Improving Employment and Reducing Recidivism Among Prison Offenders Via Virtual Interview Training Tool

Improving Employment and Reducing Recidivism Among Prison Offenders Via Virtual Interviewing Tool

The goal is to conduct a confirmatory effectiveness RCT (and an implementation evaluation) of Virtual Interview Training (VIT) by comparing employment and recidivism outcomes of offenders receiving vocational services as usual (SAU) plus VIT (SAU+VIT) with the outcomes of offenders receiving only services as usual (SAU-only).

The plan calls for participants to include offenders who are at moderate to high risk for reoffending (with an emphasis on violent-crime reoffending) who are currently enrolled in a Vocational Village.

Study Overview

• Status: Completed
• Conditions: Criminal Behavior
• Intervention / Treatment: Behavioral: Virtual Reality Job Interview Training; Behavioral: Services as usual

Detailed Description

The goal is to conduct a confirmatory effectiveness RCT (and an implementation evaluation) of VIT by comparing employment and recidivism outcomes of offenders receiving vocational services as usual (SAU) plus VIT (SAU+VIT) with the outcomes of offenders receiving only services as usual (SAU-only). To meet the criteria in the funding announcement, the plan calls for participants to include offenders who are at moderate to high risk for reoffending (with an emphasis on violent-crime reoffending) who are currently enrolled in a Vocational Village. The investigators are well prepared to conduct this study because the team has extensive experience with evaluating VIT in several settings (schools and mental-health service providers) and has previously collaborated with the Michigan Department of Corrections, which administers the Vocational Villages.

Objective 1. Evaluate whether SAU+VIT, compared with SAU-only, enhances employment outcomes and reduces recidivism among this population (i.e., effectiveness). At the individual level, the investigators hypothesize (H) that SAU+VIT trainees, compared with SAU-only trainees, will have higher employment rates (H1), greater improvement in job-interview skills (H2), and reduced recidivism (H3) by six-month follow-up. At the system level, the investigators hypothesize that SAU+VIT will be more cost-effective than SAU-only (H4). Subobjective 1. Explore whether use of the computerized VIT system frees up SAU staff time for non-interview-practice-related vocational training, relative to SAU-only (system level).Objective 2. Evaluate the mechanisms of employment outcomes and explore the mechanisms of recidivism.

Based on Corbiere's model, the investigators hypothesize that interview-skill improvement and measured role-play interview performance will mediate the effect of interview training on employment outcomes (H5). Also, The investigators will explore whether employment outcomes mediate the relationship between interviewing skills and recidivism at six-month follow-up and twelve month follow ups.

Objective 3. Conduct a multilevel, mixed-method initial process evaluation of VIT implementation to assess the acceptability, scalability, generalizability, and affordability of VIT. The investigators will use focus groups, surveys, and interviews (among offenders, staff, and leaders) to identify facilitators and barriers to implementing VIT in a prison-based vocational service program (the Vocational Village). The investigators will use budget-impact analysis to estimate the cost of implementing VIT at the Vocational Village.

1.1 Virtual Interview Training (VIT). This is an immersive simulation that allows trainees to have a virtual conversation with, and influence the behavior of, a simulated character. Previous studies suggest that virtual reality simulation training is engaging, and as a result, trainees will practice willingly for hours. The VIT simulation far exceeds typical interview training by combining video, speech recognition, and nonbranching logic (described below) to create an interactive environment in which trainees encounter complex social cues and realistic interpersonal exchanges with a virtual hiring manager. VIT was designed as part of a collaborative partnership between SIMmersion LLC and the research team to improve interview skills using behavioral-learning principles. These principles help develop sustainable changes in behavior.

