Methods: This study used waves 1 (8th/9th graders) and 2 (11th/12th graders) of the Children of Immigrants Longitudinal Study (N=2,453) that surveyed youth from California (San Diego) and Florida (Miami, Fort Lauderdale). The dependent variable was depressive symptoms in the past week (4-question version of CESD-depression scale). The primary independent variable (main stressor of interest) was ever having the experience of feeling discriminated against due to one's race/ethnicity. Other stressors were: family cohesion, conflict with parents, socioeconomic status index, and dissonant acculturation (parent-youth discordance in preference in doing “American way,” parent-youth difference in preferred language). Finally, gender, race, years in U.S., bilingual ability, self-esteem, depression and discrimination at wave 1 were controlled. The study used quantile regression with bootstrapped standard errors (1000 replications) to identify heterogeneous associations between discrimination and depression with respect to the youth's relative position in the distribution of depressive outcomes.
Results: Approximately 51% of the sample experienced perceived racial/ethnic discrimination and about 75% reported some level of depressive symptoms in the past week. Quantile regression results indicated that perceived discrimination had increasingly larger association at higher percentiles. In detail, the estimated coefficients for youth in the 25th, 50th, 75th, and 90th percentiles were, 0.08 (p=0.297), 0.25 (p=0.023), 0.29 (p=0.060), and 0.53 (p=0.037), respectively. Also, stress due to lack of family cohesion and parent-youth conflict were consistently strong predictors of depression throughout the different percentiles. Acculturative stress was only significant among youth in the 75th and 90th percentiles of depression.
Conclusions and Implications: This study found the persisting effects of perceived racial/ethnic discrimination even when accounting for other stressors in an adolescent's daily life experiences (economic hardship, family cohesion, parent-youth conflict, and acculturative stress). In particular, results indicated that adolescents in the 90th percentile for depression were most vulnerable to the detrimental effects of perceived discrimination. As such, examining the heterogeneous discrimination-mental health relationship with quantile regression can help understand the how individuals in the upper part of the mental health distribution are systematically different from those in the lower part of the distribution. Understanding that an individual's mental health status may lead to differential response to perceived racial/ethnic discrimination can guide practitioners in identifying target risk groups. Furthermore, this study also highlighted the importance of family cohesion and provided a strong support for an ecological approach to intervention programs.