Developmental Dynamics of Neighborhood, School, Family, and Internalizing Symptoms among Economically Disadvantaged Adolescents

• In Event: Poster Session 11
  In Poster Session: PS 11 Topic - Developmental Psychopathology

Abstract

Background. Although many researchers have examined the associations between contextual factors and adolescents’ internalizing symptoms (Edwards et al., 2011; Mrug & Windle, 2010), few articles have addressed the longitudinal dynamics between ecological microsystems (e.g., neighborhood, school, family) and adolescent internalizing symptoms in the context of high poverty. Therefore, we examined the longitudinal effects of neighborhood disadvantage, teacher relationship, and family cohesion on internalizing symptoms among economically disadvantaged adolescents.

Method. Participants were from the Taiwan Database of Children and Youth in Poverty (TDCYP), a panel study surveying Taiwanese children and adolescents financially supported by the Taiwan Fund for Children and Families. The investigation started in 2009, and conducted biennial follow-up surveys, resulting in three waves of data from 2009 to 2013. A total of 1,404 economically disadvantaged adolescents at W1 were enrolled (aged 12 to 14), and adolescents completed self-reports of neighborhood disadvantage, family cohesion, teacher relationship, and internalizing symptoms at each wave. We first applied latent growth modeling to examine developmental changes in each variable in separate univariate latent growth models, and then explored how the baseline levels (intercept) and slopes of neighborhood disadvantage, teacher relationship, and family cohesion were associated with those of internalizing symptoms in a multivariate latent growth model.

Results. As shown in Table 1, results of fit indices among four univariate latent growth models suggested good or acceptable fit to the data. The significant means of slopes among predictors indicated that adolescents perceived deteriorations in neighborhood quality (b = .06, p < .001) and teacher relationship (b = -.02, p < .05), but improvements in family cohesion (b = .03, p < .05) over time. Figure 1 presents the multivariate latent growth model (χ2 = 88.87, CFI = .98, RMSEA = .029). At baseline, greater family cohesion was associated with lower internalizing symptoms (β = -.31, p < .001), while greater neighborhood disadvantages was associated with higher internalizing symptoms (β = .44, p < .001). In addition, higher family cohesion slope was associated with lower internalizing symptom slope (β = -.45, p < .001).

Conclusions. The findings suggest that the microsystems varied in distinctive ways but were interrelated. The neighborhood quality and teacher relationship deteriorated over six years but adolescents perceived increases in the family cohesion. At baseline, higher levels of neighborhood disadvantages were associated with higher levels of internalizing symptoms, while higher levels of family cohesion were associated with lower levels of internalizing symptoms. Only changes in family cohesion had significant associations with changes in internalizing symptoms. Surprisingly, teacher relationship had no influence on internalizing symptoms. The findings show that family cohesion is a key asset that can prevent the worsening of internalizing symptoms among adolescents in poverty, while neighborhood disadvantage has a negative impact on depression symptoms in early adolescence.

Authors

• Shou-Chun Chiang, Pennsylvania State University, University Park
  Presenting Author

• Sunhye Bai, Pennsylvania State University, University Park
  Non-Presenting Author