May 5, 2009

NCIC critiques actuarial risk tools

The promise of violence risk prediction in corrections has "trumped actual performance," warns a report from the National Council on Crime and Delinquency.

Indeed, in its pell-mell rush to implement defensible, "evidence-based practice," the criminal justice field has abandoned clarity and parsimony in favor of a confusing hodge-podge of practices that lack proven reliability and validity, asserts the report, A Question of Evidence.

The report is authored by Christopher Blair, executive vice president of the NCCD, which is the oldest criminal justice research organization in the United States and a pioneer in evidence-based classification schemes in child protection and foster care.

The report critiques the sloppy use of buzzwords such as "criminogenic needs" and "protective factors." "These are important concepts, but ones that require a significantly deeper level of assessment than many risk models currently provide. As such, they can raise false expectations and lead to inappropriate case plans and services."

The NCIC is advocating that juvenile and adult corrections administrators step back and take a critical look at the actuarial tools, lest flawed instruments, approaches, and terminologies become so entrenched that they are impossible to change.

The report is available HERE.

No comments: