Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Feb 10.
Published in final edited form as: J Soc Serv Res. 2024 Jun 10;50(5):758–770. doi: 10.1080/01488376.2024.2358249

Impact of Perceived Racial Discrimination on Depressive Symptoms and School Outcomes among Asian American Youth

Tae Yeun Kim 1, Yoonsun Choi 2, Miwa Yasui 2, Julia Henly 2, Tracy Harachi 3
PMCID: PMC12885568  NIHMSID: NIHMS2126347  PMID: 41669193

Abstract

Using two survey datasets--the Korean American Families(KAF) Project and the Cross Cultural Families(CCF) Project, this study examined how perceived racial discrimination influences depressive symptoms and school outcomes among Asian American adolescents, and what factors mediate the association between perceived discrimination and youth outcomes, focusing on Korean American(KA) youth (n=220), Cambodian American(CA) youth (n=164), and Vietnamese American(VA) youth (n=163). Using Structural Equation Modeling, the results revealed perceived discrimination was positively associated with depressive symptoms for KA and VA youths, but school misbehavior only for CA youth. Among KA youth, perceived discrimination was positively associated with ethnic identity, in turn positively related to social support, thereby reducing depressive symptoms. Among CA and VA youths, perceived discrimination was not associated with ethnic identity, but ethnic identity was related to social support, which in turn was associated with lower depressive symptoms, less school misbehaviors and better grades. This study offers insights into the way in which perceived racial discrimination may exert an impact on children’s mental health and school outcomes and how the mechanisms that explain this relationship vary across AA subgroups. The findings of this study highlight the need for sustained research in the future to advance research and practice for AA groups.

Keywords: perceived discrimination, ethnic identity, co-ethnic friendship, social support, youth outcomes, Asian American youth

Instruction

Asian Americans (AA) are one of the fastest-growing populations in the United States, yet AA adolescents remain understudied and misunderstood. They typically receive attention for their successful academic achievement as a “model minority,” but rarely for their difficulties and problems. They show a unique developmental pattern—namely, poor psychological well-being despite fewer problem behaviors and better school outcomes (J. Lee & Zhou, 2014). For example, although AA students outperform White students in terms of their overall or average grades (GPA), grades and test scores in math, and average SAT scores (Byun & Park, 2012; Kao & Thompson, 2003), the prevalence of depression among them is equivalent or greater than that among European Americans (Gee, 2004; Kuo, 1984).

These paradoxical and unique patterns also exist at the subgroup level. For instance, Korean American (KA) adolescents report significantly higher depressive symptoms than European American children (E. Kim & Cain, 2008), although they present higher academic outcomes at school (Zhou & Kim, 2006). Although approximately 24% of children in the United States experience depressive symptoms (Duggal et al., 2001), 39% of KA children experience depressive symptoms (E. Kim & Cain, 2008). Vietnamese American (VA) youth also show remarkably high levels of academic outcomes. They surpass the levels of educational attainment of native-born Whites and African Americans despite their families’ disadvantaged economic backgrounds and lack of middle-class cultural capital (Lee & Zhou, 2014). In contrast, studies suggest that Cambodian American (CA) youth are likely to struggle academically, as well as experience high levels of depression and post-traumatic stress disorder (PTSD) (Marshall, Schell, Elliott, Berthold, & Chun, 2005; Sack, Clarke, & Seeley, 1996). High dropout rates (approximately 40%) are reported for CA youth, which are comparable to the rates for Latinos, African Americans, and American Indian groups (U.S. Census Bureau, 2000, 2010).

It is still an open question that why some students fare well academically but others suffer from poor school outcomes, and/or why Asian youth who do well academically experience psychological difficulties. Perceived racial discrimination might be one of the most important but unexplored factors for explaining the unique pattern of developmental outcomes among AA adolescents. Asian Americans face prejudice and discrimination (Alvarez, Juang, & Liang, 2006) in various ways, such as housing, school, and employment (Young & Takeuchi, 1998). Among children, discrimination often takes the form of peer discrimination, which has been found to be significantly higher for AA youth than for youth from any other group (Rosenbloom & Way, 2004). As peers are one of the most important developmental contexts for adolescents, peer discrimination against Asian Americans by their non-Asian peers is a major challenge (Rosenbloom & Way, 2004). Despite the recent increase in interest regarding the role of racial discrimination during adolescence (e.g., Benner & Kim, 2009; Greene, Way, & Pahl, 2006), studies on the impact of perceived racial discrimination on developmental outcomes among AA youth are still relatively few compared to those focusing on adults’ outcomes (e.g., Lee, 2003).

More sophisticated understanding of the relation between perceived racial discrimination and youth outcomes, for example, investigating mediating variables between them may also help explain the paradoxical pattern of development outcomes among AA youth. From an array of possible variables, this study focuses on co-ethnic friendship, ethnic identity, and social support as mediators between perceived discrimination and youth outcomes.

Present Study

A central aim of the present study is to empirically test the pathways by which perceived racial discrimination affects the mental health and school outcomes of CA-, KA-, and VA youths. Importantly, this study focuses on not only direct relations between perceived racial discrimination and developmental outcomes but also the indirect effects that other factors (i.e., ethnic identity, co-ethnic friendship, social support) may have on the discrimination-outcomes relationship (Figure 1). This study will finally see whether and how the hypothesized relationships vary across those three Asian subgroups. The present study uses the social rejection-identification model (Nyla R. Branscombe et al., 1999), and a set of several empirical studies to construct hypothesized relationships among the main study constructs—namely, perceived racial discrimination, ethnic identity, co-ethnic friendship, social support, and developmental outcomes.

Figure 1.

Figure 1.

