An Exploratory Study of the Discretionary Use of Electronic Monitoring for Individuals Upon Release to Mandatory Supervised Release (MSR) in Illinois

There is limited research on the efficacy of electronic monitoring (EM), with mixed results, specifically for individuals released from a correctional facility. Upon request from the Governor’s Office of Management and Budget (GOMB), we conducted a study to explore the discretionary use of EM for individuals released from a prison facility to mandatory supervised release (MSR) with the condition of EM that is not statutorily required upon release. Many previous studies conducted to date are poorly designed and lack methodological rigor. Moreover, few studies examined parole populations, making it difficult to compare outcomes with other community corrections populations, such as pretrial and probation.

The current study sought to fill the need for current and methodologically sound research on the use of EM and provide recommendations on the use of EM for individuals leaving Illinois Department of Corrections (IDOC) facilities to MSR. The Prisoner Review Board (PRB) imposes conditions of supervision for these individuals, and parole agents employed by the IDOC supervise them.^[1]

This study used a cohort of individuals who served a term of incarceration and released in FY16. The cohort does not include individuals who were released pursuant to a return for a technical violation or who were resentenced to IDOC after violating probation. Further, individuals included in this study were not statutorily mandated to an EM device pursuant to their admitting offense.^[2]

This study used a mixed-methods approach to answer the following research questions regarding the discretionary use of EM upon release to MSR:

1. What policies do IDOC and PRB staff have in place regarding the use of EM?
2. What are the perspectives of IDOC staff and parole agents on conditions of MSR and EM?
3. What are the demographic characteristics of those ordered to EM versus those not ordered to EM, and what are the characteristics of EM?
4. To what extent are there differences in the programming and characteristics of prison sentences for people ordered to EM compared to those not ordered to EM?
5. To what extent are there differences in recidivism patterns among those ordered to EM versus those not ordered to EM?
6. Which entity is best situated to make decisions regarding EM?

To answer these questions, researchers examined PRB orders on individual conditions for release, EM hook-up data from IDOC, arrest data on individuals released from an IDOC facility in SFY2016, IDOC exit and admission files, PRB orders,^[3] and surveys of IDOC correctional counselors/field service representatives and parole agents.^[4]

Key Findings

Policies, Directives, and Statutes

We analyzed polices, directives, and state statutes related to EM for individuals on mandatory supervised release. The PRB is the overall authority regarding any placement or removal (including early removal) related to EM and decides conditions for any person released from IDOC to mandatory supervised release (MSR) 730 ILCS 5/3-3-1(a), see also 730 ILCS 5/5-8-1, 730 ILCS 5/3-1-2(l)].^[5] The board is composed of 15 persons appointed by the governor with backgrounds and experience relevant to the criminal justice system [730 ILCS 5/3-3-1(b)]. Illinois Department of Corrections correctional counselors and/or field service representatives provide recommendations to the PRB as to conditions of MSR for each individual set to be released. In addition, IDOC parole agents must maintain records as the PRB or IDOC require and the information entered into the master file for the individual releasee [730 ILCS 5/3-14-2(f)]. The PRB may only impose EM as a condition of MSR when statutorily required or explicitly ordered by the PRB (730 ILCS 3-3-7), and any approval for adding or removing EM for an individual on MSR must go through IDOC’s administrative channels, with the ultimate decision-making authority given to the PRB. For this report, the study sample included only individuals discretionarily placed on EM using radio frequency as a condition of MSR.

Surveys of IDOC Institutional and Parole Division Staff

Survey responses from 58 IDOC correctional counselors and field service representatives indicated that most view their duties in relation to providing recommendations to PRB as providing necessary information to the PRB, making recommendations where statutorily required. Many indicated they only recommend EM as a condition for MSR when statutorily required. Respondents indicated that factors related to an individuals’ current charge were most important when making condition recommendations to the PRB, as well as when specifically recommending EM as a condition. Additionally, respondents tended to view EM favorably, as a worthwhile correctional practice.

Most respondents indicated they do not use a risk and needs assessment (RNA) tool when making recommendations to PRB and in determining whether to recommend EM, specifically. While the Illinois statute does not require the (discretionary or statutory) use of an RNA for EM or GPS consideration, research indicates the importance of RNA use in determining appropriate populations for EM use, conditions of release, supervision, and appropriate triaging of caseloads for more intensive supervision and services based on level of risk and services to match criminogenic needs.^[6] Further, Gies and colleagues (2013) highly recommend incorporating, minimally, a risk assessment, particularly for identifying risk for violent reoffending when considering the use of EM as a condition of supervision. Currently, IDOC does not use a validated RNA to provide evidence-based decision-making and ability to identify appropriate supervision levels for those exiting IDOC, as indicated in the Crime Reduction Act (CRA) of 2009. This is further supported by Illinois statute 730 ILCS 5/3-3-7(a)(21), which indicates any individual released from IDOC is to “be evaluated by the Department of Corrections prior to release using a validated risk assessment and be subject to a corresponding level of supervision.”

Survey responses from 73 IDOC parole agents were related to the research questions of the study; however, some responses were useful for understanding parole agent descriptive information. Many respondents described their duties as providing supervision, ensuring compliance, and assisting individuals in their reintegration process back into society. Additionally, about half of the respondents viewed EM as a deterrent and “a tool to ensure compliance."

Criminal Justice History and Outcomes

Of the full sample (N=1,065) of individuals released from IDOC to MSR in SFY16, 458 (43.0%) were placed on EM by the PRB. Regardless of whether an individual was placed on to EM, the majority had prior arrests for at least one felony (n = 1,029, 96.6%) and at least one misdemeanor offense (n = 850, 79.8%), and at least one prior felony conviction (n=973, 91.4%). The majority had no prior convictions for misdemeanor offenses (n=631, 59.2%) and no prior sentences to probation (n=559, 52.5%). Figure 1 provides information on the types of crimes the sample was committed to IDOC for initially, prior to their release.

