By Paul Hammel
LINCOLN — When state prison officials allowed a weekend furlough for a former gang member convicted of a felony gun crime, a scientifically devised risk assessment that is given to all prisoners helped drive that decision.
The inmate, Jermaine Lucas, was on furlough when he was shot and killed by Omaha police after he lunged for a gun in a 2012 standoff. The case led to questions about who should be furloughed.
They say that such surveys, when validated by research and supplemented with professional judgment and consideration of an inmate’s conduct in prison, can help focus limited corrections resources to the inmates who most need them.
They aren’t 100 percent reliable, as the Lucas case tragically illustrated, but such risk assessments play a key role in “data driven” corrections practices that are being used to determine who can serve out their sentences via less-costly parole or probation supervision and who needs intense rehabilitation and supervision before returning fully to society.
“Risk and needs assessment tools are an important component of offender management, both in facilities and in the community,” said new Nebraska Corrections Director Scott Frakes.
Frakes, who took the job a month ago, said he will be reviewing the department’s risk assessment tools in the near future. The Lucas case, back in 2013, already inspired a review and some changes in the furlough program.
An Iowa prison researcher who helped devise that state’s assessment tool said they’re a highly effective tool in identifying inmates who are at the highest risk of reoffending.
Since 92 percent of Iowa’s prison inmates will eventually return to society, it’s critical to determine who should get the most supervision and rehabilitation, said Lettie Prell, director of research for the Iowa Department of Corrections.
“(Assessments) just make us smarter on who to target for our programming,” Prell said.
The issue of risk assessments was thrown into the spotlight recently. An investigation by the Associated Press highlighted some high-profile failures associated with such assessments, including a Texas sex offender deemed “low risk” because of the inaccurate responses he provided. After he was released, he moved to Indiana and committed a string of rape-murders.
The AP’s report criticized some states for failing to double-check inmate responses to assessment questions and for keeping secret the findings of such risk surveys. It raised questions about the reliability of such assessments as more and more states, such as Nebraska, join the “data-driven” corrections movement.
But prison officials in Nebraska and Iowa, as well as a national authority on prison reform, defended the use of such assessments. Neither state uses an assessment contractor who drew criticism in the AP report.
Marc Pelka of the Council of State Governments Justice Center said such assessments play a key role in deciding which inmates need extensive supervision and rehabilitation programs and which ones don’t.
“The goal of the criminal justice system is to bring the risk of reoffending and harm as close to the zero as possible. These are tools to help do that,” Pelka said.
A study of Nebraska’s prison system, coordinated by CSG and involving state leaders, made several recommendations, including that the state’s parole system begin using a “validated actuarial risk and needs assessment” to help determine which parolees are the highest risks and need the most supervision. CSG concluded that the current parole assessment tool was inadequate.
Such assessment tools are similar to those used by insurance companies to set insurance rates — they predict the possibility of future outcomes, such as a car wreck, based on an analysis of past activities and current conditions, such as your past driving record and age.
For instance, the assessment used most often by the Nebraska prison and probation systems — the Level of Service/Case Management Inventory, or LS/CMI — asks questions about criminal history, education and employment background, drug and alcohol problems, and anti-social and criminal thinking.
Questions that focus on when an inmate committed his first crime and if he has a high school diploma or GED are predictors of future behavior, officials said.
Iowa’s assessment tool, coupled with a stronger focus on re-entry to society, has helped reduce the recidivism rate — the frequency of inmates committing repeat crimes and returning to prison within three years — from 35 percent in 2009 to 29.7 percent in 2014.
Prell and others emphasized that the assessments are effective only when paired with professional analysis of other factors and when considering things such as input from crime victims.
“These tools are just that — they’re tools,” she said. “They should never be used as the only deciding factor in any release.”
Such surveys should also be regularly “validated” to ensure they are accurately predicting future behavior, officials said.
Last year the Nebraska probation office contracted with psychology and legal researchers from the University of Nebraska-Lincoln to validate the LS/CMI assessment, which has been used in the state since 2006.
It concluded that the test, developed by three Canadian psychologists, predicted recidivism as well as any other measure in the country but that it might be improved with better training of those who administer the survey.
Deb Minardi of the State Office of Probation Administration said assessments are not only used in compiling presentence investigations, which help judges determine whether an offender goes to prison or is sentenced to probation, but also in deciding the intensity of probation supervision. That allows the office to make caseload decisions. The more intense supervision needed, the fewer cases given to a probation officer.
“It is the foundation on which we manage probation cases in the community,” Minardi said.