Open in a new tab

Conceptual Model

This study hypothesizes that perceived racial discrimination exerts a direct and negative effect on mental health and school outcomes. This study also hypothesizes that perceived discrimination will have an indirect and less detrimental effect when ethnic identity, co-ethnic friendship, and social support mediate the relationship between perceived racial discrimination and developmental outcomes. Specifically, perceived racial discrimination is expected to be positively associated with depressive symptoms and school misbehavior as well as negatively associated with grades in school. Perceived racial discrimination is also hypothesized to be positively associated with ethnic identity and co-ethnic friendship. Further, ethnic identity and co-ethnic friendship are hypothesized to be positively related to social support, which in turn is positively associated with developmental outcomes (i.e., less depressive symptoms, better GPA, less misbehavior in school).

Methods

Data

Data are from the Korean American Families (KAF) project and the Cross Cultural Families (CCF) project. The KAF sample is derived from a Midwest metropolitan area. In the first wave of data collection in 2007, KA youths (n = 220) and their parents were surveyed through person-to-person interviews conducted by trained bilingual interviewers. Korean American and immigrant families with early adolescents (ages 11–14) were eligible to participate. Participants were recruited from three sampling sources: phonebooks, public-school rosters, and Korean church or temple rosters. Approximately equal numbers of samples from each source participated, and further analyses revealed no statistically different demographics (e.g., age, gender) or socio-demographic data, such as the level of household income and parental education across the three sources.

The CCF is a longitudinal study examining the developmental trajectories of a panel of CA (n = 164) and VA (n = 163) youth and their mothers. In 2001, a random sample of families with a child enrolled in Grades 3–6 were contacted in an urban Pacific Northwest school district. The data collected for the five-year longitudinal study consisted of annual maternal interviews, youth interviews, teachers’ reports of student behaviors, and school records.

This study used Time 1 data for KA youth (KAF) collected in 2007 and Wave 4 data for CA and VA youth (CCF) collected in 2004, in which youth participants’ age ranges were comparable between the three groups. Although it was not possible to merge these two datasets due to reasons such as different items and response options, the current study analyzed the common variables and constructs—namely, depressive symptoms, grades, misbehavior in school, ethnic identity, social support, and co-ethnic friendship—to explore variations in the relationships across the three subgroups.

Measures

While the measures of the variables in the KAF data consist of 5-point Likert scales, those in the CCF data consist of 4-point Likert scales. The measures of KAF data were adjusted to 4-point Likert items to make the datasets as comparable as possible. Unless noted otherwise, response options for all measures were 1 (NO!), 2 (no), 3 (yes) and 4 (YES!).

Perceived Discrimination

Perceived discrimination (PD) was measured using a scale based on a composite of measures developed by Phinney, Madden, and Santos (1998) and Verkuyten (1998). In the KAF data, the PD scale included six items: Four items assessed the perceived frequency of being treated unfairly or negatively due to participants’ ethnic background by peers, teachers, and other Americans while two items assessed how often they feel that other Americans behave unfairly toward Koreans or have something against Koreans. Examples included “How often do you feel that other students/your teachers treat you unfairly because you are Korean?” and “How often do you feel that you are called names in school because you are Korean?” In the CCF data, the PD was measured using two items: “I am treated unfairly by the people in my neighborhood, because I am (Vietnamese or Cambodian)” and “I am treated unfairly by the people in my school, because I am (Vietnamese or Cambodian).” Response options were (1) never, (2) rarely, (3) sometimes, (4) often. The alpha reliability coefficient is .79 for Koreans, .58 for Vietnamese, and .72 for Cambodians.

Ethnic Identity

Ethnic identity in KAF consists of 7 items adopted from the Multi-Ethnic Identity Measure (MEIM) proposed by Phinney (1992). The questions asked how much pride the youth have in what Koreans have accomplished, how well the youth understand what being Korean means, the extent to which their lives will be affected by being Korean, how they relate to their own group, such as through activities in organizations or social groups that mostly include Koreans, etc. In CCF, ethnic identity consists of 20 items adopted from Phinney’s (1992) MEIM scale as well as eight items adopted from the General Ethnicity Questionnaire set forth developed by Tsai et al. (2000). The items from Tsai et al. (2000) include “I was raised in a way that was Vietnamese/Cambodian,” “I admire people that are Vietnamese/Cambodian,” and “I want to be accepted by Vietnamese/Cambodians.”

This study used six of the seven items that appeared in both datasets. Examples included “I have spent time trying to find out more about my own ethnic group, such as its history, traditions, and customs,” “I am active in organizations or social groups that include mostly members of my own ethnic group,” and “I have a strong sense of belonging to my own ethnic group.” The item “I am not very clear about the role of being Korean/my ethnicity in my life” was dropped due to the very low corrected item-total correlation (r= .110). The alpha reliability coefficient was .76 for KA, .63 for VA, and .61 for CA adolescents.

Co-ethnic Friendship

Friends’ ethnic composition was also assessed by the same question in both the KAF and CCF datasets. The question asking for the percentage of co-ethnic friends (i.e., Thinking about all of your friends, about what percentage of your friends are Korean, Vietnamese, or Cambodian?) was used in this study to examine the number of co-ethnic friends respondents had compared to friends from other ethnic groups. Response options were (1) none (0%), (2) a few (1%–25%), (3) some (26%–50%), (4) quite a few (51%–75%), and (5) a lot (76%–100%).

Social Support

Social support was assessed using six items in KAF and one item in CCF. The common item in the two datasets was about emotional support: “There is at least one teacher or other adult at school I can talk to if I have a problem.” The KAF included five additional items on social support, concerning social support from peers and adults in an ethnic church or the community/neighborhood. Examples include “Other students in my class help me do my best,” “Friends at church who are not in my school help me do my best,” “There is at least one adult in the community or neighborhood I can talk to if I have a problem.” (α=.71 for KA youth).