Figure 1

Sample of People Released from IDOC in SFY16, by Offense Type (N=1,065)

Source: ICJIA analysis of IDOC administrative data

Over half of the full sample was rearrested after release from IDOC (n=691, 65.0%), a higher proportion of which were not placed on EM by PRB (n=395, 57.2%). Of those in the sample who had at least one rearrest, they were most frequently rearrested within 6 months of release to MSR and more than half of rearrests occurred within one year of release to MSR. Further, the PRB ordered a greater proportion of individuals of color to EM compared to individuals who self-identified as White. Figure 2 provides criminal justice-related outcomes for individuals, by EM condition.

Figure 2

Percentage of Individuals who Exited IDOC in SFY16 with Any Rearrest, Reconviction, or Recommitment to IDOC, by Condition of EM (N = 1,065)

Source: ICJIA analysis of CHRI data, IDOC administrative data, and PRB order data.

Note: These differences do not control for any confounding variables that may mediate or moderate the actual relationship between being ordered to EM and recidivism, such as programs or services received in the community or in IDOC.

We applied a quasi-experimental research design to create a treatment group of individuals placed on EM (discretionarily) and a comparison group of those not placed on EM (who had not statutory reason to be placed on EM). We used Propensity score matching (PSM) to approximate a random comparison group based on relevant potential confounding variables against which to compare outcomes for those ordered to EM. After matching, the sample included 280 individuals in the treatment group and 280 in the comparison group for a total of 560. Chi-square tests for differences between the two groups revealed significant differences in terms of new IDOC admissions for person offenses and technical violations after release to MSR depending on whether the release included the condition of EM. A higher proportion of those ordered to EM upon release to MSR by the PRB were admitted to IDOC for a person offense and for a technical violation; however, these associations were weak, at best. Further, it is unknown whether technical violations were related to violations of EM.

Discussion and Recommendations

Overall, findings suggest that there is a higher proportion of those admitted to IDOC for a person offense and for a technical violation for those who are ordered to EM, discretionarily, upon release to MSR compared to those who are not ordered to EM upon release to MSR. It is unknown whether any IDOC commitments on or after MSR date were a result of a technical violation(s) related to EM while on MSR, as the data did not include information regarding the specifics of technical violations related to IDOC recommitments or any information on technical violations that may not have resulted in commitment to IDOC.^[7] While this makes sense, as greater surveillance is likely to detect more violations due to more rigorously enforced compliance monitoring,^[8] the initial development and purpose of EM was not to increase compliance and detect non-compliance, but to assist in the facilitation of reintegration, building self-esteem, and gaining socially valued skills for justice-involved individuals.^[9] For example, in a systematic review of EM for system-involved individuals on reducing recidivism,^[10]researchers found heightened surveillance of individuals released from a correctional facility and placed on EM as a condition of release can contribute to increased technical violations (i.e., missing curfew, missing appointments). While these infractions are not inherently criminal in nature, they can, and may often, result in reincarceration for those who may not otherwise pose an increased risk to public safety.^[11]

Previous research findings are generally inconclusive on the efficacy of EM in reducing recidivism for individuals on probation or intensive supervision and those on parole/post-release supervision.^[12] Further, other research indicates the potential for disruption and barriers to successfully meeting conditions of release and reintegration of individuals back into their communities.^[13] Some research indicates limited efficacy of EM on recidivism, specific to certain sub-populations of justice-involved individuals, assessed via actuarial RNA as high-risk to recidivate. However, this includes some indication of a dissipating effects at three years, as well as use of EM for individuals that are already at lower likelihood for recidivism.^[14] Further, though limited, some research indicates greater efficacy with the use of GPS compared to RF devices for EM purposes^[15] and, if used effectively, can help improve treatment and employment outcomes.^[16] Further, if EM is to be used as a deterrent to engaging in pro-criminal behavior, the data above suggest it may not be an effective deterrent as overall, a higher proportion of those ordered to EM upon release to MSR were admitted to IDOC, admitted to IDOC for a technical violation, and admitted to IDOC for a person offense on or after release to MSR; however, it is unknown if these recidivism outcomes occurred while an individual was connected to an EM device.

The following recommendations are based on the study findings and additional communications and documentation provided by IDOC and the PRB.

Recommendation #1: Implement an actuarial risk and needs assessment (RNA) for IDOC to support supervision recommendations, PRB condition decisions, and IDOC MSR case plans and services. This is a requirement of the CRA. Further, reduce reliance on offense severity and offense type in condition decision-making and supervision levels for both IDOC recommendations and PRB conditions, as offense severity does not equate to level of risk for recidivism in the community.^[17]

Surveys indicated no RNA is conducted within IDOC prior to an individual’s release to MSR nor while they are on MSR, though a pilot risk assessment tool created in-house by IDOC is being piloted in two districts. Per communication with IDOC personnel, approximately 20 parole agents monitoring around 1,500 individuals currently use the pilot tool, nominally updated since its launch in October 2019; however, this tool will be retired once the Ohio Risk Assessment System (ORAS) is implemented. In correspondence, the IDOC administration indicated it is awaiting training and implementation of the ORAS.^[18] While prior criminal history is one, albeit static (non-changeable) aspect of an RNA, there are several other areas of risk that are cumulatively related to an individual’s risk for recidivism (or future law enforcement contact).^[19]

Based on the review of statutes, IDOC and the PRB are required, as part of the Illinois CRA of 2009,^[20] to implement and use an RNA (or a risk, assets, needs assessment, RANA) and create policies, rules, and regulations regarding the use of RNAs.^[21] The CRA of 2009 states:

