Can Child Abuse Be Predicted?


Predictive analytics has fundamentally changed the way retailers operate — helping Netflix guess what movie a viewer might want next or Amazon market to a buyer’s needs and wants — but can it be used to protect minors from child abuse?

A new paper in the journal Child Abuse & Neglect examines the potential use of predictive analytics for child welfare agencies.

“Increases in data quantity and data diversity, along with advances in computing technologies, have made predictive analytics a powerful tool for helping to guide decision making,” researcher Jesse Russell writes in the paper.

Russell writes in the paper that predictive analytics are increasingly used in the public sector, but he warns that child welfare agencies should use caution when embracing the new technology.

“The applications of predictive analytics are not academic endeavors; they often involve private, for-profit partners, and trade secrets may apply,” Russell writes. “Based on publicly available information, it is difficult to assess how valid the models are, how often they get it right, how trade-offs between false positives and false negatives were made, and how different approaches compare to each other in terms of accuracy.”

Read the paper HERE.

Comments are closed.