Family structure and cohesion, and depressive symptoms in adolescents

This study assessed whether the presence of both natural parents in the home and the level of perceived emotional bonding in the family, as measured by the cohesion component of the Family Adaptability and Cohesion Evaluation Scales, were independent predictors of depressive symptoms as measured by the Center for Epidemiologic Studies Depression Scale (CES-D) in a diverse community sample of 3,191 young adolescents. In cross-sectional analysis, cohesion was significantly associated with depressive symptoms after controlling for family structure and parent education, but there were significant interactions of cohesion with race and gender. Higher CES-D scores were associated with lower cohesion scores, with partial R² values from 0.11 for African American female adolescents to 0.29 for White female adolescents. In longitudinal analysis, the Year 1 cohesion score explained only a small amount of the variance in Year 2 CES-D after controlling for Year 1 CES-D; however, the change in cohesion score from Year 1 to Year 2 was a significant predictor after controlling for baseline CES-D and cohesion scores. Neither family structure nor parent education was a significant predictor. Depressive symptoms in young adolescents may be more strongly affected by the perception of emotional bonding in the family than by the particular family structure. Apparent effects of family structure are likely to be confounded by and a reflection of the level of perceived emotional bonding or cohesion. The strength of these associations varies across race and gender groups, perhaps reflecting differences in external events, extended family structures, and stress.