A Multilevel Analysis of the Association Between School Climate Dimensions and Adolescent Depressive Symptoms

Purpose: Prevalence estimates indicate 11% of youth meet the criteria for a major depressive disorder with approximately 8% having impairment (Merikangas et al., 2010). Given the impairment related to depressive symptoms, identifying risk and protective factors is critical. The literature points to individual and family factors associated with depressive symptoms (Dallaire et al., 2008). The school context also exerts a significant influence on youth development including mental health outcomes (Gadeyne et al., 2006). However, there are gaps in the current literature including a need for more research that uses multilevel modeling to relate school climate with depressive symptoms. Also, there is a need for examination into whether subgroups of youth such as sexual or racial minority youth are at greater risk or will benefit more from specific factors relative to depressive symptoms (Kuperminc et al., 2001). To address these gaps, the aims of this study were: (1) to examine the association between school climate dimensions and adolescent depressive symptoms and (2) to assess the extent to which this relationship varies for racial and sexual minority youth.

Methods: A secondary analysis was conducted using the National Longitudinal Study of Adolescent Health (Add Health). School climate was conceptualized and measured with five constructs: perceived school connectedness, perceived teacher support, harshness of discipline policies, presence of mental health and social services at school, and median school-level income. Depressive symptoms were measured with 19-items from the Center for Epidemiologic Studies Depression (CES-D) Scale. The study sample included adolescents (N=9,524) and schools (N=105) with completed responses on the key study variables. Multilevel mixed-effects linear regression was employed to address the research aims.

Results: Controlling for individual and school level demographic variables, higher perceived school connectedness (β=1.29; SE=.1; p<0.001), higher perceived teacher support (β=.84; SE=.09; p<0.001), and higher median school level income (β= -.01; SE=.006; p<0.001) were significantly associated with fewer depressive symptoms in the multilevel model. There was a statistically significant difference between racial majority and minority youth (t (9524) =59.70, p=0.00) as well as between sexual majority and minority youth (t (9524) =17.51, p=0.00) on depressive symptoms at Wave II.  Racial (M=12.3) and sexual minority youth (M=13.5) had significantly more depressive symptoms than majority youth. Belonging to a racial minority (β=.73; SE=.14; p<0.001) or sexual minority group (β=.85; SE=.27; p<0.01) were strong, significant predictors of depressive symptoms. The relationships between each school climate dimension and depressive symptoms did not vary for racial or sexual minority youth.

Implications: The results from the current study provide implications for social work practice, policy, and future research. School social workers can play a critical role in promoting adolescent connectedness to school and fostering positive student-teacher relationships. Policies to support school climate initiatives particularly in relation to accountability and technical support for school climate initiatives are also needed. Directions for future research are provided including additional exploration into the relationship between school discipline policies and mental health outcomes in youth and a review of school climate measures to ensure they accurately assess the school experiences of minority youth.