Association for Public Policy Analysis & Management

Although socioeconomic gradients in health are broadly acknowledged, little research has carefully assessed the distributional and contextual complexities of associations between income and health.  This paper focuses on mental and behavioral health outcomes (i.e., depression, anxiety, substance use, delinquency) which typically emerge during adolescence and can have far-reaching implications for lifelong health and economic productivity.  Research has long pointed to family and neighborhood poverty as a risk factor for heightened mental and behavioral health problems among youth, but recent work suggests that there may be risks associated with affluence as well. Specifically, a series of studies by Luthar has identified high levels of mental and behavioral health problems among youth in high affluence suburban schools.  Given the conflated nature of economic resources across the numerous proximal contexts in which human development unfolds, we argue that it is essential to identify the unique origins of risk across multiple contexts of poverty and affluence.

This paper uses data from a nationally representative sample of high school students (Add Health; N=13,304) drawn from 132 schools across the U.S. Data were drawn from student, parent, schoolmate, and administrator interviews as well as Census data.  Family income was reported by parents, and then aggregated to the school level to assess school income. Neighborhood income was delineated at the census block level.  Youth reported on psychosocial functioning in the arenas of depression, intoxication, illicit drug use, violent behavior, and property crime using validated measures. 

Multilevel 0-inflated negative binomial and OLS regression analyses found that school income was consistently associated with youth outcomes, with few effects of family or neighborhood income (see Tables 1 and 2).  Youth attending schools with more affluent peers reported a heightened likelihood of intoxication, drug use, and property crime, but also a lower likelihood of interpersonal violence and fewer depressive symptoms.   Neighborhood and family income, in contrast, were far less predictive, suggesting that economic risks for youth mental and behavioral health problems may be driven by school and peer contexts.  Patterns were largely similar for male and female youth.  Additional analyses will consider interactive effects between family, school, and neighborhood income, and will further assess whether patterns vary across racial/ethnic group membership.

Results highlight the central role of schoolmates in promoting or protecting against psychosocial risk, and raise concerns for educational policy makers seeking to improve the functioning of youth.