Many states are focusing expensive prison beds on violent and career criminals with new policies that divert lower-level offenders into non-prison sanctions or reduce the time they spend locked up, yet most states cannot readily determine whether the new policies are working any better than those they replace, Adam Gelb and Craig Prins of the Pew Public Safety Performance Project write in the Washington Times. Beyond a simple count of prisoners, the typical state report offers basic demographic information and breaks down how many inmates are serving time for violent, property, drug and other crimes. These numbers reveal only fragments of the information necessary to paint a meaningful portrait of inmate populations. For instance, an offender serving time for a relatively minor crime may have a string of prior violent convictions that make him a higher risk to society than someone in prison for a more serious offense not likely to be repeated.
A more holistic look at prison use would blend current offense, prior record and risk of recidivism, say Gelb and Prins. By joining some combination of these elements into a single measure — a prison composition index — policymakers and the public could develop a better understanding of how their prison beds are being used and whether their reforms are succeeding. Pennsylvania may be the first state to use a sophisticated prison composition index. Under Secretary John Wetzel, the Pennsylvania Department of Corrections uses an “Offender Violence Risk Typology” tool, which merges information about current offense, prior record and risk level to create three categories of inmates. The index says 69 percent of state prison admissions and 59 percent of the total population in 2013 fell into the least serious of the three categories, figures that have changed little since 2010.