Youth Outcomes: Depressive Symptoms, Grades, Misbehavior in School

Items to measure depressive symptoms were identical across all three subgroups. 14 items from the Children’s Depression Inventory (Angold, Costello, Messer, & Pickles, 1995) and the Seattle Personality Questionnaire for Children (Kusche, Greenberg, & Beilke, 1988) assessed the frequency of depressive symptoms during the two weeks prior to the survey. Examples included “I was very restless,” “I felt miserable or unhappy,” and “I feel like crying a lot of the time.” (α=.91 for Koreans, α=.91 for Cambodians, and α=.92 for Vietnamese).

For grades, the KAF asked four items, “During the most recent grading period or last grading period, what was your grade in English or language arts/mathematics/history/science?” while the CCF asked one item (i.e., “In general, what grades do you mostly get?”). Grades were calculated by averaging the four items for KA youth. Response options were ranged from 4 for A to 1 for D. The alpha reliability coefficient was .70 for KA youth.

For misbehavior in school, the KAF originally included five items: repeating a grade, receiving an out-of-school suspension, being expelled from school, skipping classes, and being sent out of the classroom. Among these items, two were used: “In the past year, how many times have you skipped classes or school without an excuse?” and “In the past year, how many times have you been sent out of the classroom for doing something wrong?” Repeating a grade, suspension, and expulsion items were excluded from the analysis because most KA youths have not had these behaviors (i.e., 100%, 95.9%, and 94.4%). Response options were the number of the times that they had experienced these behaviors. Similar to the KAF data, the CCF originally had five items assessing misbehavior in school, including suspension, expulsion, and cheating on school tests. Most VA and CA youths did not report the experiences of suspension (i.e., 90% of VA youth, 80% of CA youth) or expulsion (99.4% of VA youth, 97.4% of CA youth). Thus, the two items that corresponded with those in KAF were used. Response options were (1) Never, (2) 1 or 2 times, (3) 3 or 4 times, and (4) 5 or more times. The alpha reliability coefficient was .42 for KA, .54 for VA, and .51 for CA adolescents.

Demographic Controls

Three control variables were adjusted in the model—age of youth, gender of youth, and parent report of family socioeconomic status (SES)—to account for their impact on youth outcomes. Family SES was coded 1 (Yes) if they have received public assistance and 0 (No) if they had not. The gender variable was coded 1 (boys) and 0 (girls).

Analytic Procedures

The analysis of the present study involved several steps. Before testing the conceptual model, we conducted several univariate and bivariate descriptive analyses including means and standard deviations at the individual item and construct levels, item-total correlations, and reliability of each construct and pairwise correlations among main study constructs. Items that did not meet the criteria (e.g., less than .3 item-total correlations) were removed. Specifically, descriptive statistics were generated to examine three outcome variables (i.e., depressive symptoms, grades, and misbehavior in school) as well as perceived discrimination, social support, ethnic identity, and co-ethnic friendship. Preliminary analyses were also conducted for the demographic variables of gender, socioeconomic status, and age. Second, an independent-samples t-test was conducted for all variables to assess whether CA and VA adolescents were significantly different or not. KA youth were not compared because KA youth samples were from a different dataset (KAF) that could not be merged with that of the VA and CA samples (CCF). Third, bivariate correlations were examined to test the relationships among the study variables.

Upon completion of the descriptive analyses, path analysis within a structural equation modeling (SEM) framework was conducted to test the relationships in the hypothesized model. When testing the proposed models, three parallel models were analyzed to test the predictions for depressive symptoms and school outcomes for each subgroup (KA, CA, and VA adolescents). We used M Plus 8.10 (Muthén & Muthén, 2013) to examine the path models. Path models were run with correlations among mediators, and controlled for youth age, gender, and parent reported SES.

The fit of all models was assessed by examining the model chi-square (χ2), the Comparative Fit Indices (CFI, Bentler, 1990), and the Root Mean Squared Error Approximation (RMSEA, Browne & Cudeck, 1993). Because no single index is ideal, the fit of all models is reported as a minimal set of fit indices. A good fit is indicated by CFI values of greater than .90 (Byrne, 1994), and RMSEA values of less than .05 (Browne & Cudeck, 1993). Values between .05 and .08 for RMSEA indicate a fair fit, and values greater than .10 represent a poor fit (Browne & Cudeck, 1993).

After testing the path models separately for each group, a multiple group path analysis was also conducted to determine whether the hypothesized relationships were equivalent across VA and CA youth. A multiple-group path analysis was conducted only for CA and VA youths, not for KA youth, because—as previously explained—the KA youth samples could not be merged with the CA and VA youth samples. In conducting the multiple-group path analysis, the fit of the model was first examined without constraints for each ethnic group (i.e., allowing them to vary), followed by a constrained model that was equivalent across subgroups (Kline, 2010). The analysis can demonstrate whether the constrained model has statistically significantly poorer fit than the unconstrained model using the change of χ2 relative to the change in degrees of freedom (Muthén & Muthén, 2013). If the fit of the constrained model is significantly worse than that of the unconstrained model, it indicates that statistically significant group differences exist in the magnitudes of path parameters across the two ethnic subgroups.

We used Maximum Likelihood (ML) for estimating the models and handling missing data, because ML estimators are appropriate for continuous dependent variables (Wang & Wang, 2012), which is the case for depressive symptoms, grades, and misbehavior in school.

Results

Path Models

To examine the difference among subgroups, the path models were tested separately by subgroups. The results of the path models are summarized in Figure 2, 3, and 4. In the path model of KA youth (Figure 2), fit indices were χ2 (17) = 29.879, p =.0272, CFI = .858, and RMSEA = .067 (R2 = 21% for depressive symptoms, R2 = 7% for grades, R2 = 7% for misbehavior in school). Statistically significant paths with a p-value of .05 or less for KA youth (Table 1) were the following: (1) perceived racial discrimination ➔ depressive symptoms (β = .153), (2) perceived racial discrimination ➔ ethnic identity (β = .147), (3) ethnic identity ➔ social support (β = .208), and (4) social support ➔ depressive symptoms (β = −.391). More specifically, perceived racial discrimination was positively associated with depressive symptoms. Yet, perceived racial discrimination was not associated with the other two outcome measures (i.e. grades and school misbehavior). Perceived racial discrimination was positively related to ethnic identity, which was positively associated with social support, which was in turn related to less depressive symptoms. Tests of indirect effects indicated that perceived discrimination was indirectly related to depressive symptoms via ethnic identity and social support (p < .05).

