Our Projects

Proactive Child Welfare Predictive Risk Model

Allegheny County, PA

Developing a screening tool can potentially enable earlier intervention to prevent adverse outcomes. The research team will deliver a draft methodology paper to the County, so that the County can decide how it wishes to proceed.

Allegheny Family Screening Tool (AFST)

Allegheny County, Pennsylvania

Researchers have modelled, designed and supported implementation of this world-first child welfare predictive analytics tool.  The AFST uses rich administrative data to generate a screening score for each incoming call alleging child maltreatment. The score is an additional piece of information that helps call screeners as they decide whether to open an investigation.  Allegheny County hopes this decision support tool will improve the accuracy and consistency of call screening decisions.  The research team continues to collaborate with the County on active quality assurance and regular rebuilds, designed to improve the accuracy and usability of the tool.

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

California State

The goal of the project is to establish whether the statistical modelling of historical child protection records can be used to improve the initial screening and triaging of child abuse and neglect referrals.

Douglas County Decision Aid

Douglas County, CO

In early 2017 researchers developed a prototype child welfare (maltreatment) predictive risk model for Douglas County using data from child welfare and public welfare eligibility systems.  Douglas County leadership has decided to proceed with the infrastructure build stage of the model.  The DCDA will be implemented as a randomised controlled trial.

Homelessness Predictive Risk Model

Allegheny County, Pennsylvania

CSDA researchers are working with the Allegheny County Department of Human Services on the possible use of predictive risk modeling to address homelessness. The models under consideration and development include one that would allow early intervention when the risk of homelessness is high, and another that would allow the County to better prioritise requests for homelessness services. .

Community perspectives on the use of algorithms by Government in NZ and the US

Wellington, NZ & Allegheny County, Pennsylvania

A pilot project exploring whether it is possible to create guidelines for trusted use of algorithms informed by the community.  Researchers will systematically gather and analyse community concerns and views about the use of algorithms.  The study will rely on structured participatory design workshops with professionals and families who have experience of the child welfare systems in New Zealand and the US.

Protective Factors of Children and Families at risk of Adverse Childhood Experiences

Auckland, New Zealand

Exploring protective factors of children and families at highest risk of experiencing adverse childhood experiences (childhood experiences in the categories of abuse, neglect and household dysfunction that are associated with health and social problems in adulthood.) This study uses data collected for Growing up in New Zealand, a longitudinal study led by the University of Auckland.

MyDay Workplace Wellbeing Survey (Junior Doctors)

Wessex, England

Researchers will use MyDay, a purpose built web-based survey tool, to collect anonymous data from up to 2100 junior doctors about their wellbeing at work.  Participants will list tasks they did in five short windows of time, and how they felt doing each task.  The research sets out to provide a rich and useful picture of the experienced wellbeing of this key workforce.