Evaluation of the Honest Opportunity Probation with Enforcement Demonstration Field Experiment (HOPE DFE), 4 United States counties, 2012-2014 (ICPSR 36857)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Study Purpose View help for Study Purpose

The Honest Opportunity Probation with Enforcement Demonstration Field Experiment (HOPE DFE) evaluation was designed to answer the following research questions, grouped by type of evaluation:

Process evaluation

• What was the structural context for the implementation of HOPE in the four sites?
• Was HOPE implemented with fidelity in the four sites?
• What lessons were learned for implementation success, replicability, and sustainability?
• How do intensive drug treatment services offered with the HOPE programs compare with the principles of effective offender intervention?
• What were the communication pathways among HOPE stakeholders? Did these vary from site to site?
• How did HOPE probationers view their supervision experience?

Outcome evaluation

• Does HOPE participation improve compliance with conditions of supervision and reduce violations?
• Does HOPE participation reduce recidivism, measured by arrest, conviction, and probation revocation?
• What is the impact of HOPE on jail days served and prison days sentenced?
• What is the impact of HOPE on drug use?
• Does HOPE participation change potential mediators, including dynamic recidivism risk factors such as employment and housing stability?
• Does HOPE participation change attitudes that are potential mediators, including participants' criminal thinking/attitudes, perceptions of locus of control, and perceptions of the criminal justice system fairness/legitimacy?

Cost evaluation

• What is the cost of starting and implementing HOPE?
• What are the costs and (any) savings? How are these distributed among the agencies (level of government) participating in HOPE?
• Is HOPE cost effective?

Study Design View help for Study Design

Enrollment, baseline interviews, and randomization. Study enrollment periods varied by site: August 2012 to December 31, 2013 (Oregon), October 2012 to July 31, 2014 (Massachusetts), and August 2012 to September 30, 2014 (Arkansas and Texas). In each site, after an offender was sentenced to probation or otherwise identified as potentially HOPE eligible, a probation office administrator collected basic information from the probationer and referred them to either a HOPE probation officer or intake/assessment personnel for pre-screening and study referral. The evaluation research coordinator then approached probationers for baseline interviews, which were completed via audio-computer assisted self-interview (ACASI). Upon interview completion, participants were randomized into either HOPE or probation as usual (PAU) and remained in that status throughout the evaluation period.

Weekly telephone-based interviews. To determine whether attitudes of HOPE participants changed over time compared to those on PAU, following the baseline interview, a sub-sample of participants were randomly selected for telephone audio-computer assisted self-interview (T-ACASI). For a six month period, participants who consented were directed to call twice a week based on an assigned schedule and answer attitudinal scale questions from the full ACASI interview. Enrollment began in June 2013 (Massachusetts, Oregon, Texas) and September 2013 (Arkansas); enrollment ended in December 2013 for all sites. Individuals who completed interviews were compensated with a rechargeable gift card or probation fee credit, receiving a site-specific amount for each interview done (up to $130-$140 total).

Probationer interviews (baseline, 6 months, and 12 months). At each timepoint, interviews were administered using ACASI technology and were conducted in research coordinators' offices for privacy. Upon completion, non-incarcerated individuals received either a $20 gift card, $20 cash, or $20 probation fee credit (site-dependent).

Oral swab drug tests. Probationers who completed 6- and 12-month interviews were sometimes asked to provide an oral swab for drug testing to determine if HOPE was reducing drug use compared to PAU. Those who agreed were compensated with a $5 gift card or fee credit.

Fidelity analysis. Administrative data about court probationers were collected to monitor site-level fidelity. These data include HOPE court activity (warning hearing dates, violation hearing dates), drug testing (date, results, testing frequency), probation violations (date, type), warrant service (date issued, date served, agency), sanctions (date imposed, type, incarceration admission and release date, judges' exceptions), treatment (admission and discharge date, type, setting), and recidivism (arrest date and charge, revocation date and reason, conviction date and offense, incarceration term). Depending on the item, analysis was conducted at the incident or individual level. The research team also conducted three visits to each of the four sites (baseline, intermediate, and final) and arranged interviews with members of the local site HOPE team (stakeholders), which typically consisted of the HOPE judge, HOPE project coordinator, HOPE probation officers, probation management, county jail administrator, and local law enforcement. Site visits lasted 3-4 days, and interviews typically lasted 30-90 minutes.

Probation experiences interviews. During the final visit to each HOPE site, the team conducted interviews with a small number of probationers (n=21) selected by the research coordinators to obtain their thoughts about HOPE program operations and their experiences participating in the program. Interviewees were primarily on probation for at least six months.

Network data and analyses. During each site visit, HOPE stakeholders were asked to rate levels of communication with each other about the HOPE program and levels of involvement and importance of each stakeholder to the program.

