Formative Evaluation of a New Practice Model to Support Permanence for African American and American Indian Children in Foster Care

Background and Purpose: The California Department of Social Services (CDSS) received PII funding for the California Partners for Permanency (CAPP) project to develop an innovative casework practice model for reducing long-term foster care and improving permanency outcomes, particularly among African American and American Indian children who were disproportionately represented in the state’s child welfare system. This represented a recognition by CDSS of the role that institutional bias plays in child welfare and the importance of systemic barriers that particularly affect African American and American Indian children and keep bias entrenched, including poverty, racial bias and discrimination, system processes and resources, and geographical factors such as neighborhood segregation and jurisdiction-specific child welfare policies and processes.

Methods: Because the practice model was new and represented sweeping change in a large, diverse child welfare system, the evaluation team emphasized formative evaluation, calling first for trial use of administrative data to analyze casework events and permanency outcomes while collecting survey data on proximal outcomes from the biological parents or legal guardians of CAPP-served children. The lessons from the first formative evaluation led to a second formative evaluation, rather than proceeding to summative evaluation.  

The CAPP permanency analysis used state administrative data from AFCARS and NCANDS. These data were supplemented with data from the Child Welfare Services Case Management System and the project’s data system to identify CAPP-trained social workers' active cases. Survival analysis was used to examine the time to permanence for CAPP-served and matched children in each CAPP site.

Results: The CAPP practice model was associated with positive changes on permanency measures in two counties.  In Fresno, exit to permanency within 12 months was three percent higher among CAPP-served children than it was among historically matched children.  The difference was greater for African American CAPP-served children (6 percent higher) and Hispanic CAPP-served children (12 percent higher).  CAPP-served children in Fresno were also more likely to be placed with a permanent family within 12 months; placement was nine percent higher for all CAPP-served children combined, and 12 percent higher for Hispanic children.  In Pomona, placement with a permanent family within 12 months was three percent higher for CAPP-served children than it was for matched children; among Hispanic children, it was five percent higher.

Conclusions and Implications: Though CAPP has not yet reached full implementation and measured proximal outcomes essential to substantiating the theory of change, it has created a culturally sensitive practice model, trained hundreds of social workers, served thousands of children and families, established lasting community and tribal partnerships, and built internal organization supports.   The second formative evaluation suggests these accomplishments resulted in positive associations between the practice model and shorter foster care stays, especially in Fresno and Pomona Counties, which started early with strong coaching and supportive leadership. A key challenge going forward will be sustained attention to core model components, such as coaching and fidelity assessment, and further development of leadership-level organizational supports that address systemic barriers to consistent, enduring use of the practice model.