Poster #55 - Neighborhood Cohesion as a Moderator between Familial Stress and Externalizing Behavior in Infancy

• In Event: Poster Session 17

Abstract

Household stress results in elevated externalizing behaviors in children through mechanisms such as depleted caregiver mental resources which, in turn, lead to harsh, inconsistent parenting and a negative emotional climate in the home (Neppl, 2015). Social support is associated with less parenting stress and positive childhood outcomes (McConnell, 2011). We examined one potential moderator of the relationship between parental stress and externalizing behavior —neighborhood cohesion—which captures the network of relationships and values among neighbors. Importantly, we explored this potential moderating effect in infancy. We hypothesized that with higher household stress (as reported by the caregiver), infants would display greater levels of externalizing behavior at 24-months. We expected to find that high neighborhood cohesion would minimize the effects of household stress on externalizing behavior.
Data were collected as part of a multi-site, longitudinal study of attention and temperament. Participants included 357 caregiver-infant dyads for which data were collected at 8-, 12-, 18-, and 24-months infant age. Neighborhood cohesion was measured using a subscale on the Inter-university Consortium for Political and Social Research (ICPSR) community survey at 8-months, 12-months, and 18-months. Household stress was measured at 8-months, 12-months, and 18-months using the Parenting Daily Hassles (PDHSR) questionnaire. Finally, externalizing behaviors were measured using a subscale on the Child Behavior Checklist (CBCL) at 18-months and 24-months.
Moderation analysis was used to test if neighborhood cohesion buffers the effects of family stress on early externalizing symptoms. Externalizing symptoms at 18-months were entered into the model as a covariate to control for initial levels in behavior problems. Infant sex, study site, child birthweight, maternal education, and prenatal drug use were also entered into the model as covariates. Missing data were handled using multiple imputation.
Initial correlations revealed a positive relation between concurrent family hassles and externalizing problems at 18 (r=0.37, p<.001) and 24 months (r=0.32, p<.001). However, the moderation analysis indicated a significant negative main effect of average family hassles across 8-to-18 months on externalizing symptoms at 24-months while holding constant externalizing symptoms at 18-months (β=-0.47, p=0.03). In addition, the moderation model indicated a significant negative main effect of neighborhood cohesion averaged across 8-to-18 months on externalizing symptoms at 24-months (β=-6.20, p=0.04). Lastly, there was a significant positive association between average neighborhood cohesion and average family hassles on 24-month externalizing problems with each unit increase in average neighborhood cohesion associated with a 0.14 increase in the relation between average family hassles and externalizing problems (Figure, p=.03).
These results are not in line with the hypotheses and require further analysis. As seen in the figure, at one standard deviation above the mean on neighborhood cohesion, the relation between average family hassles and externalizing problems becomes positive, in line with prior literature. This could suggest that neighborhoods more ‘at-risk’ may swamp relatively nuanced relations between family factors and outcomes that are more evident in better resourced contexts. This analysis also highlights how controlling for initial levels may reveal imbedded relationships and changes in directionality given that concurrent relations were in line with initial hypotheses.

Authors

• Morgan Alayna Gilmer, Pennsylvania State University, University Park
  Presenting Author

• Marisa N Lytle, Pennsylvania State University, University Park

• Kristin A. Buss, Pennsylvania State University, University Park

• Vanessa LoBue, Rutgers University

• Koraly Elisa Perez-Edgar, Pennsylvania State University, University Park