyon & Zhou, 2005). Thus, discrimination appears to be racially interpreted for African American youth and developmentally interpreted for European American youth.
The results of the present study enhance empirical and theoretical literature examining adolescent perceptions of discrimination. Specifically, the majority of African American and Caribbean Black youth perceive they have been the victims of discriminatory treatment within the past year. Although discrimination is pervasive, all minority youth are not equally affected in that male adolescents and those at later stages of development perceive more discrimination than their counterparts. Lastly, perceptions of discrimination are associated with diminished psychological well-being for African American and Caribbean Black youth, although Caribbean Black youth appear to be more vulnerable in the context of high levels of perceived discrimination. The results are consistent with theoretical formulations suggesting that discrimination is a stressor for minority youth, during a normatively stressful period of development.
This research was supported by contract (U01-MH-57716) from the National Institute of Mental Health and the Office of Behavioral and Social Sciences Research at the National Institutes of Health. Preparation of this article was supported by a fellowship to the first author through a National Institute of Mental Health Postdoctoral Training Grant (NIMH T32 MH 067555). We thank everyone who participated in conducting the National Survey of American Life (NSAL).
Publisher's Disclaimer: The following manuscript is the final accepted manuscript. It has not been subjected to the final copyediting, fact-checking, and proofreading required for formal publication. It is not the definitive, publisher-authenticated version. The American Psychological Association and its Council of Editors disclaim any responsibility or liabilities for errors or omissions of this manuscript version, any version derived from this manuscript by NIH, or other third parties. The published version is available at http://www.apa.org/journals/dev/
Eleanor K. Seaton, Department of Psychology, University of North Carolina at Chapel Hill.
Cleopatra H. Caldwell, School of Public Health, University of Michigan.
Robert M. Sellers, Department of Psychology, University of Michigan.
James S. Jackson, Institute for Social Research and Department of Psychology, University of Michigan.