The Parole Division of the Department of Corrections and the Prisoner Review Board shall adopt policies, rules, and regulations that, within the first year of the adoption, validation, and utilization of the statewide, standardized risk assessment tool described in this Act, result in at least 25% of supervised individuals being supervised in accordance with evidence-based practices; within 3 years of the adoption, validation, and utilization of the statewide, standardized risk assessment tool result in at least 50% of supervised individuals being supervised in accordance with evidence-based practices; and within 5 years of the adoption, validation, and utilization of the statewide, standardized risk assessment tool result in at least 75% of supervised individuals being supervised in accordance with evidence-based practices. [730 ILCS 190/10(b)(1)]

Based on review of this Act and a recent auditor general report of PRB and IDOC, IDOC and the PRB have not aligned policies and practices with the CRA.^[22] Currently, there are no validated, reliable, RNAs used to create “standardized individual case plans that follow the offender through the criminal justice system,” [730 ILCS 190/10-b(1)(A)] and that an individual case plan should be provided, including those supervised in prison, which is to be based upon: assets, risks, and needs of the individual identified through “the assessment tool as described in this Act” [730 ILCS 190/10-b(1)(A)(i)] and incorporates treatment and “supervision services appropriate to achieve the purpose of this Act” [730 ILCS 190/10-b(1)(A)(ii)]. At the time of this report, IDOC had not adopted a validated and reliable RNA. This also suggests that the PRB may not be receiving scores from an RNA, nor have policies or procedures in place for use of an RNA in condition decision-making. Further, the Parole Division of IDOC has also not adopted an RNA.^[23] Further, the CRA indicates the establishment of a system of graduated responses (or graduated sanctions) to violations, in which:

The system of responses shall take into account factors such as the severity of the current violation, the supervised individual’s risk level as determined by a validated assessment tool described in this act, the supervised individual’s assets [emphasis added], his or her previous criminal record; and the number and severity of any previous supervision violations. [In addition], the system shall also define positive reinforcements that supervised individuals may receive for compliance with conditions of supervision. [730 ILCS 190/10-b(D)(iii) and (iv)]

Available research on use of EM, while limited, indicates EM is most appropriate for those identified as highest risk for recidivism via a validated and reliable RNA.^[24] Risk to recidivate and criminogenic needs and strengths are most accurately identified with a reliable, validated, and locally normed actuarial assessment rather than personal, potentially subjective, discretion, offense severity, and offense type. Consistent with evidence-based policy and practice and the CRA, IDOC institutional staff should provide RNA scores to the PRB,^[25] using individuals’ scores to:

• Identify conditions that target criminogenic needs and potential responsivity factors.^[26]
• Determine intensity of service needs based on risk.
• Better identify individuals in which EM may be effective.^[27]

Further, actuarial RNAs are research- and behavioral science-based and can help prevent or minimize biases and subjectivity, reduce unnecessary discretion in decision-making, help identify proper and appropriate resources for individuals, and/or increase fairness in the judicial process.^[28] Additionally, validated and reliable RNAs can help identify responsivity factors—or barriers to being successful on supervision, including compliance with conditions, treatment, and services (e.g., mental health/stability, language barriers, trauma, motivation, transportation, childcare, etc.).

It also is important that the tool is based on factors directly associated with risk for recidivism (or, the desired outcome, such as risk for institutional misconduct and reentry planning), validating, standardizing, and norming the tool to the local population to improve utility and accuracy.^[29] The customization of the tool andthe ability to select the tool that may be best suited to the population being served and corresponding point in the criminal justice system (i.e. pre-trial, incarceration, reentry, community supervision), can potentially enhance stakeholder buy-in, extended use and sustainability of the tool, and quality of tool classification and utility.^[30] Further, it is important to consider how the RNA will improve or enhance communication between agencies, resulting in better outcomes for its clients.^[31]

The Parole Division of IDOC indicated they were piloting a risk assessment tool that assessed factors that are largely static in nature (factors that are not amenable to change and do not identify criminogenic need areas that are directly associated with reoffending risk). The assessment was in use by approximately 20 parole agents supervising about 1,500 releasees.^[32] The reliability, validity, and accuracy of the instrument in its ability to determine risk for recidivism, appropriate supervision levels, and appropriate linkage to treatment and services is unknown,^[33] which could result in adverse or unintended effects that may be detrimental to an individual, including inaccurately identifying risk level, resulting in over- or under-supervising individuals.

It would be more efficient and effective for IDOC to begin training and implementation of the appropriate ORAS tools (given the different decision points within the system, for reentry, and in the community), as it has a wealth of research on its validity and reliability. Further, IDOC has already identified that once training and implementation of the ORAS occurs, the current piloted assessment will become non-existent.^[34] The ORAS needs only to be validated and normed or standardized to the Illinois MSR/parole population (for community supervision) and, separately, on the ORAS institutional and release tools for individuals incarcerated in an IDOC facility.

