Bridging Disciplinary Boundaries (January 11 - 14, 2007) |
Methods: The study is based mainly on a secondary analysis of a database, which focuses on the status of 4,900 children in care who reside in 66 institutions, their background characteristics and their interactions with their biological parents. Additionally, I collected data on supervisors' assessments on the facilities. I applied Hierarchical Linear Regression Modeling (HLM) to examine factors on multiple levels, children and facilities that explain variance in children's behavioral-emotional and academic condition. I also examined cross-level interactions (child by facility).
Results : It was found that both facility and child-level factors helped in explaining the condition of children in-care. Thus, institution climate (e.g., the interactions with staff and children's violence against peers), as well as size and other physical aspects of the institution are important predictors of the child's behavioral-emotional condition. Children's age, gender, length of stay in the institution and the quality and frequency of the children's encounters with their biological parents predict some of the child's condition. There were also significant statistical interactions between children's personal and institutional characteristics.
Implications for practice or policy: This study has theoretical implications for an ecological perspective on children in out-of-home care. It also suggests strongly that current methods focusing on either the child in care or the facility should be modified to multi-level approaches that examine 'children within settings'. From a practice and policy point of view, the study shows the importance of specific aspects of the residential facility climate (e.g., children's interactions with the staff). The study also emphasizes the contribution of support for the child-family contact while in care.