Research That Matters (January 17 - 20, 2008)

Purpose: One of the strengths of the FFCWS is that the large number of White, African American, and Hispanic families is sufficient to allow for subgroup analyses. The second presentation examines race/ ethnic differences in the model linking community characteristics to parenting stress and mastery, and risk for child physical abuse and neglect. Such analyses are important for a variety of reasons. First, it is well documented that African American children are disproportionately involved with child protective services and the high rate of poverty experienced by minority families does not seem to fully account for this overrepresentation, thus pointing to the importance of examining a broader set of community characteristics such as collective efficacy and social disorganization. Second, it has been proposed that parenting practices may differ significantly as a function of race/ ethnic status. Minority parents may experience higher levels of parenting stress, shaped in part by their neighborhoods. Finally, there is relatively limited past research examining community characteristics and parenting behaviors of Hispanic families.

Methods: Predictor variables were mothers' perceptions of community disorganization and collective efficacy, maternal parenting stress, and mastery. Physical child abuse, psychological abuse, and neglect proxies were drawn from the CTS-PC. The final models controlled for maternal age, income, and martial status. Mplus was used to conduct multiple group analyses testing the structural model for equivalence across race/ ethnic group (White, African American, and Hispanic).

Results: Findings indicated that the model fit equally well for the White, Hispanic, and African American mothers. However, while the fit and misfit measures and chi-square indicated that overall lack of race/ ethnic invariance in the relationships linking collective efficacy, social disorganization, parenting stress and mastery to physical aggression, psychological aggression, and neglect, path coefficients pointed to specific ways that race/ ethnic status influenced mothers' abuse risk. For example, being an older mother was a more significant protective factor for African American mothers; relationship status (not being married) was negatively associated with higher levels of mastery and also with risk for physical child maltreatment, but only for White mothers.

Implications: This study is important given that few studies have adequate sample size to conduct subgroup analyses, particularly examining Hispanic mothers, comparing across a range of variables measuring CAN. Results suggest that once appropriate demographic variables are controlled for there are few overarching significant race/ ethnic differences in terms of how community factors shape individual parenting behaviors. We find little evidence that maternal mastery and perceived parenting stress differ significantly across race/ ethnic groups; or that race/ ethnic status differentially influences risk for maternal CAN. These findings are consistent with other recent studies using the FFCWS focusing on fathers that also point to few race/ ethnic differences in CAN risk except in terms of how race/ ethnicity interacts with demographic variables. In sum, a key implication of this research is that it contributes to a better understanding the relative influence of individual, community and socio-demographic characteristics.