The team developed the VIT learning objectives, which the investigators re characterized using terminology from the literature. The job-related interview content objectives target dependability, negotiating skills (e.g., asking for a day off), teamwork, and honesty (e.g., following company policies). The interviewer performance objectives target comfort level during the interview, sharing information in a positive way, sounding interested in the position, sounding professional, and establishing a rapport with the interviewer. VIT includes the following components to help the trainee achieve the learning objectives:

• Study Type: Interventional
• Enrollment (Actual): 101
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Michigan
    • Ann Arbor, Michigan, United States, 48109
      • University of Michigan

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

1. 18 years or older
2. Identified as at moderate to high risk for reoffending with violent crimes (determined at the time of enrollment in the Vocational Village via the COMPAS Risk Assessment Classification Instrument [60])
3. Within three months of their earliest release date
4. Actively enrolled in a Vocational Village
5. Have at least a 6th grade reading comprehension

Exclusion Criteria:

1. Has uncorrected hearing or visual problem that prevents him or her from using the training
2. Has a medical illness that compromises their cognition (for example, moderate to severe traumatic brain injury).

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Services as usual + Virtual Reality Job Interview Training; In addition to the services as usual comparator, participants will participate in Virtual Reality Job Interview Training.	Behavioral: Virtual Reality Job Interview Training; Participants will engage in training related to preforming well during a job interview as well as repeated job interview practice with a virtual hiring manager. Participants will receive feedback and tips on improving skills throughout the training.; Behavioral: Services as usual; Study participants will be receiving Vocational Village services as usual that may include but is not limited to vocational skill training, daily living skill training, and social skill training.
Active Comparator: Services as Usual; Study participants will be receiving their Vocational Village services as usual that may include but is not limited to vocational skill training, daily living skill training, and social skill training.	Behavioral: Services as usual; Study participants will be receiving Vocational Village services as usual that may include but is not limited to vocational skill training, daily living skill training, and social skill training.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Employment as measured in Outcome Questionnaire	Employment rate will be measured by participant self-report using a questionnaire. The questionnaire will include information about their job status (unemployed, employed); if they completed any job interviews (yes/no) and how many interviews they completed, if they received a job offer (yes/no), and information about jobs they may have had since their release date (employer name, job type, average hours worked, hourly wage, start date, end date). It will also include questions about their housing status (staying at a friend's, having own apartment, having own house/condo, homeless, or other); and if they thought the virtual interview training helped them prepare for an interview. The employment status will be validated with administrative data collected by MDOC.	6 month and 12 month follow up after villagers are released; will occur through study completion, average of 4 years
Mock Interview Scoring	Videos of mock interviews taken pre-intervention and post-intervention will be rated using a developed scale by the UM study team. Performance will be rated using a 1-5 likert scale with 1 being "Poor", 3 being "Average" and 5 being "Excellent"	Assessments occur pre-intervention and post-intervention and data will be assessed throughout study completion, average of 4 years
Recidivism as measured in Outcome Questionnaire	Recidivism status will be collected during the 6 month and 12 month follow up survey. Questions include if they've been re-arrested (yes/no) and if they were convicted of a crime (yes/no). The recidivism self-report data will be validated by administrative data collected from MDOC.	6 month and 12 month follow up after villagers are released; will occur through study completion, average of 4 years
Cost Effectiveness of VR-JIT as measured by Pre-Delivery Cost Capture Questionnaires	The Pre-Delivery cost capture surveys ask for MDOC staff to list the number of hours they worked on certain topics, including job interview training, mock interviews, job development/supports, job application skills, educational team meetings, educational team correspondence, project team correspondence, SIMersion team correspondence. This will provide a baseline number of total hours worked on these tasks before VR-JIT use. The Installation Cost Capture Questionnaire will ask MDOC Staff and Research Staff to list the number of hours they worked on activities related to getting VR-JIT installed and up and running at the prisons.	Through study completion, average of 4 years
Cost Effectiveness of VR-JIT as measured by Installation Cost Capture Questionnaires	The Installation Cost Capture Questionnaire will ask MDOC Staff and Research Staff to list the number of hours they worked on activities related to getting VR-JIT installed and up and running at the prisons. This includes time spent on phone calls between sites, between IT members, UM staff, Simmersion staff, etc.). This will provide a total number of hours spent on getting VR-JIT up and running in these agencies.	Through study completion, average of 4 years
Cost Effectiveness of VR-JIT as measured by Delivery Cost Capture Questionnaire.	The Delivery cost capture surveys ask for MDOC staff to list the number of hours they worked on certain topics (including job interview training, mock interviews, job development/supports, job application skills, educational team meetings, educational team correspondence, project team correspondence, SIMersion team correspondence, VR-JIT teaching, and VR-JIT support while using VR-JIT. This will provide a total number of hours worked on these tasks while delivery of the VR-JIT tool is occurring. The Installation Cost Capture Questionnaire will ask MDOC Staff and Research Staff to list the number of hours they worked on activities related to getting VR-JIT installed and up and running at the prisons.	Through study completion, average of 4 years

