Race Matters in School-Based Preventive Interventions: Disaggregating Program Effects on Positive Youth Development Assets

Background:  Recent studies of school-based prevention programs have illustrated the potential for propensity scoring methods to establish links between service use and student outcomes, but few have used this approach to test program impact on youth development assets. Moreover, it is unclear whether the relationship between school-based service use and positive outcomes found in previous studies holds across subgroups of students. In light of growing evidence of cultural influences on help-seeking and ongoing racial mistreatment in the educational system, the effects of school-based interventions are likely attenuated for students of color. To test this hypothesis, this study used propensity score methods to estimate the impact of participation in school-based prevention programming on youth development assets by race.

Methods:  The current study uses epidemiological data from the California Healthy Kids Survey administered in the spring of 2009 to all students attending 15 high schools with school-based mental health prevention programs in a large urban district. A 70% response rate yielded a sample of 7,314 youth that were predominantly Asian, Latino, and Black. Dependent variables were student-reported youth development assets at school (e.g., caring relationships, high expectations, and meaningful participation). The primary independent variable captured students’ frequency of utilization with five response categories (never, 1 or 2 times, 3-5 times, 6-10 times, and 10+ times). Pre-treatment and other covariates included student and school demographics and youths’ self-reported early risk behaviors. Propensity scoring methods were used to estimate relationships between service use and assets.

Results:  Students’ use of services was positively associated with all youth developmental assets of interest (Cohen’s ds ranged from .10 to .50). However, there were differences in the strength of the relationship between program participation and assets by race, depending on the type of asset measured and frequency of service use. Among Black students, positive relationships between service use and assets were only evident at high rates of service use (more than 10 times). For Latino students, moderate program participation (3-5 and 6-10 times) was associated with assets. Finally, among Asian students, only very low service use (1 or 2 times) or very high service use (more than 10 times) was positively associated with assets.

Conclusions and Implications:  The findings suggest that participation in school-based prevention programs is positively related to youth development assets at school. Theories of cultural influences on adolescent service use and treatment outcomes suggest three possible explanations for differences in the relationships between service use and assets by race: (1) different pathways to services (e.g., voluntary or coercive), (2) varied levels of service engagement, and (3) service type (e.g., case management vs. therapy). Results suggest the need for further clarification of the interplay between student racial background, pathways into program participation and types of services.