Assessing Children’s Risk Using Administrative Records: A Proof of Concept Predictive Risk Modeling (PRM) Project

Summary

The increased availability and quality of administrative data during the last several decades have led to growing interest in tools and statistical models that can be deployed in real time to predict future events. Predictive risk modeling (PRM) is one such class of tools. PRM is used to automatically generate a risk score for each individual in a given data system, providing a efficient means of screening populations without requiring any additional data entry.

The goal of the project is to establish whether the statistical modeling of historical child protection records can be used to improve the initial screening and triaging of child abuse and neglect referrals. Although this project will not result in a tool without future technological investments, it will lead to the development of data that can inform (in an open and transparent fashion) the opportunities and limitations of PRM as an additional decision aid for child welfare workers in California.

Overview of Findings

Analysis in Progress.

A Community Workgroup on Risk Assessment comprised of agency and community partners has been established to discuss key ethical, legal, implementation and transparency issues when using PRM to improve referral screening decisions.

Opportunities for Participation

To submit items for discussion by the Community Workgroup on Risk Assessment, or to pose other questions you would like answered by the research team, click here.

Products

Media

Investigators
  • FirstName LastName, PhD
  • FirstName LastName, PhD
  • FirstName LastName, PhD
Funders
Data and Research Partners
Timeline
  • August 2016 – June 2018
1150 South Olive Street, Suite 1400
Los Angeles, CA 90015
General Questions
Connect
Newsletter
Sign up for our mailing list to stay up to date with the Children's Data Network's latest research and news developments.

©2020 Datanetwork.org

#