Careful considerations must be made in tool development, implementation, and use, and its actual utility, validity,^[35] and reliability.^[36] It is unknown whether the risk tool in use incorporates dynamic risk factors (criminogenic needs), which are necessary for parole agent case planning to link individuals on MSR/parole to appropriate services and the intensity (or “dosage”) of treatment and services needed to impact behavior change, based on factors most highly associated with risk for recidivism. Further, while the Service Planning Instrument (SPIn), an RNA previously used within IDOC correctional facilities, it is unknown how frequently SPIn scores were provided by IDOC correctional counselors or field service representatives within their clinical recommendations.^[37] It is unknown what and how much information is provided to the PRB for each individual hearing and whether and how long the SPin has been out of use by IDOC.^[38]

The use of RNAs are particularly important for a variety of reasons, but most notably for:

1. IDOC in developing individualized, evidence-informed housing and security placement decisions, understanding risk for institutional misconduct and programmatic and services decisions within IDOC institutions based on the findings of the RNA assessment(s).
2. IDOC in developing evidence-informed release plans (from IDOC correctional counselors and field service representatives) that are evidence-informed based on an individual’s highest criminogenic needs (or risk factors known to be most highly associated with risk for recidivism).
3. PRB in making evidence-informed release decisions and appropriate and relevant condition decisions that minimize bias and subjectivity and providing a basis for the conditions placed on individuals being released to MSR or parole.
4. IDOC Parole Division in creating individualized case plans in the community that target the criminogenic needs—or risk factors known to be most highly associated with risk for reoffending—to reduce these risks for potential recidivism while in the community and identify how to build on individuals’ strengths and assets.^[39]

It is further important as effective conditions and case plans based on an objective, actuarial assessment of an individual’s risk, needs, and assets (or protective factors) can decrease likelihood of recidivism,^[40] compared to that of gut instincts or more subjective assessments (e.g. interviews, professional judgment), which may pose issues of inconsistency in assessing risk and needs, misalignment of resources, and potential bias and/or stereotyping (intentionally or otherwise).^[41] Further, the foundation of evidence-based decision-making and effective supervision is the use of a validated, reliable RNA and adherence to the Risk-Need-Responsivity (RNR) model of supervision.^[42] Therefore, current release decisions and conditions are not based on evidence-informed and evidence-based practices.

The Administrative Office of the Illinois Courts recently implemented the ORAS,^[43] which includes nine separate instruments for use in various contexts and settings, including institutional and community settings. Risk and needs assessments must be conducted while an individual is incarcerated to facilitate appropriate programming, prior to release, and once in the community, as risk and needs at these points are likely to be different. These tools should be validated regarding the IDOC population and IDOC parole/MSR population, a standard practice for adopting and implementing RNAs.^[44]

Recommendation #2: Allow PRB to place individuals onto EM based on statutory requirements, providing the IDOC parole agents limited, discretionary use of EM to assist non-compliant releasees; limit PRB discretionary EM use to cases where substantive reasons justify its use, and that use is uniformly applied and documented. The IDOC Parole Division should also consider whether and how EM can be used to enhance rehabilitative goals, including whether EM may be impeding an individual’s compliance and ability to successfully complete conditions of supervision, as the current study data indicates it is not necessarily an effective deterrent to engaging in pro-criminal behavior.

Little is known and understood about how the PRB makes MSR condition decisions. Without a validated, reliable RNA, the PRB is limited in its ability to make objective, informed, and evidence-based decisions regarding discretionary EM use for individuals being released from an IDOC facility to MSR. While research is limited, there is some evidence to support EM efficacy with highest risk system-involved individuals.^[45] Further, by initially placing individuals on EM upon release into the community, parole agents are left without a potential redress for non-compliance, a point that was heavily noted in parole agent surveys. Overuse of EM may limit tools available to parole agents (or rewards through early removal of EM), in addition to potentially creating more barriers for individuals’ successful reentry and reintegration into their communities.

Qualitatively, research indicates those placed on EM feel stigmatized or shaming effects, felt their relationships were negatively impacted (i.e. stress, inconvenience, cost), privacy issues (e.g. not being able to get away from family or roommates when desired), social restrictions as a result of movement restrictions (i.e. not being able to go to the store or take a walk when desired, inability to meet up with support networks, friends), and obtaining and/or retaining employment due to EM’s visibility, not being able to stay late when needed, or having work interruptions from law enforcement calls.^[46]

In a memo dated July 15, 2019, to IDOC Acting Director and Chief of Parole, drafted in response to the Spring 2019 legislative session, the PRB outlined changes to conditions of release and compliance reporting for individuals released to MSR/parole and placed on EM as a condition (Appendix C in the full report). The memo described intentions for changes and clarified the change to 12-hour movements for individuals placed on EM as a condition of supervision (discretionary use). The memo included an example document for compliance reporting for individuals placed on EM or GPS, which provided IDOC parole agents with another avenue to secure early removal of EM or GPS, in addition to parole agents’ current ability to do this through PRB amendment order requests.^[47] However, the compliance report had not been used by parole agents at the time of this report, per IDOC and the PRB.

In survey findings and personal communication with the PRB and IDOC, EM with 12-hour movements was described as merely a “curfew monitor” for individuals, creating a span of 12 hours that may or may not be structured and could be done without EM. Curfew is not specific to those only on EM; however, it eliminates the issue related to timely approvals for movement. Parole agents indicated that the new policies regarding 12-hour movements eliminate the usefulness of EM as “that defeats the purpose of what EM is to be used for.” Agents said, “You can’t monitor an offender with that much movement per day,” and, “a ‘blanket policy’ regarding allowable movement is a very poor decision.”

Despite the parole agents’ reaction to the change, it is also largely unknown how allowing12-hour movements impacts EM (and its efficacy or lack thereof) and individuals’ successful community reentry. This is something outside the scope of this report but should be evaluated, as merely placing an individual on EM as a condition of release does not provide sufficient understanding as to how parole agents relay policies, expectations, and rules other than documents provided to their clients, nor parole agents’ actual application of EM for that individual on MSR. Further, incorporating feedback from advocates, information on parole agents’ use and understanding of EM’s application and purpose, and feedback from individuals directly impacted by EM, particularly as it relates to societal reintegration, would be important to gather for this purpose as well.