Figure 2.

Figure 2.

Open in a new tab

Path model of KA youth. Note: Paths are shown when they are significant; Fit indices: χ2(17) = 29.879, p = 0.0272, CFI = 0.858, RMSEA = 0.067; Path coefficients are standardized, *p < .05; ** p< .01; ***p < .001; Several paths are not shown for simplicity, e.g., three controls for the outcome variables (age, gender, and family SES) and covariations among three independent variables on the far left-hand side and among mediating variables.

Figure 3.

Figure 3.

Open in a new tab

Path model of VA youth. Note: Paths are shown when they are significant; Fit indices: χ2(17) = 41.991, p = 0.0007, CFI = 0.835, RMSEA = 0.099; Path coefficients are standardized, *p < .05; ** p< .01; ***p < .001; Several paths are not shown for simplicity, e.g., three controls for the outcome variables (age, gender, and family SES) and covariations among three independent variables on the far left-hand side and among mediating variables.

Figure 4.

Figure 4.

Open in a new tab

Path model of CA youth. Note: Paths are shown when they are significant; Fit indices: χ2(17) = 28.294, p = 0.0416, CFI = 0.867, RMSEA = 0.068; Path coefficients are standardized, *p < .05; ** p< .01; ***p < .001; Several paths are not shown for simplicity, e.g., three controls for the outcome variables (age, gender, and family SES) and covariations among three independent variables on the far left-hand side and among mediating variables.

Table 1.

Path Statistics for KA Adolescents

Estimates Estimate/SE P-value
Perceived Discrimination➔ Depressive Symptoms .153 2.207 .027
Social Support➔ Depressive Symptoms −.391 −5.972 .000
Perceived Discrimination➔ Grades −.023 −.312 .755
Social Support ➔ Grades .054 .725 .469
Perceived Discrimination➔ Misbehavior in School .112 1.498 .134
Social Support ➔ Misbehavior in School −.032 −.431 .667
Ethnic Identity ➔ Social Support .208 2.701 .007
Co-ethnic Friendship ➔ Social Support .041 .506 .613
Perceived Discrimination ➔ Ethnic Identity .147 1.955 .040
Perceived Discrimination ➔ Co-ethnic Friendship .039 .502 .615
Open in a new tab

In the path model of VA youth (Figure 3), fit indices were χ2 (17) = 41.991, p =.0007, CFI = .835, and RMSEA = .099 (R2 = 20% for depressive symptoms, R2 = 23% for grades, R2 = 15% for misbehavior in school). Statistically significant paths with p-values of .05 or less for VA youth group (Table 2) were the following: (1) perceived discrimination ➔ depressive symptoms (β = .149), (2) ethnic identity ➔ social support (β = .349), (3) social support ➔ depressive symptoms (β = −.411), (4) social support ➔ GPA (β = .353), and (5) social support ➔ misbehaviors in school (β = −.219). Similar to KA adolescents, perceived discrimination was positively associated with depressive symptoms; however, there was no significant path between perceived discrimination and ethnic identity and between perceived discrimination and co-ethnic friendship. Instead, ethnic identity was positively associated with social support, which in turn was negatively associated with depressive symptoms and misbehavior in school, and positively associated with grades. Path models for VA youth showed no mediated effect of perceived racial discrimination on youth outcomes, yet, ethnic identity had significant indirect effects on depressive symptoms (p < .01), grades (p < .01) and misbehavior in school (p < .05) via social support.

Table 2.

Path Statistics for VA Adolescents

Estimates Estimate/SE P-value
Perceived Discrimination➔ Depressive Symptoms .149 1.975 .048
Social Support➔ Depressive Symptoms −.411 −5.852 .000
Perceived Discrimination➔ Grades .099 1.342 .180
Social Support ➔ Grades .353 4.912 .000
Perceived Discrimination➔ Misbehavior in School .003 .044 .965
Social Support ➔ Misbehavior in School −.216 −2.805 .005
Ethnic Identity ➔ Social Support .349 3.726 .000
Co-ethnic Friendship ➔ Social Support −.126 −1.553 .120
Perceived Discrimination ➔ Ethnic Identity .077 .950 .342
Perceived Discrimination ➔ Co-ethnic Friendship .076 .932 .352
Open in a new tab

In the path model of CA youth (Figure 4), fit indices were χ2(17) = 28.294, p = .0416, CFI = 0.867, and RMSEA = 0.068 (R2 = 5% for depressive symptoms, R2 = 16% for grades, R2 = 19% for misbehavior in school). Statistically significant paths with p-values of .05 or less for CA youth group (Table 3) were the following: (1) perceived discrimination ➔ misbehaviors in school (β = .139), (2) ethnic identity ➔ social support (β = .209), (3) co-ethnic friends ➔ social support (β = .157), (4) social support ➔ depressive symptoms (β = −.184), (5) social support ➔ GPA (β = .195), and (6) social support ➔ misbehaviors in school (β = −.224). Notably, perceived racial discrimination was positively associated with misbehaviors in school, but not with other two outcomes, such as depressive symptoms or grades. In addition, no significant relationship emerged between perceived racial discrimination and ethnic identity, and between perceived racial discrimination and co-ethnic friendship. Instead, positive relationships existed between ethnic identity and social support as well as between co-ethnic friendship and social support. The path model of CA youth showed indirect effects of ethnic identity on youth outcomes via social support.

