By Evelyn Sanchez Hernandez
The media portrait of Chicago as a dangerous place is sadly accurate for the South and West areas of the city.  These areas report the highest concentration of gang violence and criminal activities. Residents in these areas are thus more susceptible to falling prey to this vicious cycle of violence as victims or perpetrators. The life path of these individuals is further impacted once they become part of the criminal justice system. Incarceration decreases productivity, limits employment opportunities, and in many cases, produces trauma. 
Due to these harmful results, Cook County, where Chicago is located, has diversified actions to reduce the number of people behind bars in recent years. The strategies used to reach this goal have included setting affordable bail amounts for nonviolent felonies, reducing mandatory probation periods, helping people attend court appointments, and promoting alternatives to arrests for residents facing a mental health crisis. These strategies have proven to be successful: the number of inmates in Cook County jail has dropped to around 6,000 in 2018 from a high of 10,000 inmates in 2007. 
The Justice Advisory Council (JAC), under the Office of the President of Cook County, manages a traditional grant program that has invested more than 14 million dollars from 2013 to 2018 to complement the county’s goal to reduce the population behind bars. These grants allocate resources to organizations in the county that offer services to “at risk” populations, in order to prevent violence, reduce recidivism, and conduct activities of restorative justice such as peace circles and mediation services.
As a fellow assigned to JAC under the University of Chicago Harris School of Public Policy’s Applied Data and Governance Program, one of my first tasks was to help JAC refine the definition of “at-risk”. One of the requirements to access JAC grants is that services should be directed towards “at-risk” populations. However, there was no clear standard to ensure grantees followed this directive. JAC was concerned that this definitional ambiguity had resulted in grants accruing to organizations focused on serving economically deprived populations without considering other pertinent risk factors, such as mental health or academic performance. Additionally, without a standard of risk, it was difficult to define which groups needed the most support and prioritize accordingly. For instance, one of the questions that came up in discussion was: were gang members more at-risk than students who have to walk every day into the gangs’ territories as part of their schedule?
To tackle this challenge, I followed two strategies. First, I conducted an exercise to understand the status quo in which I interviewed Cook County grantees to tease out their understanding of “at risk” populations. I interviewed 7 organizations that currently receive support from JAC. From this activity, I learnt that the perception of risk factors was similar among all groups, and almost all estimated that people in their programs were at high risk or very high risk. The service providers generally pointed to the same risk factors: previous criminal justice involvement, gang membership, and a violent community environment. While the organizations agreed on risk factors, they differed in how they interact with the populations at-risk, as well as their recruitment strategies and methods of risk assessment.
Recruitment strategies were diverse. Some organizations have agreements with the Probation Department to enroll individuals already identified as “high-risk”, while others maintained school networks that refer students under adverse situations. In addition, some have an open strategy that allows participants to voluntarily register their programs or participate based on their own needs and interests.
Looking at the different types of recruitment strategies was useful to understand the population profile engaged by the organizations. Open registration could create “self-selection” of low-risk participants, since individuals that most need intervention may be less likely to enroll voluntarily. Additionally, referral networks are helpful to target participants “at-risk”, especially when the organization has limited capacity to identify risk factors among their potential participants.
Risk assessment protocols differed as well, impacted most significantly by the scale of the organization. Larger and more experienced organizations use certified assessment and screening instruments to detect risk factors, mental health and substance abuse for their participants. Smaller organizations rely on self-report data or intake questionnaires to detect risk factors.
This difference was relevant because certified risk assessments instruments are more effective to identify special needs and degree of risk (low-high), in order to provide specific services to the participants. Intakes questionnaires and self-reported data can provide useful information too, but they are not entirely reliable since the participants can omit information necessary to identify risk. Additionally, some risk factors, as mental health problems, are not often easy to detect through general questionnaires. 
My second strategy was to look towards research and academia to set a baseline definition. One remarkable study conducted by Chapin Hall in collaboration with the Cook County Probation Department and Get In Chicago, a nonprofit focused on youth violence prevention, captured my attention. The study created a predictive profile of youth between 18 to 30 years old, in certain areas of Chicago, that are most likely to be involved in criminal activities in a one-year period. To create this profile, the study analyzed the criminal backgrounds of youth in probation and their families, and examined their school and social service records.  According to the results, the main predictive factors of recidivism for this population were repeated arrest records (more than 4 arrests), gang involvement, chronic school absenteeism and a history of mental health illnesses. 
Through interviews with Nick Mader, a researcher from Chapin Hall, and former and current staff member of Get In Chicago, I learned that regardless of the wonderful insights this study offered, it was not a definitive step to understanding the type of population that needs to be prioritized under JAC grants. I was also warned that due to the specificity of the study, this risk profile might differ for a different population even in other neighborhoods of Cook County.
Another main insight from the interviews with the researchers is that even if we could identify the "perfect" profile of high at-risk population, this wouldn't be helpful if the service providers do not have the capacities to attract this population, or cannot detect risk factors among their potential participants. For this reason, I determined that the main challenge was to establish grant funding conditions that attracted organizations effectively targeting correct “at-risk” populations, without excluding organizations that do not have a sophisticated enrollment or assessment process as of today, in the hope that funding would enable them to develop these capacities.
I recently presented the results from my research and interviews and recommended adding sections to the grant’s Request for Proposals to require organizations to describe in detail their target population, risk factors of the target population, recruitment strategies, and assessment risk capacities. I believe that these modifications will help JAC to not only to have better criteria for selecting the most effective programs, but will also be useful to pinpoint areas where service providers need more support to reach their objective of helping the Cook County’s vulnerable populations break out from generational cycles of violence
 In 2016, 50 percent of the shootings reported to the Chicago Police Department concentrated in neighborhoods in the South and West areas of Chicago. https://www.ipr.northwestern.edu/about/news/2018/crime-in-chicago-research.html
 According to the Pew Charitable Trust, in U.S. “serving time reduces hourly wages for men by approximately 11 percent, annual employment by 9 weeks and annual earnings by 40 percent”. The Pew Charitable Trusts, 2010. Collateral Costs: Incarceration’s Effect on Economic Mobility. Washington, DC: The Pew Charitable Trusts. For the electronic version, click here
 For information on the jail population in 2010, please check Olson, D.E., Tahier, S. (2012). Population Dynamics and the Characteristics of Inmates in Cook County. Chicago, Illinois, Cook County Sheriff's Reentry Council. For jail population from 2017 to present check Sheriff’s Office website, www.cookcountysheriff.org.
 There is a wide diversity of risk assessment tools. For instance, by 2014 more than 20 were used in the Juvenile System across the U.S., according to the Office of Juvenile Justice and Delinquency Prevention. Some examples of risk/needs are: 1) Youth Assessment and Screening Instruments (YASI): measures an individual youth’s level of risk for delinquent conduct, key areas of programming or service need and protective factors which can be fostered to produce positive outcomes for the youth and family. 2) Structure Assessment of Violence Risk of Youth: assessment designed to estimate the risk of youth committing a specific offending behavior (like violent crime).
 For more information, visit https://getinchicago.org/understanding-youth-risk-level/. For a similar study on gun violence and recidivism please review https://getinchicago.org/wp-content/uploads/2018/07/Juvenile-Probation-Report-Chapin-Hall-GIC.pdf
 Remarkably, this study that used data matching through machine learning, omitted factors, such as postal code, school district, etc., that could be associated with race, in order to avoid interpreting race as a risk factor. Including race as a factor could bias the results since it might reflect possible justice system’s unequal responses to the offending behavior of different racial groups.