The PRB and IDOC, as well as those directly involved in reentry planning and community supervision, should be in concordance regarding terminology—application of specific conditions (re: electronic monitoring, home detention, and electronic home detention)—so that their applications and uses are consistent throughout IDOC (including parole districts and agents) and among the PRB; and specifically, for those individuals responsible for overseeing individuals in the community with the ability to limit, restrict, or otherwise grant movements. Further, because EM devices are operated by a third party through a monitoring center, another entity must be included in the discussions for policies and procedures related to EM application, use, and policies, particularly as it relates to approved movements and violation of those movements. As DeMichele (2014) notes, “these tools [EM via RF or GPS] are all dependent on humans and only work as well as the infrastructures supporting them and the people operating them [emphasis added]” (p. 396). In addition, these tools are not infallible. They can malfunction, including breaking and failing to report correctly. An overreliance on EM devices as the “silver bullet” for compliance can decrease the one-on-one contact with the supervising officer, potentially creating complacency in actual check-ins with clients, and increasing workload for supervising officers in the way of data on movements, responding to alerts (real or due to a malfunction), in addition to already large caseload sizes.^[48] It is suggested that treating EM as a program rather than a technological tool can create unintended negative consequences; rather than finding ways to catch people when they are non-compliant, EM should be considered for its ability to improve long-term outcomes.^[49]

Additionally, imperative to this study and the understanding of what EM is, as DeMichele (2014) notes, “GPS, radio-frequency devices, and other forms of electronic monitoring are only tools that officers can use,” and that, “researchers and policy makers need to step away from treating these tools as programs or strategies” (p. 396). Moreover, these tools can actually increase an officer’s job workload and costs related to supervision, with the potential for both positive and negative effects of the tools’ use.^[50] It can increase a parole agent’s workload related to receiving and verifying movement requests, continually going back and forth to the PRB, and the requirements necessary to effectively and consistently monitor and check for EM related technical issues or violations of EM. For example, one parole agent indicated that EM and the process for EM placement and removal, “…is more of a hindrance to me than to the offenders,” while another parole agent indicated, “it is more of a hassle than it’s worth,” and that, “as currently used [sic] it is a liability to the agents, and the system.”

The overall purpose of EM, at its impetus, was for rehabilitative purposes—where one could track an individual’s location while providing a means for two-way communication.^[51] However, since the 1980s, EM has been punitive in nature.^[52] Research and information from the parole agents survey identifies one potential, yet understudied use, of EM—to help support an individual to come back into compliance (from non-compliance with conditions of supervision), using it as a last resort prior to revocation.^[53] Placing individuals on EM upon release may send the message that they are already not in compliance or that, somehow, putting an individual on EM makes them more compliant (though there is limited research to indicate this). Electronic monitoring does not provide “intrinsic supervisory powers,” may provide minimal information as to where an individual is (or where they are not), and does not provide information relevant to what an individual may be doing, prosocial or otherwise.^[54]

The efficacy of the 12-hour movement policy is unknown outside of likely decreasing individual movement requests. The policy also hampers the ability for parole agents to restrict movement when appropriate and necessary. At the same time, too much discretion given to parole agents ability to restrict movements may also inhibit an individual’s ability to reintegrate into their communities and successfully complete their release supervision and attending conditions.^[55] Further, what is known about EM indicates, “little appreciable effect on recidivism rates” (Belur et al., 2020, p. 2). While EM has generally been identified as cost-effective, those savings have been tied to EM use in lieu of jail or prison, as it is generally more costly than traditional community supervision, but less than incarceration.^[56]

Considerations for Parole Agent Use of EM

Electronic monitoring, as currently used, does not serve rehabilitative goals or substantially impact recidivism, based on the findings from the statute, policy, and procedure reviews; survey findings; and administrative results. Therefore, it is important to understand how EM may be used to enhance other community correctional programming and practices that support behavior management and change, as the current study suggests it may not be an effective deterrent to engaging in pro-criminal behavior. However, little research is available in the area of EM’s utility in achieving rehabilitative goals. More rigorous, substantive research is needed in this area if it is to be a rehabilitative tool, including how best to implement the technology in a way that increases compliance and supports prosocial behavioral change. Some research suggests there is potential for utility when used appropriately and with fidelity to contingency management^[57] and behavioral management principles. Specifically, EM used as a method of delivering positive reinforcement, positive punishment, and negative punishment could be successful (Figure 3).^[58]

Figure 3

Descriptions of Reinforcement, Punishment, and Associated Theories

Source: Van Voorhis, P., & Salisburgy, E. J. (2016). Correctional counseling and rehabilitation (9th Ed.). Routeledge. Akers, R. L., & Jennings, W. G. (2016). Social learning theory. In A. R. Piquero (Ed.) The handbook of criminological theory. John Wiley & Sons, Inc.; Cullen, F. T., & Agnew, R. (2011). Criminological theory: Past to present (4th Ed.). Oxford University Press, Inc.

However, positive punishment, such as adding EM to bring an individual under supervision back into compliance if they are currently non-compliant with their conditions of supervision, is insufficient in rehabilitating justice-involved individuals because it merely acknowledges an unwanted or undesirable behavior but does not sufficiently provide support or teach the individual what to do instead.^[59] Research on the use of operant conditioning—or use of reinforcements—has been shown to be more effective in teaching new, more prosocial behaviors for justice-involved individuals, particularly compared to classical conditioning (or punishments).^[60] For example, Pattavina et al. (2010) noted that with traditional EM usage, positive reinforcement for compliant behavior is only provided to individuals during infrequent contact meetings, and thus, the effect of EM as a reinforcement is weakened compared to its use as a reinforcement provided in real-time. Instead of delaying positive reinforcement, individuals on EM with a mobile connection could be sent motivational messages encouraging compliant behavior, particularly at times where they are at highest risk for relapse.^[61] Therefore, the researchers suggest that EM with GPS monitoring and a mobile connection could be used as a notification system in which individuals are texted appointment reminders, alerted to nearby job opportunities or treatment services (at least until the individuals are able to strengthen their own skillsets), and sent motivational messages to reduce recidivism and supervision failure.^[62] With research, this could prove as one potential use of EM for individuals released to MSR or parole. Negative reinforcement could be applied by parole agents via early removal of EM if ordered as a condition of release to MSR/parole. Though the avenue of early removal of EM exists for parole agents, its use is limited (Figure 4).