Administrative data for outcomes evaluation. The research team worked with probation staff across all sites to procure administrative data on probationer demographic information, education level, employment status, criminal history, current probation, urine testing, violations of probation, treatment experiences, and recidivism. State-level criminal history data were also obtained from site-specific Crime Information Centers and historical court data repositories. Data acquisition completion varied by site: December 2014 (Oregon), September 2015 (Texas), February 2016 (Massachusetts), and March 2016 (Arkansas).

Economic evaluation. To track resources per individual, the team identified three subgroups based on the length of time post-intake data was available by probationer: 1,494 individuals with 6 months of data, 1,291 individuals with 12 months of data, and 625 individuals with 24 months of data. Quantity data came from administrative records, fidelity data, interviews with probation officers, and surveys with sites. Price data were obtained using data from the Bureau of Justice Statistics (BJS), Bureau of Labor Statistics (BLS), and other secondary sources. All prices were converted to 2015 dollars using the Consumer Price Index.

Sample View help for Sample

Eligible probationers were identified from new probationers and existing probationers with one or more recent probation violations. Eligibility criteria for probationers across all sites included risk level and at least one year of probation remaining. Although HOPE was originally proposed for high-risk probationers, medium- or moderate-risk probationers were included in Arkansas and Massachusetts, medium-risk cases with violations were included in Oregon and Texas, and low-risk cases with violations were included in Arkansas. Juveniles, non-English speakers, transfers, and some special caseloads were excluded from the study. While several attempts were made to enroll 400 individuals at each of the four sites, 1,580 individuals were assessed for eligibility. Of those, 1,504 probationers were randomized, with 743 assigned to the HOPE intervention and 761 to probation as usual. By site, 342 participants were from Arkansas, 392 from Massachusetts, 394 from Oregon, and 376 from Texas. Overall, HOPE and probation as usual (PAU) participants were similar in baseline characteristics, though this varied by site. On average, participants were about 31 years old, male, white, and high-risk, with seven prior arrests and three or more prior convictions.

Time Method View help for Time Method

Universe View help for Universe

• English-speaking adults under probation supervision in one of the four study sites.
• Key stakeholders for the HOPE program at each study site, including but not limited to judges, probation officers, probation management, jail administrators, project coordinators, and local law enforcement.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Description of Variables View help for Description of Variables

Fidelity analysis. The research team identified eleven key markers of expected HOPE model implementation: Leadership (documentation and consensus around HOPE leadership), Probationers at High Risk (percentage of moderate to high-risk probationers), Warning Hearing Compliance (compliance with HOPE model hearing script), Initial Drug Testing Frequency (percentage of probationers who received at least eight drug tests during the first two months), Stepped Down Drug Testing Frequency (percentage of probationers who received at least one drug test per month after two months), Exceptions for Missed Drug Testing (percentage of missed drug tests with consequences), Time to Violation Hearing (time between violation and hearing), Sanction Type (percentage of confinement sanctions), Sanction Dosage (percentage of sanctions that are 19 days or less), Sanction Certainty (percentage of violations issued a specific sanction), and Sanction Swiftness (time between hearing and sanction start).

Probationer interviews. Items for probationer interviews consisted of demographics (age, race, ethnicity, veteran status, education), employment situation, housing situation, family background, peer relationships, physical and mental health, criminal history, and program services needed and received. Attitudinal items included self-efficacy, readiness for change, legal cynicism, substance abuse treatment motivation, and community involvement. For the T-ACASI sub-sample interview, items included attitude toward probation officer and judge, deterrence related to drug use and probation compliance, perceived fairness of sanctions, tolerance for law violations, identification with criminal others, and attitudes toward the law.

Response Rates View help for Response Rates

Probationer interviews: Among 1,504 probationers, 978 (65%) completed a baseline interview, and 536 probationers (36%) completed the 6-month follow-up interview. Of the 1,312 probationers eligible for the 12-month follow-up interview, 459 (35%) completed it. For the T-ACASI sub-sample, 282 participants were approached, 250 consented, and 127 called in at least once.

Presence of Common Scales View help for Presence of Common Scales

• Domain scales originally from Returning Home (Urban Institute): Neighborhood Information, Family Emotional Support, Quality of Intimate Partnership, Relationship with Children, Peer Criminal Behavior, Friendships, Readiness for Change
• Domain scales originally from Multisite Adult Drug Court Evaluation (Center for Court Innovation in 2000): Attitude Toward Supervision Officer, Deterraece Related to Drug Use, Deterrence Related to Compliance with Conditions
• Legal Cynicism (Sampson and Bartusch, 1998)
• Mastery Scale (Pearlin and Schooler, 1978)
• Civic Attitudes and Skills Questionnaire (Moely, Mercer, Illustre, Miron, and McFarland, 2002)
• Criminal Sentiments Scale-Modified (Simourd, 1997)
• Texas Christian University (TCU) Correctional Residential Forms: Self-Rating at Intake

Original Release Date View help for Original Release Date

2023-07-13

Version History View help for Version History

Weight View help for Weight

Not applicable.