Discrimination in Multi-Phase Systems: Evidence from Child Protection.

Area of Science:

• Social Sciences
• Public Policy
• Child Welfare Studies

Background:

• Child protective services involve multiple decision-making phases.
• Racial disparities in foster care placement are a significant concern.
• Understanding the propagation of discrimination is crucial for equitable systems.

Purpose of the Study:

• To develop empirical tools for analyzing discrimination in multi-phase systems.
• To investigate racial disparities in foster care placement decisions.
• To identify the sources and extent of racial bias in child protective services.

Main Methods:

• Utilized quasi-random assignment of hotline screeners and investigators.
• Analyzed a large dataset of over 200,000 maltreatment allegations.
• Employed empirical tools to quantify racial disparities in placement decisions.

Main Results:

• Black children are 55% more likely to be placed in foster care than white children with similar maltreatment risk.
• Call screeners contribute up to 19% of the unwarranted racial disparity.
• Investigators account for the majority of the remaining unwarranted disparity.

Conclusions:

• Racial disparities in foster care placement are evident and propagate through the system.
• Both initial screening and investigative phases contribute to unwarranted racial bias.
• Potential under-placement of white children in high-risk situations warrants further investigation.