Figure 4

Parole Agent Requests for Adding and Early Removal of an EM or GPS Device, 2019

Source: Data provided by the Illinois Prisoner Review Board Order Requests.

However, and frequently cited in criminal justice research decision-making, is understanding that decision-makers for EM may over-place individuals upon release to EM/GPS and/or over-revocate individuals based on a continuum of less to more serious violations as a way to be risk averse.^[63] Findings regarding placement on EM as a condition set by the PRB may indicate over-placement of individuals to (discretionary) EM upon release to MSR; however, this is largely unknown as the PRB has not identified policies and procedures, other than those in statute, for when and why EM is used for individuals being released into the community from an IDOC facility. Further, comments from open-ended questions on the parole agent survey indicate, “[EM should] only be used as a sanction for non-compliance…not implemented for new releases except for those required by statute,” and that the state should, “Eliminate it” and “Get rid of it all together [sic]. It is not an effective deterrent for most parolees that are required to be monitored.” In addition, agents reported “there is no consistency in who qualifies for EM and who doesn’t.” Though limited in responses and not generalizable, of 10 respondents, 70% indicated they were extremely dissatisfied with the PRB process of identifying EM placement for individuals being released to MSR. While this study could not identify the potential for over-revocation of individuals, it is possible to analyze PRBs’ decisions to put more restrictions on individuals upon release or in a higher revocation rate for those who must go back to the PRB due to a technical violation(s) (in addition to low-release decisions for those sentenced under indeterminate sentencing structures) as possible identification of risk aversion rather than best practice.

Recommendation #3: Develop more transparent policies and processes related to IDOC correctional counselor and field service representative recommendations, limiting recommendations to those that are supported by actuarial and clinical assessments, with offense information used for statutory release and condition purposes. Similarly, develop more transparent policies and processes related to PRB condition setting, limiting conditions to those that are related to actuarial assessments or, minimally, professional structured judgment tools. At minimum, conditions should require a reason or purpose, preferably grounded in evidence/research, as indicated in the CRA.

For the discretionary use of EM, there are no requirements or policies that indicate when, who, or how the PRB should make these decisions, aside from the use of an RNA and adherence to evidence-based practices per the CRA. The same is true for policies and practices for parole agents’ application of EM upon its order from the PRB (except for those statutorily indicated). Based on data from PRB orders, when there was an indication of why someone was being placed on EM, the most common reasons were due to

• Criminal (and/or juvenile) arrest and conviction history (largely unspecified).
• the current or instant offense (largely unspecified).
• A Class 1, Class X, or Class M offense (instant offense or unspecified).
• Variations (aggravated or otherwise) of residential burglary, robbery, home invasion, DUI, battery, firearm(s) arrest(s)/conviction(s)/history, assaults (among some others).
• An active order of protection or violation of an order of protection (ICJIA analysis of PRB order data).

This is supported by a Sentencing Policy & Advisory Council (SPAC) report which indicated when justifications for EM were provided (by either IDOC staff or the PRB), the reasons were not universally applied in all cases.^[64]

Based on our surveys and review of PRB orders, many of the recommendations provided by IDOC correctional counselors and/or field service representatives (aside from the actual conditions set by the PRB) had limited clinical or actuarial assessment information and/or program information. More frequently, these recommendations included information on an individual’s criminal and/or juvenile history or related to current offense (n=383, 36.0%), with limited clinical information. The 383 individuals who had IDOC recommendations that consisted of criminal/juvenile histories excludes instances in which the current or instant offense^[65] is mentioned; this would increase the number and percentage of offense information reported in IDOC’s “clinical” recommendations. While this is otherwise included in a master file provided to the PRB, it may be beneficial for IDOC correctional counselors and/or field service representatives to provide a more clinical or assessment-based set of recommendations that exclude those tools that are otherwise non-rehabilitative in nature (e.g., EM, GPS, electronic detention) and other non-clinical information that is/can be provided in individuals’ master file.^[66] This also includes the recommendation of close supervision, as this, per CRA, should be based on a validated RNA.

As mentioned previously, TCU scores have some inconsistencies in when substance use disorder treatment is recommended and/or ordered by the PRB as a condition; there is limited information on the level of intensity or service recommended for substance use disorder treatment, which may be of more assistance to the PRB when making conditions for substance use disorder treatment. Further, it is important to note that programming and treatment for substance use, mental health, or other cognitive and/or behavioral programming within an institution does not negate the need for those same services upon release from the institution, as the setting has changed for that individual to one that may be more risky or require more support when it comes to community reintegration.^[67]

Recommendation #4: Create an electronic database for PRB orders to better understand PRB decision-making, processes, and condition-setting, revocations, and other order information. This includes development of performance indicators to illustrate how EM—among other conditions—are being used, including supervision outcomes and potential unintended impacts on those outcomes. This further includes the incorporation of interviews and information from those directly impacted, negatively or positively, by EM.

While most PRB orders were legible, we found a substantial amount of illegibility when there were hand-written notes or comments on the PRB order. Further, we had to hand-enter PRB order data from PDF scanned forms into an Excel spreadsheet to identify who ordered to on EM upon release to MSR, among other conditions and release requirements. To provide, minimally, descriptive information regarding individuals incarcerated in an IDOC facility and their release decisions, the PRB should maintain an electronic database, one that preferably connects or is incorporated with IDOC’s current data management system, Offender360 (O360) or can be connected through individual’s IDOC numbers.

