Diversion programs offer alternatives to the initial or continued processing of youth within the juvenile delinquency system with the purpose of redirecting youthful offenders through programming, supervision, and supports. While it may be ideal to prevent problematic behavior before a youth demonstrates delinquent behavior leading to involvement with the juvenile justice system, it is often difficult to determine which diversion program has demonstrated the most success with the unique population of youth teetering between typical adolescent impulsivity and true delinquency. So let’s look at three ways predictive analytics can assist in selecting, developing, adopting, or choosing an array of diversion programs.
Collecting data on whether or not a youth is successful (i.e. no subsequent arrests, no additional criminal complaints filed) after completing a diversion program will help determine what characteristics of the youth and the program led to a positive (or negative) outcome. Predictive analytics has the ability to determine which youth are most appropriate for different types of diversion programming, leading to a more individualized approach.
Diversion programs that are ineffective will be identified:
Predictive analytics can assist in determining if a diversion program is not producing the desired outcomes (no re-arrests, no further involvement in the system) by sifting through historical data and outcomes of the program. This idea is more complicated than simply tracking “successful completions;” it is using machine learning to predict the future in terms of rates of success among various youth and interventions. If the youth characteristics or the diversion program characteristics shift, stakeholders will be informed and a decision can be about the efficacy of the diversion program.
Predictive analytics can identify protective factors (or the lack thereof) to be addressed through diversion programming or other community support services:
Though protective factors can significantly decrease a youth’s likelihood of continued misbehavior, they are often underutilized in many “risk” assessments which focus mostly, of course, on “risk.” Uniquely, predictive analytics can filter through both protective factors and risk factors simultaneously. Diversionary programs or other community support services will likely see encouraging outcomes when protective factors are identified and strengthened through their interventions.
There is a fine line between diverting youth away from a risky lifestyle and catapulting them into a system with labels such as “delinquent” and “criminal.” A predictive analytics tool can assist various agencies in finding the best match between a youth and a diversion program. Furthermore, it has the ability to predict the likelihood a youth would complete the intervention and have no further involvement with the justice system — an investment that benefits the youth, their family, the community and society as a whole.