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Psychological Distress Survey (SCL-10)	The SCL-10-R is a 10-item self-report checklist of psychological distress based on the Symptom Checklist-90 (SCL-90). This survey will be administered at pre-test, as well as 6 month and 12-month follow up. This survey asks participants, "During the past 30 days, how much have you been stressed by" and then lists items such as "Feeling blue, difficulty making decisions, trouble catching your breath", etc. They are then able to choose responses varying from "Not at all, A little bit, Moderately, Quite a bit, and Extremely".	At 3 points throughout the study completion, average of 4 years.
DSM-5 Level 1 symptom measure	The DSM-5 Level 1 symptom measure was created to aid in a comprehensive mental status assessment by measuring mental health symptoms. This survey will be administered at pre-test, as well as 6-month and 12-month follow up. This 20-item self-report survey asks participants, "During the past two weeks, how much (or how often) have you been bothered by the following problems?". They are then able to choose responses varying from, "None, Slightly Rare, Mild, Moderate, and Severe".	At 3 points throughout the study completion, average of 4 years.
Interviewing Anxiety/Comfort	Self-report survey that will be administered at pre-test and post-test. It includes 34 statements where participants can indicate whether they "Strongly Disagree, Disagree, Neutral, Agree, or Strongly Agree" to those statements about job interviewing.	At 2 points throughout the study completion, average of 4 years.
Interviewing Self-Efficacy/Confidence	Self-report survey that will be administered at pre-test and post-test. It includes 1 statement about how comfortable a participant is going on a job interview, with the options being "Extremely comfortable, Very Comfortable, Moderately Comfortable, Neutral, Slightly Uncomfortable, Very Uncomfortable, and Extremely Comfortable". The rest of the questions asks how skilled they are at certain tasks, with the options being, "Extremely skilled, Very skilled, Moderately skilled, Neutral, Slightly unskilled, Very unskilled, and Extremely unskilled.	At 2 points throughout the study completion, average of 4 years.
Interviewing Motivation/Enjoyment	Self-report survey that will be administered at pre-test and post-test. It includes 30 statements asking participants how true it was for them regarding preparing for a job. The options provided on a 1 to 7 scale, with 1 being "Not at all true", 4 being "Somewhat true", and 7 being "Very true"	At 2 points throughout the study completion, average of 4 years.
Acceptability/Usability of VR-JIT	Self-report survey administered only to those in the intervention group. The 15-item surveys asks about participant's general thoughts about the acceptability and usability of the program using rating scales from 1 to 4 and 1 to 7. This will provide a total score of participant's acceptability and usability of VR-JIT.	At one point throughout study completion, average of 4 years.

Collaborators and Investigators

• Sponsor: University of Michigan
• Collaborators: SIMmersion, LLC
• Investigators: Principal Investigator: Matthew J Smith, PhD, University of Michigan

Study record dates

Study Major Dates

• Study Start (Actual): September 16, 2019
• Primary Completion (Actual): January 1, 2024
• Study Completion (Actual): December 1, 2025

Study Registration Dates

• First Submitted: April 4, 2019
• First Submitted That Met QC Criteria: May 2, 2019
• First Posted (Actual): May 3, 2019

Study Record Updates

• Last Update Posted (Actual): January 22, 2026
• Last Update Submitted That Met QC Criteria: January 19, 2026
• Last Verified: January 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Behavior; Criminal Behavior
• Other Study ID Numbers: HUM00155161; AWD011726 (Other Grant/Funding Number: Department of Justice)

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No