While performance measures are not indicative of efficacy of programs, practices, or in this case, tools (as EM is a technological tool and not a program or practice), measures can be used to understand the landscape of what is being provided to individuals upon release into the community from an IDOC facility. The PRB created new PRB order documents, and recently approved by the PRB Chairman, Chief Counsel, and Chief of Program Operations, to improve clarity in the orders (PDF fillable forms that can be used for data entry or data points that could be incorporated into O360, including definitions of conditions). The Chief Counsel of the PRB, stated,

It is important to have transparent and clear set of guidelines and definitions for these orders because due process is best served by making sure that expectations and the goals and restrictions being set are clear and understood from the outside. (J. Sweat, personal communication, June 29, 2020)

The recently approved version of these documents are provided in Appendix D of the full report.

The Governor’s Office of Management and Budget (GOMB) inquired about what measures could be collected to better understand how EM is used and its potential impact on individuals being supervised upon release from IDOC. If the authority for EM placement remains with the PRB, the PRB should be the entity collecting this information. If the authority of EM changes to the Parole Division of IDOC, then it should be the entity collecting this information. Both entities can collect data relevant to their respective positions. For example, the PRB could collect data based on orders whereas parole could collect data on order requests for adding or removing EM, reasons for this, and technical violations associated with EM violations or other violations while on EM (it is likely that parole keeps information on technical violations, in general). Some potential performance indicators to collect on a regular basis for EM could include:

• Demographic and offense information related to individuals placed on and not placed on to EM upon release into the community (i.e. age, race/ethnicity, offense class, offense type, arrest history, IDOC history, institutional misconduct).
• Length of time an individual is connected to EM (in addition, looking to see how length of time may impact risk for recidivism or not).
• Number of parole agents requesting amendment orders to enact EM for an individual they are supervising and reasons.
• Number of parole agents requesting amendment orders to disengage EM early for an individual they are supervising and reasons.
• Purpose of EM placement upon release to the community.
• Purpose of EM placement while in the community.
• Other quality-of-life indicators such as educational/vocational attainment, employment, (socio-emotional) skill development, stable housing, prosocial recreational/leisure activities, and/or supports in the community.
• Number of recommitments based on EM violations and number of recommitments based on other violations.
• The frequency with which EM violations result in recommitment to IDOC and attending descriptions of EM-related violations (what they were doing when they violated their EM movement).
• Number of parole/MSR conditions placed on an individual at any one time, as more conditions can increase recidivism. ^[68]
• Collecting RNA information to identify, more generally, highest needs among individuals under MSR or parole supervision, with reassessment every six months to identify any changes in risk and needs.
• Frequency (or number of) parole agent contacts and provider contacts.
• Dosage and intensity of services received in the community (number of hours per week, number of weeks, criminogenic need or responsivity factor being targeted).

Further, there are several measures identified in the Crime Reduction Act of 2009, specifically 730 ILCS 190/10(f).

While this list is not exhaustive, it can provide a good start to get a better understanding related to who is placed on EM, how it may impact them—negatively or positively—and any potential issues regarding EM use/application. In addition, (re)evaluation is vital to continuously understanding the use of EM post-incarceration, its impact, and whether it is having the intended effect—in theory—would be recidivism reduction and increase of compliance (if looking to deter individuals). However, as noted, use of EM as a deterrent is not necessarily effective as currently used, and consideration for how EM can be used as a rehabilitative technology is important and whether discretionarily ordering an individual to EM serves a legitimate purpose. This is especially relevant given the limited reasons behind the discretionary placement of individuals to EM upon release to MSR. Continuous evaluation also would provide for quality assurance on EM performance and application, helping to identify, clarify, and revise the use and purpose as the technology and population evolves. This information should be gathered and discussed by both the PRB and parole, as ordering an individual to EM is merely a condition whereas parole agents are responsible for the actual supervision, instruction, and application of EM in the field that can directly impact individuals’ outcomes. The Robina Institute is a good place to start in looking for information, research, policies, and procedures regarding measuring and identifying appropriate measures for EM and PRB decision-making. Most important is the collaboration, communication, and data sharing between IDOC and the PRB for EM application, use, and rules to identify how information can be shared on both ends to create more informed processes and decisions. The collaboration could enhance efficacy and efficiency in the flow of information and the clarity of information among these agencies, but also between administration, supervisors, line staff, board members, other stakeholders, and individuals released into the community regarding expectations, smoothing out the reentry process for both entities. This also includes getting a better understanding as to how parole agents use EM for those released to their supervision.

Study Limitations

First, data was a limitation. We specifically chose individuals released to MSR from SFY16 who were initially admitted direct from court; this limits the analyses above to those who are only directly admitted from court, generally on their first commitment (99.7% had no prior IDOC commitments). While this provides for a “blank slate” to look at the impact of EM and recidivism, it limits the generalizability by excluding those who were discharged and recommitted, those recommitted on a technical violation, or other ways an individual may be recommitted or committed to IDOC. Future research should consider the different types of admits and releases in relation to EM, potentially looking at any new sentence to IDOC admissions rather than only direct from court admissions. Further, future research could analyze those individuals who are released from IDOC on a technical violation who are placed on EM or specifically analyze those individuals whose release from IDOC stems from a technical violation related to EM. What makes the ability to repull this data and rerun the current analyses difficult is that PRB orders are currently in the form of PDF scanned files, requiring hand-entry of all PRB order information to identify who was or was not placed on EM upon release from an IDOC institution.