Table 3.

Path Statistics for CA Adolescents

Estimates Estimate/SE P-value
Perceived Discrimination➔ Depressive Symptoms .005 .065 .949
Social Support➔ Depressive Symptoms −.184 −2.233 .026
Perceived Discrimination➔ Grades .114 1.468 .142
Social Support ➔ Grades .195 2.534 .011
Perceived Discrimination➔ Misbehavior in School .139 1.826 .050
Social Support ➔ Misbehavior in School −.224 −2.964 .003
Ethnic Identity ➔ Social Support .209 2.024 .043
Co-ethnic Friendship ➔ Social Support .157 2.039 .041
Perceived Discrimination ➔ Ethnic Identity −.050 −.596 .551
Perceived Discrimination ➔ Co-ethnic Friendship −.074 −.889 .374
Open in a new tab

Multiple Group Path Analysis

In addition to testing the path models separately for each group, statistical tests of group differences were made across VA and CA youths. The unconstrained path model, where all parameters were allowed to vary between the two subgroups, provided an adequate fit to the data [χ2 (34) = 70.285, p = .0003, CFI = .847, RMSEA = .085]. Constraining the parameters in the three-path mediation model to be equal across the two AA subgroups did not provide as good of a fit [χ2 (62) = 170.445, p =.0000, CFI = .109, RMSEA = .541]. When the fit of the constrained models was compared with that of the unrestricted models without the equality constraints, the fully unconstrained model fit significantly better, as shown by the statistically significant change in chi-square [Δχ2 = 100.160, Δdf = 28; p < .05]. These results suggested that making the structural parameters equal across the two subgroups resulted in a statistically significant worsening of the overall model fit. Thus, CA and VA groups are significantly different and should not be constrained to be equal.

Discussion

Experiences of racial discrimination among AA youth in the United States have received little attention. AA youths tend to be perceived as model minorities, implicating a sense of imperviousness to youth Experiences of racial discrimination among AA youth in the United States have received little attention. AA youths tend to be perceived as model minorities, implicating a sense of imperviousness to youth outcomes (Tummala-Narra, Alegria, & Chen, 2012). To date, evidence is lacking on the impact of perceived discrimination among AA youths. Particularly, existing research has less commonly investigated the pathways through which perceived discrimination leads to youth developmental outcomes. This study is the first step to fill the gap in the literature. This study affirmed that AA adolescents do perceive racial discrimination in the present day and that these perceptions are indeed highly related to a variety of developmental outcomes, specifically depressive symptoms and school outcomes. Also, this study examined the way in which the effects of perceived racial discrimination could be mediated by various critical factors (e.g., ethnic identity, co-ethnic friendship, social support), and whether and how the mechanisms vary across AA subgroups.

Consistent with perceptions of racial discrimination as being stressful and capable of harming mental health and school outcomes (e.g., Gee, Spencer, Chen, Yip, & Takeuchi, 2007; Wei, Ku, Russell, Mallinckrodt, & Liao, 2008), the results obtained in this study provide empirical evidence that the direct relationship between perceived racial discrimination and well-being was negative. Specifically, when KA youth perceived racial discrimination, they were more likely to have depressive symptoms; this was also the case for VA youth. On the other hand, when CA youth perceived racial discrimination, they were more likely to become involved in misbehaviors at school, without experiencing more mental health consequences.

The findings also provided insights into the pathways through which perceived discrimination leads to youth developmental outcomes. We found evidence in support of the social rejection–identification model among KA youth. As hypothesized, perceived racial discrimination exerted a direct and harmful impact on depressive symptoms, however, at the same time, perceived discrimination had a positive effect on psychological well-being when mediated by ethnic identity and social support. When KA youth perceived racial discrimination, they came to have a clearer sense of ethnic identity. KA youth with a strong ethnic identity were more likely to obtain social support from peers or adults in school, their community, ethnic church, etc. In turn, they felt less depressed. This finding indicates that the perception of adversity can be positive (if not, less detrimental), to some extent, to youth outcomes if the adversity can serve as an opportunity to strengthen a protective factor (i.e., ethnic identity).

On the other hand, the social rejection-identification model was not well supported by the results of VA and CA youths. Among VA youth, perceived racial discrimination had a direct and negative effect on depressive symptoms, as hypothesized. However, in contrast to the results of KA youth, perceived discrimination was not significantly associated with ethnic identity. In other words, ethnic identity did not mediate the effect of perceived discrimination on depressive symptoms. Among CA youth, perceived racial discrimination had no direct relationship with psychological well-being (i.e., depressive symptoms) as well as ethnic identity.

This study also showed that CA youth had the lowest grades and the most frequent misbehavior in school among the three subgroups. Unlike CA youth, VA youth had high grades (similar with those of KA youth) and less misbehavior in school (but a little more misbehavior than KA youth). In general, VA youth are known for their high school achievement that overrides their low SES (Zhou & Bankston, 1994). The current study provided insights on how VA youth can succeed despite their low SES as well as which factors will work to promote CA youth’s poor school outcomes. Among CA and VA youth, social support was found to be a strong and focal factor for their depressive symptoms and school outcomes. Consistent with existing literature (e.g., Mossakowski, 2003), findings of this study revealed that co-ethnic friendship and ethnic identity helped them receive critical social support (e.g., educational resources, social networks, role model), which in turn promoted school outcomes (and decreased depressive symptoms). Co-ethnic ties and social support can be important sources of social capital for students’ school outcomes and mental health outcomes. These results are consistent with previous literature that found that the Vietnamese community can promote academic achievement in various ways (Zhou & Bankston III, 1994).

