Predictors Child Neglect Among Low-Income Families with Young Children

Background and Purpose: Extensive research exists on the risk factors associated with child neglect. However, most child maltreatment research is inadequately designed for identifying true predictors of neglect because sampling strategies do not afford a precise understanding of the timing of predictors vis-à-vis maltreatment episodes. Appropriate studies require prospective designs, wherein potential risk and protective factors are measured prior to neglect episodes. The present analysis relies upon data from three separate longitudinal studies of low-income families with young children. The research questions addressed include (1) what types of factors predict involvement with child protection systems (CPS) for reasons of neglect? (2) Do similar factors predict a validated self-report measure of neglect? And, (3) to what extent are there consistencies across these studies in the predictors of both neglect outcomes?

Methods: All three studies (Fragile Families and Child Wellbeing [FFCW]; Healthy Families New York [HFNY]; Illinois Families Study-Child Wellbeing [IFS]) involve probabilistic samples or subsamples of low-income families with young children. FFCW and HFNY involve birth cohorts, and the IFS-CWB tracks a subsample of low-income families with children under five years of age. None of the samples were selected on the basis of past or current child maltreatment or CPS intervention, and all were followed for multiple years using several measures of child neglect. Furthermore, these three studies share a host of similar measures for predicting child neglect, including demographic characteristics, economic indicators, markers of parent wellbeing, and parenting characteristics. Multivariate logistic regressions predicting CPS neglect reports and a dichotomous indicator of neglect from the Parent-Child Conflict Tactics Scale (CTS-PC) were conducted separately for each study, using common sets of predictors derived from baseline or initial survey waves. Various robustness checks were conducted to determine if additional measures unique to each study significantly altered associations observed between neglect outcomes and the common set of predictors employed across studies.

Results: Across the three studies, consistencies emerged with respect to the predictors of CPS neglect reports. Number of children, employment status, welfare receipt, reliance on charitable assistance, housing hardships, self efficacy, and parental depression predicted investigated CPS reports of neglect in all of the studies. Additionally, marital/cohabitation status, maternal education level, reliance on family and friends for economic assistance, domestic violence, substance abuse, and maternal physical health were predictive of CPS neglect in at least two of the three studies. There was less consistency across studies for the CTS-PC measure of neglect, but within studies, predictors of the two measures of neglect trended in similar directions.

Conclusions and Implications: Understanding the predictors of child neglect is of critical importance to the development of child maltreatment prevention strategies, since a clearer understanding of the risk and protective factors associated with neglect would enable more effectively targeted and tailored interventions.

More research is needed to further understand the nature of the associations between predictors of neglect and neglect outcomes, such as whether certain predictors reflect family problems and strengths, or whether they may reflect heightened surveillance of some families involved with various service systems.