The inability to pull PRB data also was a limitation as they are not collected in a database or provided in an easily usable format. Hand-entering of data opens the door to possible human error. While this could be true for a database as well, there may be better protocols in place in a web-based or electronic database rather than hand-written PDF forms scanned for data entry. Further, some of the PRB orders were illegible or missing necessary information. For example, 262 (24.6%) of PRB orders for this sample did not identify a “Board Action” (MSR approval, release prior to hearing, or statutory parole approved).

Another limitation was the response rates for both surveys. With significant attrition, the survey information was rendered non-generalizable to IDOC staff or parole agents, but potentially non-generalizable to the total number of individuals that submitted the surveys due to small sample size. Low response rates are common among organizational surveys, including criminal justice organizations, compared to surveys distributed to the general population.^[69]

Another limitation was due to the time-frame in which this study was conducted. While the study was to be started in the Fall 2019, an intergovernmental agreement between ICJIA and GOMB did not commence until February 2020, resulting in an extremely limited time-frame in which to conduct this study.^[70] This ultimately limited our ability to talk to individuals on MSR or parole and their families, gaining their insights into the use of EM, its impacts, and its utility from their perspective. Future research should incorporate those affected by these policies, positively or negatively. In addition, future research should evaluate the actual application of EM for individuals supervised by parole agents, as the actual application on the ground of EM once ordered by the PRB is of limited knowledge.

Lastly, due to legal concerns of the PRB, we were unable to talk to or survey PRB members. This significantly hampered our ability to know or understand policies, procedures, and processes regarding how and why certain conditions are ordered for individuals. Therefore, most of the PRB-related information comes from Illinois statutes, the PRB website, PRB orders provided by the PRB, and other publicly available documents. Seeing as the PRB is the entity charged with making conditions for individuals released to the community, including the decision of EM, it would have been vital to get insight into the group of individuals that make these recommendations. However, recommendations below may provide for more transparency and understanding on policies, processes, and procedures related to PRB decision-making for individuals being released into the community from an IDOC facility.

Conclusion

The present study filled a gap in knowledge regarding EM use in Illinois. A document review revealed the PRB has the statutory authority over EM—whether by ordering an individual to EM upon release from an IDOC institution, placing an individual back on EM per a parole agent request, or early removal of EM per a parole agent request, in addition to MSR and parole revocation decisions and overall condition setting. While the PRB has overarching authority over EM per statute, the current evaluation is in accordance with SPAC’s conclusion that neither IDOC nor the PRB has sole authority, or has taken sole authority, for the implementation of EM or its practical application on the ground. This includes its discretionary application, use, rules, and policies, especially as it relates to what actually occurs between parole agents and their clients. We, too, could not identify regulatory consistency in the application of EM for the PRB nor how those released to MSR with the condition of EM are actually being supervised in the community. Clearer guidelines and policies as they relate to parole agents’ use, including policies regarding restriction of movement, are necessary to create transparency and clarity regarding expectations related to EM for individuals released to MSR. Further, clarity in legislation could better differentiate the roles and responsibilities of IDOC and the PRB.

Though limited in response rate and completeness, we analyzed surveys of IDOC correctional counselors and field service representatives—those responsible for making recommendations to the PRB regarding conditions of supervision and general “clinical” recommendations—and parole agents charged with supervising individuals upon release into the community. Among IDOC correctional counselor and field service representatives, the individual’s current charge was the most important factor in making condition recommendations. Neither the Parole Division of IDOC (parole agents) or IDOC correctional counselors and field service representatives use a validated, reliable RNA.

We analyzed many sources of administrative data to offer information regarding the use of EM and EM’s impact on recidivism outcomes. Most frequently, individuals in the final random sample (placed on EM and not placed on EM) (N = 1,065) had prior felony and misdemeanor arrests as well as prior felony convictions and 66.4% of the total sample had at least one arrest on or after their MSR discharge date. When conducting analyses on the full random sample, a higher proportion of individuals of color were ordered to EM upon release to MSR compared to their White counterparts. PSM was conducted, there were some differences on recidivism outcomes between those placed on EM and those not on EM who exited IDOC in SFY16 on a direct from court admission. Though weak associations, those placed on EM had a slightly higher proportion of post-discharge IDOC admits for person offenses and IDOC admits for technical violations on or after release to MSR based on the PSM analyses. There was no difference regarding race/ethnicity, as individuals in each of the groups for the PSM analyses were matched on race/ethnicity to create groups that were as equal as possible on a variety of measures.

While this report provides a wealth of knowledge largely unknown in Illinois, information remains unknown on PRB decision-making, input from impacted individuals and families, and MSR technical violations and use of EM by parole agents. It is also important for IDOC and the PRB to create a more collaborative relationship that provides for data sharing necessary to make evidence-informed decisions while enhancing compliance with the CRA. Further, implementation of policies and procedures should be discussed between IDOC and the PRB to create a more fluid, efficient, and effective process of communication between the two entities and the individual being released from IDOC to MSR, as “poor implementation can make things worse and add to the confusion, doubt, and uncertainty, leading to a lack of support among staff” (Burrell, 2018, p. 40).

Future research should include interviews and/or focus groups with parole agents, individuals on MSR, individuals who have been subjected to EM whether upon release or at any point in their post-release supervision, and, if possible, with PRB members. This report provides limited information on how individuals placed on EM view the tool, including positive or the negative aspects. The lack of voice provided to individuals under MSR or parole supervision who are or have been placed on EM limits the findings of this report, as does the limited information related to parole agent use and views on EM, how parole agents enact policies and procedures related to EM, and information regarding allowable (and restricted) movements. Additionally, future research should consider incorporating a larger group of individuals of different admission types; however, the use of PSM does enhance the rigor of the current study analyses. The shortened study period and lack of time and resources to enter thousands of paper PRB orders into a database limited analyses to what could be feasibly conducted.