This study adds nuances to our understanding of how perceived discrimination exerts an impact on children’s mental health and school outcomes as well as how the mechanisms vary across AA subgroups. To date, much of research explored correlates or predictors for outcomes separately but not together in a single model. This study focused on how those predictors interplay one another in a single model. Also, most of previous research used aggregated AA adult samples as a monolithic group. Although the results might not be generalizable to all AA groups, this study contributes to filling the gap in the literature by exploring both East Asians and Southeast Asians. It is believed that this study can help us revisit the pervasive model minority stereotype and contribute to our understanding of the diversity among Asian populations and their divergent youth outcomes. Additional comparative analyses across AA subgroups are warranted to discern both common and unique mechanisms in youth development.

Further, findings of the KA path model were in support of the social rejection-identification model. It revealed that perceived racial discrimination had a direct and negative impact on depressive symptoms and simultaneously had a positive impact when mediated by ethnic identity. Yet, the social rejection-identification model was not supported by the results of CA and VA youths. Even when ethnic identity played a mediating role, whether ethnic identity is a mediator or a moderator may not be clear cut. Our findings suggest that this question is not yet fully answered. Thus, this question is another inquiry that future studies should explore.

This study provides detailed empirical evidence to guide the development of effective interventions for AA youths. The results offer several implications for practice. The findings imply that researchers and practitioners should attend to the way in which AA youths’ lives are affected by perceived discrimination. In addition, ethnic identity, co-ethnic friendship, and social support should be targeted for intervention as such efforts could improve youths’ mental health and school outcomes across all AA subgroups. The positive association between perceived discrimination and depressive symptoms for KA and VA youths suggests the necessity for mental health services that specifically target these groups. Developing specialized interventions, such as counseling, support groups and workshops on coping strategies, could help mitigate the psychological effects of discrimination. Observations that CA youth are prone to misbehavior in response to perceived discrimination underscore the need for behavioral interventions and mentorship programs within schools. These programs should aim to help students manage and channel their responses to discrimination constructively, addressing the unique stressors faced by these students and promoting positive behavioral outcomes.

The mediating role of ethnic identity and social support in improving outcomes indicates that social services should focus on strengthening these aspects. Social services could implement cultural competence programs that foster a strong sense of ethnic identification among AA youths. Programs that enhance cultural pride and connectivity can play a crucial role in reducing depressive symptoms and improving academic performance. Initiatives could include community group activities that celebrate cultural heritage and facilitate peer connections.

This study not only contributes to the academic understanding of discrimination’s effects but also provides a practical framework for developing interventions tailored to the unique needs of different AA subgroups. The differential impact on Korean, Cambodian, and Vietnamese American youths highlights the importance of customized service delivery. Social services should consider the unique cultural, social, and psychological needs of each subgroup when designing interventions. This could mean different strategies for engagement, support, and intervention based on the specific challenges and strengths identified within each group. Social workers and educators should be trained to recognize and address the unique needs of each subgroup, ensuring that interventions are culturally and contextually appropriate. Further, the findings from this study are essential for informing policy makers and stakeholders about the specific needs of AA subgroups in combating the effects of racial discrimination. Advocacy efforts could focus on promoting policies that provide greater resources for mental health and educational support tailored to these ethnic groups. Additionally, training programs for educators and social workers should incorporate elements of cultural competence and strategies to foster supportive and inclusive environments.

By implementing these strategies, social service providers can more effectively support AA youths in navigating the challenges of racial discrimination. Such insights are crucial as they suggest that well-targeted interventions not only address immediate needs but also contribute to the long-term well-being and resilience of these AA youths, thereby enhancing community resilience and empowerment.

Despite the unique contributions that this study makes to the literature, its limitations need to be addressed in future research. The first limitation is the directionality of the model. Due to the use of cross-sectional data, it is possible that the relationships among the constructs may conceivably be run in the opposite way. For example, although perceived discrimination may determine ethnic identity, it is also possible that people who have a high ethnic identity might be more likely to perceive discrimination against the in-group (Crocker & Major, 1989). We need more longitudinal data to establish temporal sequences and account for the interactive nature of youth development. Second, although the study included three AA subgroups to examine and compare the hypothesized models, the results may not be generalizable to other Asian American groups. Although Asian Americans share a common culture, they vary widely in the impact of perceived discrimination on youth outcomes. Additional comparative analyses across Asian subgroups (e.g., Filipino) will be conducted to discern common as well as unique mechanisms in youth developmental outcomes. Finally, although the models accounted for family SES, the KA samples were predominantly from middle-class and highly educated families, whereas CA and VA samples were largely low income and undereducated families. Thus, additional data with greater SES variability are needed to better understand AA families and their children in the United States.

The Asian Americans are the fastest-growing populations in the United States, and their numbers are expected to grow at an even faster rate. Yet due to the pervasive model minority stereotype, the diversity within AA groups has been largely ignored or inadequately described. It is hoped that this study enhances our understanding that Asian Americans are a heterogeneous group that needs more attention. It is believed that the findings of this study highlight the need for sustained research in the future to advance both research and practice for Asian American groups.

Acknowledgments

This study was supported by the National Institute of Mental Health (NIMH, K01 MH069910) to the second author.

