The Effects of Kinship Care on Child Behavioral Problems Across Age Groups: A Propensity Score Analysis

Background and Aims: Placement with a relative after substantiation of maltreatment is often viewed as the least restrictive and most family-like out-of-home placement for children who must be removed from their biological parents due to abuse or neglect. Although kinship care has become a preferred option in the U.S. child welfare system, this preference has largely been based on “soft evidence” rather than rigorous evaluation of the risks and benefits of kinship care. Therefore, more evaluations of the impact of kinship care on child behavioral problems are needed to guide child welfare practice and policy. In addition, because children of different ages or developmental stages adjust differently to placement changes and maltreatment experiences, evaluations of kinship care should explore its effects on child behavioral problems across age groups.

Methods: The analysis used data from Waves 1 and 2 of the second National Survey of Child and Adolescent Well-Being (NSCAW II). The sample included 1054 children, ages 0 to 17.5 years old, who were living with the same caregivers at Waves 1 and 2. Of these included in the sample, 584 children had been placed in kinship care, and 470 children had been placed in non-kinship care. We examined the impact of kinship care on behavioral problems in the entire sample and then in 2 age groups: younger children (0 to 5 years), and older children (6 to 17.5 years).

The estimated treatment effects of kinship care were examined by combining both propensity score weights and NSCAW II survey weights in regression analyses in order to account for selection bias and ensure that results were generalizable to the original target population. Also, the child behavior in Wave 1 was treated as a predictor of child behavior in Wave 2 to examine whether behavior changed significantly between the two waves.

Results: Children in kinship care had average total behavior problem scores that were 5.1 points lower (p < 0.05), and externalizing behavior problem scores that were 4.77 points lower (p<0.05) than the scores of children in non-kinship care. Kinship care was not associated with internalizing behavior problems in this sample. For younger children, those in kinship care had lower total and externalizing behavior problem scores, but the effects were not statistically significant. In contrast, older children had average total behavior problem scores that were 8.47 points lower than children in non-kinship care (p < 0.01). Placement in kinship care also had significant negative impacts on older children’s externalizing (p < 0.5) and internalizing (p < 0.01) behavior problems. 

Conclusion and implication: Findings from this study suggest that children in kinship care, especially older children, exhibit lower levels of problematic behaviors compared with children in non-kinship care. These results support the protective role of kinship care for youth in the child welfare system. Findings also suggest that practitioners should work to ensure the stability of these placements by providing supports to increase the caregiving capacity of kinship caregivers.