Research That Matters (January 17 - 20, 2008)


Empire Ballroom (Omni Shoreham)

How Neighborhoods Influence Risk for Physical Child Maltreatment: Path Analyses from a National Study

Neil B. Guterman, PhD, University of Chicago.

Purpose: Qualities of neighborhoods have long been thought to play a central role in shaping the occurrence of physical CAN, and views of their importance have substantially influenced national and local policies and practices. However, recent research employing multi-leveled analytic strategies has pointed out the important role of individual-level factors when considered in conjunction with neighborhood factors in predicting physical CAN. Such findings have raised questions about the precise role of neighborhoods and their interplay with individual and family-level influences on the occurrence of physical CAN. We examine this interplay, drawing from the FFCW study, a national population-based study.

Methods: Predictor variables included measures of community disorganization, collective efficacy and community violence, as well as maternal mediator variables such as parental stress and mastery. Physical CAN proxies were drawn from mothers' self-reports on the CTS-PC. Path diagrams were developed to trace direct and indirect pathways from community factors to maltreatment risk. The final model linked neighborhood characteristics and parenting characteristics to CTS-PC variables measuring physical aggression, psychological aggression, and neglect, as well as observed HOME scale outcome measures. The model also controlled for maternal age, race, family income, and marital status.

Results: The model demonstrated very good fit across all outcome variables. All RMSEA estimates were equal to or less than .04; all goodness of fit and misfit measures were between .97 - .99. SEM models indicated a significant but mild direct pathway from community social disorganization to physical child abuse risk, but not to neglect risk. There were no significant direct pathways from community collective efficacy to either physical abuse or neglect risk. A key finding is that an indirect effect from community social disorganization to both physical abuse and neglect risk was identified via mothers' sense of mastery and stress, as mediators. Less robustly, community collective efficacy showed an indirect pathway to physical abuse risk through maternal mastery, and no indirect pathway to maternal neglect.

Implications: These findings are consistent with the most recent empirical studies challenging previous notions regarding the influence of neighborhood characteristics on parenting behaviors. We find only a mild direct role for community factors in risk for physical CAN. However, for the first time, these findings document that community factors likely play a relatively stronger indirect role in their influence on parenting behaviors, specifically, shaping risk for physical CAN by challenging mothers' sense of control and stress in the parental role. This research highlights potential intervention strategies, at both the individual and neighborhood level, that might interrupt community-level influences in risk for maltreatment.