Family Functioning and Adolescent Behavior Problems: A Moderated Mediation Model of Caregiver Depression and Neighborhood Collective Efficacy

Background:Adolescents’ behavioral problems arewidely known to be the result of interactions with their environments. However,the long-term influence of the ecological risks in early adolescence on individual’s behavior problems in middle adolescence is less clear.Applying Bronfenbrenner’s ecological systems theory, the present study employed a longitudinal approach to explore the effects of family functioning (FF), caregiver depression (CD), and neighborhood collective efficacy (NCE), on adolescent internalizing problems (AIP) and adolescent externalizing problems (AEP). Four hypotheses were tested: (H₁) CD mediates the relationship between FF and AIP; (H₂) CD mediates the relationship between FF and AEP; (H₃) the mediating effect of CD on the relationship between FF and AIP varies depending on levels ofNCE; (H₄) the mediating effect of CD on the relationship between FF and AEP varies depending on levels of neighborhood collective efficacy.

Methods:Data were drawn from the Longitudinal Study of Child Abuse and Neglect (LONGSCAN). A total of 450 participants for which complete data were available at ages 12, 14, and 16 were included in this study. FF was an independent variable assessed with 36 items (α = .85) using the self-report family inventory (Beavers, Hampson, & Hulgus, 1990). CD as a mediator was measured with the self-reported Center for Epidemiologic Studies depression scale (CES-D; Radloff, 1977; α = .88). NCE as a moderator was measured with 11 items developed by LONGSCAN investigators (α = .86). AIP and AEP were dependent variables measured with 113 items (α = .89 and α = .91, respectively) with the child behavior checklist/4–18 (Achenbach, 1991). Mediation and moderated mediation analyses were conducted using PROCESS Macro 3.4, and the regression bootstrapping method was applied to test the hypothesized models. In the analyses, the PROCESS macro generated 5,000 bootstrap samples with a 95% confidence interval (CI).

Results:The results partially supported our hypotheses. First, CD significantly mediated the relationship between FF and adolescents’ behavioral problems. Specifically, FF was significantly related to AEP and mediated by CD in the mediation model. However, FF was not significantly related to AIP, but it was significantly mediated by CD. Furthermore, in the two moderated mediation models, the indirect effect of FF on AIP through CD was significantly different based on levels of NCE, but it was not significantly different for AEP. Specifically, the indirect effect of FF on AIP through CD was greater for adolescents living in neighborhoods with lower levels of collective efficacy than those in neighborhoods with moderate and high levels of collective efficacy.

Implications:The present longitudinal data revealed that both lower levels of family functioning and higher levels of caregiver depressive symptoms were significant risk factors for both internalizing and externalizing behaviors. Specifically, the moderated mediation results highlighted the protective role of neighborhood collective efficacy in reducing the negative impact of negative family functioning and caregivers’ psychological problems on adolescents’ internalizing problems. This information about environmental risks and protective factors provides important insights that can be used to create early interventions against behavioral problems among adolescents.