Delinquency in Child Welfare: Investigating Kinship Care Effects

Introduction: Kinship care placements are common practice in child welfare systems. Approximately 200,000 (40%) of the children in substitute care settings are in kin placements (Mayfield, Pennucci, & Lyon, 2002). The literature on kin placements is mixed. Children in kin homes experience more stable placements, are more likely to maintain ties with their biological family, and are less likely to re-enter substitute care placement (Testa, 2001; Courtney, 1995; Berrick, Needle, Barth, & Jonson-Reid, 1998). Yet, kin providers are more likely to be poor, single, older, and have fewer years of education as compared with non-kin providers. Kin providers also have less contact with caseworkers, and receive fewer support services (Barth, Courtney, Berrick, & Albert). Overall, children in kin placements face significantly more environmental hardships than children in non-kin homes (Ehrle & Green, 2002). We hypothesize that the exposure to such hardships increases the likelihood of delinquency for adolescents in kin placements.

Methods and Results: The initial sample includes 11,857 youth between 7 and 16 years of age in Los Angeles County. The sample is limited to adolescents with only kin or non kin placements. We use propensity score matching (race, age, gender, maltreatment type) to help address selection effects. The matched sample includes 4,274 youth in kin placements, and 4,274 youth in non kin placements. The matched sample is 31% African American, 49% Hispanic, 17% white and 3% Asian. Forty-four percent of the sample is male. Of the 8,548 adolescents, 585 (7%) are associated with at least one arrest subsequent to placement. Survival analyses indicate that controlling for a range of covariates, adolescents in kin placements are significantly more likely to experience at least one arrest, but the kinship care effect is limited to males (Expâ = .78, p <.05). There is no kinship care effect associated with females. A closer examination of the kinship care and gender interaction reveals that much of the male effect is associated with African American males, and to a lesser extent, white males. For African American males, 16% in kinship care placements experienced at least one arrest compared with 10% in non kinship care homes (X2 = 8.76, p< .01). For white males, 10% in kinship care placements experienced at least one arrest compared with 6% in non kinship care homes (X2 = 2.26, p =.09).

Conclusions: There is evidence to suggest that delinquency rates in child welfare vary by placement type. Yet to date there exist no studies focused specifically on kinship care. The findings from the current study indicate that kin placements are associated with an increased likelihood of delinquency, but only for African American males, and to a lesser extent, white males. These findings raise important questions about the potential limitations of kinship care placements. These findings also indicate a critical need to understand why the risk of delinquency increases for some males in kin placements. Finally, these findings may offer some insight into how the child welfare system contributes to the overrepresentation of African American males in the juvenile justice.