References

  1. Alvarez AN, Juang L, & Liang CTH (2006). Asian Americans and racism: when bad things happen to “model minorities.”. Cultural diversity & ethnic minority psychology, 12(3), 477–492. doi: 10.1037/1099-9809.12.3.477 [DOI] [PubMed] [Google Scholar]
  2. Abrams D, Hogg MA, & Marques JM (2005). A social psychological framework for understanding social inclusion and exclusion.
  3. Angold A, Costello EJ, Messer SC, & Pickles A (1995). Development of a short questionnaire for use in epidemiological studies of depression in children and adolescents. International Journal of Methods in Psychiatric Research, 5(4), 237–249. [Google Scholar]
  4. Benner AD, & Kim SY (2009). Experiences of discrimination among Chinese American adolescents and the consequences for socioemotional and academic development. Developmental Psychology, 45(6), 1682–1694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bentler PM (1990). Fit indexes, Lagrange multipliers, constraint changes and incomplete data in structural models. Multivariate Behavioral Research, 25(2), 163–172. [DOI] [PubMed] [Google Scholar]
  6. Branscombe NR, Schmitt MT, & Harvey RD (1999). Perceiving pervasive discrimination among African Americans: Implications for group identification and well-being. Journal of Personality and Social Psychology, 77(1), 135–149. doi: 10.1037/0022-3514.77.1.135 [DOI] [Google Scholar]
  7. Branscombe NR, Schmitt MT, & Harvey RD (1999). Perceiving pervasive discrimination among African Americans: Implications for group identification and well-being. Journal of personality and social psychology, 77(1), 135. [Google Scholar]
  8. Browne MW, & Cudeck R (1993). Alternative ways of assessing model fit. In Bollen KA & Long JS (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage. [Google Scholar]
  9. Byrne BM (1994). Structural equation modeling with EQS and EQS/Windows: Basic concepts, applications, and programming. Thousand Oaks: Sage. [Google Scholar]
  10. Byun S. y., & Park H (2012). The Academic Success of East Asian American Youth: The Role of Shadow Education. Sociology of Education, 85(1), 40–60. doi: 10.1177/0038040711417009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Caplan NS, Choy MH, & Whitmore JK (1991). Children of the Boat People: A Study of Educational Success: University of Michigan Press. [Google Scholar]
  12. Cassidy C, O’Connor RC, Howe C, & Warden D (2004). Perceived discrimination and psychological distress: The role of personal and ethnic self-esteem. Journal of Counseling Psychology, 51(3), 329–339. doi: 10.1037/0022-0167.51.3.329 [DOI] [Google Scholar]
  13. Choi Y (2008). Diversity within: Subgroup differences of youth behaviors among Asian Pacific Islander American adolescents. Journal of Community Psychology, 36(3), 352–370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Choi Y (2008). Diversity within: Subgroup differences of youth problem behaviors among Asian Pacific Islander American adolescents. Journal of community psychology, 36(3), 352–370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Choi Y, He M, & Harachi T (2008). Intergenerational Cultural Dissonance, Parent–Child Conflict and Bonding, and Youth Problem Behaviors among Vietnamese and Cambodian Immigrant Families. Journal of Youth and Adolescence, 37(1), 85–96. doi: 10.1007/s10964-007-9217-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chong V, Um K, Hahn M, Pheng D, Yee C, & Auerswald C (2009). Toward an intersectional understanding of violence and resilience: An exploratory study of young Southeast Asian men in Alameda and Contra Costa County, California. Aggression and Violent Behavior, 14, 461–469. [Google Scholar]
  17. Eccles JS, Wong CA, & Peck SC (2006). Ethnicity as a social context for the development of African-American adolescents. Journal of School Psychology, 44(5), 407–426. [Google Scholar]
  18. Gee GC, Spencer M, Chen J, Yip T, & Takeuchi DT (2007). The association between self-reported racial discrimination and 12-month DSM-IV mental disorders among Asian Americans nationwide. Social Science & Medicine, 64(10), 1984–1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Greene ML, Way N, & Pahl K (2006). Trajectories of perceived adult and peer discrimination among Black, Latino, and Asian American adolescents: patterns and psychological correlates. Dev Psychol, 42(2), 218–236. doi: 10.1037/0012-1649.42.2.218 [DOI] [PubMed] [Google Scholar]
  20. Hwang W-C, & Goto S (2008). The impact of perceived racial discrimination on the mental health of Asian American and Latino college students. Cultural Diversity and Ethnic Minority Psychology, 14(4), 326–335. [DOI] [PubMed] [Google Scholar]
  21. Kao G (2004). Parental Influences on the Educational Outcomes of Immigrant Youth. International Migration Review, 38(2), 427–449. [Google Scholar]
  22. Kao G, & Thompson JS (2003). Racial and ethnic stratification in educational achievement and attainment. Annual Review of Sociology, 417–442.
  23. Kim E, & Cain KC (2008). Korean American adolescent depression and parenting. Journal of Child and Adolescent Psychiatric Nursing, 21(2), 105–115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kim I (1981). New urban immigrants: the Korean community in New York. Princeton, N.J.: Princeton University Press. [Google Scholar]
  25. Kim RY (2002). Ethnic differences in academic achievement between Vietnamese and Cambodian children: Cultural and structural explanations. Sociological Quarterly, 43(2), 213–235. doi: 10.1525/tsq.2002.43.2.213 [DOI] [Google Scholar]
  26. Kinzie JD, Fredrickson RH, Ben R, Fleck J, & Karls W (1984). Posttraumatic stress disorder among survivors of Cambodian concentration camps. American Journal of Psychiatry, 141(5), 645–650. [DOI] [PubMed] [Google Scholar]
  27. Kristof ND (2006). The model students. New York Times. [Google Scholar]
