Florida’s Department of Juvenile Justice is diving into the field of predictive analytics – a tool that big business has come to rely on to forecast what customers will want next – as a way to predict recidivism. But it’s more like splashing in the shallow end of the pool than the deep dive heralded by Chicago-based software company SPSS.
In a release last month, the firm said Florida would use its predictive analysis software (PASW) “to reduce recidivism by determining which juveniles are likely to reoffend. Identified at-risk youth can then be placed in programs specific to the best course of treatment to ensure offenders do not re-enter the juvenile justice system.”
Rather than abandoning its own risk assessment instrument, called Positive Achievement Change Tool Assessment (PACT), which is used to assess youths at intake and monitor their progress while they are involved with the department, DJJ will use the software from SPSS to augment it.
“It’s an interesting and cool endeavor,’ said Mark Greenwald, DJJ’s chief of research and planning.
Greenwald’s predecessor, Bob Dale, first expressed interest in predictive analytics in 2008. The department actually bought the software in 2009, but Greenwald’s staff has just finished training by SPSS on its use. In the meantime, SPSS was bought by IBM for $1.2 billion in 2009.
The SPSS predictive analytics software cost Florida $12,000, which Greenwald said pays for its use indefinitely. An annual maintenance fee buys access to SPSS product support and updates to the software. (Florida isn’t currently paying for the maintenance because of budget issues.)
Greenwald’s research and planning staff will take data from the PACT risk assessments, such as current and past offenses and diagnosed mental health needs, and enter that into the SPSS software. Other data, not collected during risk assessment, will be included, including information about specific treatments and services a youth received while in DJJ custody and how frequently he received them.
DJJ has two central objectives: to predict shifts in juvenile delinquency and to determine if the right programs were used for juveniles in the department’s custody.
Predicting any changes in the nature of juvenile delinquency, Greenwald said, can help DJJ and the governor’s office determine the use of resources, which is, of course, a major issue in fiscally challenged Florida. For example, if the input data predict there will be more female offenders in the coming year, the state could put more resources toward programs that serve girls.
The other objective is more evaluative than predictive.
The SPSS software will allow the department to determine whether youths sent to one type of program did better than “the same types of kids” sent to another.
The software programs will allow the department to dig down further than a placement or program to determine the outcomes of treatment. A hypothetical question: How predictive of non-recidivism is multi-systemic therapy (regardless of which program conducts it) for juveniles who are found delinquent of violent offenses?
The department could start using the software to forecast delinquency trends soon, Greenwald said, but using it to do macro-analysis of placement decisions is a few years off. After a few years of data have been entered, his department will start conducting evaluations on a “semi-annual or annual” basis.
“There is nothing wrong with our screening process,” he said. “We’re looking at the aggregate level to see if we can tweak some decisions.”
At $12,000 for the software, plus the man hours needed to do analysis, Greenwald said, “you don’t have to redirect many kids before you see a return on investment.”