Via a public–private philanthropic partnership, this project will provide a rigorous and transparent exploration of the feasibility and potential value of predictive risk modeling by determining whether existing administrative data can improve hotline screening decisions for children and families involved with the child welfare system due to allegations of abuse and neglect.
The objective of this project is to develop a proof of concept for how existing state-level child welfare data might be used to support improved decision making and more informed intervention and prevention strategies for children and families who come into contact with the children welfare system due to allegations of abuse and neglect. Although this project will not result in a tool that can be implemented without further technological investments and development, it will showcase whether administrative data available at the moment a referral is made to a child welfare hotline can be used to improve initial screening decisions. This project will rely on existing data from the California Child Welfare Services/Case Management System, records on near-fatality and fatality incidents from the California Department of Social Services, and other relevant state-level administrative data. The final product will include an annotated list of variables, source tables, predictive weights, and software code for predictive risk models that rely on available data to predict future adverse outcomes such as a near-fatality, chronic child welfare involvement, and foster care placement. In addition, a Community Workgroup on Risk Modeling will be established to provide a forum to discuss ethical and legal concerns, initial findings, and implementation considerations.
California Department of Social Services ($200,000) and Laura and John Arnold Foundation ($100,000)
John Prindle, Coinvestigator; Postdoctoral Scholar, Children's Data Network
Regan Foust, Coinvestigator; Director of Strategic Partnerships, Children's Data Network
Rhema Vaithianathan, Coinvestigator; Professor, School of Economics, Auckland University of Technology; Senior Research Fellow, School of Economics, Singapore Management University
Michael Mitchell, Coinvestigator; Senior Statistician, Veterans Administration; Statistician, Children's Data Network
Daniel Webster, Coinvestigator; Director, UC Berkeley California Child Welfare Indicators Project