Predictive Analytics Enlisted to Fight Child Maltreatment

Researchers say the technology could pinpoint Fort Worth neighborhoods — down to the block — where abuse may be most likely.

Research being conducted in Fort Worth is using Big Data techniques to help protect children from maltreatment.

Cook Children’s Center for Prevention of Child Maltreatment in Fort Worth and researchers at Texas Christian University are using ‘big data’ technology to predict neighborhoods where children are most at risk — and early results show the technique could be a promising tool to  fight the problem that often goes unreported.

“Data suggests that the underreporting rates of crimes against children is exceedingly high,” Dyann Daley, executive director of Cook Children’s Center for Prevention of Child Maltreatment said in a TCU release. “For example, the majority of child maltreatment fatalities occur in families not known to the child welfare system.”

“Data suggests that the underreporting rates of crimes against children is exceedingly high.”
Dyann Daley

How big is the problem? In 2014, more than 6,000 cases of child abuse or neglect were reported in Tarrant County. That was more than any other county in Texas, according the release.

Researchers applied known analytical methods to attack maltreatment of children.

“Predictive analytics has been used successfully to predict shootings and other violent crime,” Michael Bachmann, associate professor of criminal justice at TCU, said in the release. “We applied it for the first time to child mistreatment.”

According to the release, researchers collected data from over a dozen known risk factors for child maltreatment, including poverty and violence to alcohol sales. 

CHILD MALTREATMENT ANALYSIS PROVES ACCURATE

Then, researchers used “Risk Terrain Modeling” to predict likely areas of child abuse in Fort Worth, the release said.

“When we compared our 2013 model with the actual cases of child maltreatment in 2014, we found we correctly identified 98 percent of all substantiated future cases,” Bachmann said. ”Only 2 percent of future instances were outside our predicted areas.”

“Predictive analytics has been used successfully to predict shootings and other violent crime.”
Michael Bachmann

The accuracy of the model will ease the alignment of prevention and treatment resources with the neighborhoods that are most in need — down to the city block, the release said. 

The research team said it is expanding the study to evaluate other Texas cities, including Dallas, Houston, Austin, and El Paso.

The study, titled Risk Terrain Modeling Predicts Child Maltreatment, will be published in the December issue of the journal Child Abuse & Neglect, the release said.

More Reading:

ZipRisk Map Uses Data to Fight Child Abuse at ZIP Code Level


Delivering what’s new and next in Dallas-Fort Worth innovation, every day. Get the Dallas Innovates e-newsletter.

Add Comment