Rajendran, K., Smith, B. D., & Videka, L. (2015). Association of caregiver social support with the safety, permanency, and well-being of children in child welfare. Children and Youth Services Review, 48, 150-158. Doi: 10.1016/j.childyouth.2014.12.012.
Positive social support can have a buffering effect against stress, strengthen parenting styles, and in turn, lead to positive child outcomes (e.g., MacKenzie, Kotch, & Lee, 2011). This study examined the association between social support in caregivers’ social environment and maltreated children’s safety, permanency and child well-being.
The authors used data from Child Protective Service (CPS) cohort of the first National Survey of Child and Adolescent Well-being (NSCAW-I), which is a nationally representative, longitudinal survey of families who had a child maltreatment investigation. This study analyzed a sub-sample of 4,034 children (from birth to 15 years of age, mean age = 7.02) who were living at home at Time 1 (i.e., 2 to 6 months after the initial child welfare investigation) because the researchers believed that caregivers’ social support may not have had much influence over children who were out-of-home at Time 1. Caregivers’ perceptions of their satisfaction with social support, number of support providers, and caseworkers’ reports of inadequate social support were analyzed separately to predict child safety, permanency, and child behavioural problems, controlling for demographic information (e.g., age, gender and race of the child, and caregiver depressive symptoms, etc.).
The NSCAW-I has a longitudinal design: Time 1 data were collected 2 to 6 months after the initial child welfare investigation; Time 2 data were collected 36 months after the first investigation; Time 3 data were collected 59 to 97 months after the first investigation. Demographic information, including family income, caregiver depressive symptoms, child race, age, gender, and general health status, was collected at Time 1. Both social support measures from the caregiver and from the caseworker were collected at Time 1. Child safety was measured by whether the investigation was substantiated or not by Time 2. Child well-being was measured by internalizing and externalizing behavioural problems, assessed using Child Behavior Checklist at Time 2. Child permanency was measured by whether the child was placed in out-of-home care for an extended period of time (more than 5% of the time) between Time 1 and Time 3.
Results of this study revealed that different indicators of social support had a differential impact on child welfare outcomes. Caseworkers were more likely to report low levels of social support for caregivers whose investigations were later substantiated. When there was inadequate social support for the caregivers reported by caseworkers, children were almost twice as likely to be in out-of-home placement in comparison to the ones with adequate social support reported by caseworkers. Moreover, a greater number of support providers available was related to less child internalizing and externalizing behavioral problems. Caregiver satisfaction with social support was positively related to child permanency only. Similar to previous research, social support in general was associated with positive child outcomes. This study provides empirical evidence that not only formal but also informal support is an important resource for families in child welfare system. The findings in this study also showed a consistent association between caregiver depressive symptoms and all three types of child outcomes -- higher levels of caregivers’ depressive symptoms were related to less child safety and permanency, and more child behavioural problems. Even though caregiver depression was not the primary focus of the study, it certainly deserves some attention in future research to investigate whether caregiver depression could interact or mediate the relation between social support and child outcomes.
The strength of this study is that the use of two independent sources of reports on social support and a longitudinal approach could significantly reduce common variance bias. However, the authors analyzed each independent variable (i.e., caregiver support satisfaction, number of support providers and caseworker’s report on social support) separately for each outcome variable (12 regression models in total). Although the Bonferroni correction was applied, these analyses are not the most parsimonious, nor can they provide information regarding which sources of social support report have unique contributions to specific child outcomes while controlling for other sources of reports on social support. Future research should consider including both caregiver and caseworker reports of social support in the same model for analysis and examine how informal support could relate to positive child outcomes through parenting styles, and caregiver depressive symptoms.