  28. Kusche CA, Greenberg MT, & Beilke R (1988). Seattle Personality Questionnaire for Young School-Aged Children. Department of Psychology. University of Washington. Seattle. [Google Scholar]
  29. Lazarus RS, & Folkman S (1984). Stress, appraisal, and coping. New York: Springer Pub. Co. [Google Scholar]
  30. Lee J, & Zhou M (2014). The Success Frame and Achievement Paradox: The Costs and Consequences for Asian Americans. Race and Social Problems, 6(1), 38–55. doi: 10.1007/s12552-014-9112-7 [DOI] [Google Scholar]
  31. Lee RM, Su J, & Yoshida E (2005). Coping with intergenerational family conflict among Asian American college students. Journal of Counseling Psychology, 52(3), 389–399. [Google Scholar]
  32. Lee SJ (1996). Unraveling the “model minority” stereotype. New York: Teachers College, Columbia University. [Google Scholar]
  33. Leong FTL (2007). Handbook of Asian American psychology (2nd ed.). Thousand Oaks, Calif.: Sage Publications. [Google Scholar]
  34. Long PDP, Dư PL, & Ricard L (1997). The dream shattered: Vietnamese gangs in America: UPNE. [Google Scholar]
  35. Louie VS (2004). Compelled to excel : immigration, education, and opportunity among Chinese Americans. Stanford, Calif.: Stanford University Press. [Google Scholar]
  36. Marshall GN, Schell TL, Elliott MN, Berthold SM, & Chun C-A (2005). Mental health of Cambodian refugees 2 decades after resettlement in the United States. Jama, 294(5), 571–579. [DOI] [PubMed] [Google Scholar]
  37. Masten AS (2001). Ordinary magic - Resilience processes in development. American Psychologist, 56(3), 227–238. doi: Doi 10.1037//0003-066x.56.3.227 [DOI] [PubMed] [Google Scholar]
  38. Mollica RF, Wyshak G, & Lavelle J (1987). The psychosocial impact of war trauma and torture on Southeast Asian reguees. American Journal of Psychiatry, 144(12), 1567–1572. [DOI] [PubMed] [Google Scholar]
  39. Mossakowski KN (2003). Coping with perceived discrimination: does ethnic identity protect mental health? Journal of Health and Social Behavior, 318–331. [PubMed]
  40. Muthén LK, & Muthén BO (2013). Mplus User’s Guide. Los Angeles, CA: Muthen & Muthen. [Google Scholar]
  41. Ngo B, & Lee SJ (2007). Complicating the image of model minority success: A review of Southeast Asian American education. Review of Educational Research, 77(4), 415–453. [Google Scholar]
  42. Portes A, & Rumbaut RG (2006). Immigrant America: a portrait: Univ of California Press.
  43. Portes A, & Rumbaut RG (Eds.). (2001). Legacies: the story of the immigrant second generation. Berkeley, New York: University of California Press & Russell Sage Foundation. [Google Scholar]
  44. Portes A, & Rumbaut R n. G. (2006). Immigrant America : a portrait (3rd ed.). Berkeley: University of California Press. [Google Scholar]
  45. Rhee S (2009). The Impact of Immigration and Acculturation on the Mental Health of Asian Americans: Overview of Epidemiology and Clinical Implications. In Trinh N-H, Rho YC, Lu FG & Sanders KM (Eds.), Handbook of Mental Health and Acculturation in Asian American Families (pp. 81–98). Totowa, NJ: Humana Press. [Google Scholar]
  46. Rosenbloom SR, & Way N (2004). Experiences of discrimination among African American, Asian American and Latino adolescents in an urban high school. Youth & Society, 35(4), 420–451. [Google Scholar]
  47. Rumbaut RG (1995). The new Californians: comparative research findings on the educational progress of immigrant children. In Rumbaut RG & Cornelius W (Eds.), California’s immigrant children (pp. 17–70). San Diego: Center for U.S.-Mexican Studies, UC San Diego. [Google Scholar]
  48. Sack WH, Clarke GN, & Seeley J (1996). Multiple forms of stress in Cambodian adolescent refugees. Child Development, 67(1), 107–116. [PubMed] [Google Scholar]
  49. Suárez-Orozco C, & Suárez-Orozco MM (2001). Children of immigration. Cambridge, Mass.: Harvard University Press. [Google Scholar]
  50. Sue S, & Okazaki S (1990). Asian-American educational achievements: A phenomenon in search of an explanation. American Psychologist, 45(8), 913–920. [DOI] [PubMed] [Google Scholar]
  51. Tajfel H, & Turner JC (2004). The Social Identity Theory of Intergroup Behavior. In Sidanius JTJJ (Ed.), Political psychology: Key readings (pp. 276–293). New York, NY, US: Psychology Press. [Google Scholar]
  52. Takeuchi DT, Chung RC-Y, Lin K-M, Shen H, Kurasaki K, Chun C-A, & Sue S (2014). Lifetime and twelve-month prevalence rates of major depressive episodes and dysthymia among Chinese Americans in Los Angeles. [DOI] [PubMed]
  53. Tang W, Weil MH, Sun S, Duggal C, Johnson B, & Maldonado F (1991). Gastric Intramural Pco2 as an Indicator of Organ Ischemia during Hemorrhagic-Shock. Faseb Journal, 5(6), A1724–A1724. [Google Scholar]
  54. Thorsteinsson EB, Ryan SM, & Sveinbjornsdottir S (2013). The mediating effects of social support and coping on the stress-depression relationship in rural and urban adolescents. Open Journal of Depression, 2(01), 1. [Google Scholar]
  55. Tummala-Narra P, Alegria M, & Chen CN (2012). Perceived Discrimination, Acculturative Stress, and Depression Among South Asians: Mixed Findings. Asian American Journal of Psychology, 3(1), 3–16. doi: Doi 10.1037/A0024661 [DOI] [Google Scholar]
  56. Wang J, & Wang X (2012). Structural Equation Modeling: Applications Using Mplus: Wiley. [Google Scholar]
  57. Wei MF, Ku TY, Russell DW, Mallinckrodt B, & Liao KYH (2008). Moderating Effects of Three Coping Strategies and Self-Esteem on Perceived Discrimination and Depressive Symptoms: A Minority Stress Model for Asian International Students. Journal of Counseling Psychology, 55(4), 451–462. [DOI] [PubMed] [Google Scholar]
  58. Zhou M, & Bankston CL III (1994). Social capital and the adaptation of the second generation: The case of Vietnamese youth in New Orleans. International Migration Review, 821–845.
  59. Zhou M, & Gatewood JV (2000). Contemporary Asian America. New York: New York University. [Google Scholar]

RESOURCES