National Institute of Justice National Institute of Justice. Research. Development. Evaluation. Office of Justice Programs
Crime Solutions.gov
skip navigationHome  |  Help  |  Contact Us  |  Site Map   |  Glossary
Reliable Research. Real Results. skip navigation
skip navigation Additional Resources:

skip navigation

Program Profile

Project Greenlight

Evidence Rating: No Effects - One study No Effects - One study

Program Description

Program Goals

Project Greenlight was an institution-based transitional services demonstration program that was piloted in New York State’s Queensboro Correctional Facility. The program was designed to be a short, intensive intervention that could serve a greater number of offenders with reentry services at a lower cost. The 8-week program concentrated on addressing key issues that face offenders when they transition from prison to the community, including housing, employment, and drug treatment.

 

Program Theory

Components of the Project Greenlight program were based on research that has identified principles that characterize effective correctional treatment programs. These principles are that treatment should address dynamic (criminogenic) factors; programs should employ cognitive–behavioral, skills-oriented, or multimodal treatment approaches; interventions should concentrate on the needs of participants, with higher risk offenders receiving more intensive services; and interventions should be implemented appropriately (Wilson and Davis 2006).

 

Services Provided

Project Greenlight provided prerelease services and connected participants to community-based services, but there was no follow-up in the community. So they could participate in program services, inmates were transferred to the Queensboro Correctional Facility roughly 60 to 90 days before their scheduled conditional release date. The first week provided participants with an orientation that introduced them to each of the program components and to the staff. Program staff provided participants with the information they needed to find and keep a job, avoid drug and alcohol relapse, and make better decisions in all areas of their lives following release from prison. Programming was designed to be nonsequential, to accommodate people working at different levels, and to adjust services as new participants entered the program at different times.

 

One component of the program emphasized changing antisocial behaviors and thinking of offenders by providing cognitive–behavioral treatment. The Reasoning and Rehabilitation (R&R) program was adapted to run for 60 days, which is a shorter period than the program was originally designed for and a significant deviation from the program model. Job readiness training was provided, which included preparing for and conducting job interviews and guidance in workplace behavior. The program also aimed at strengthening the practical living skills of inmates. Sessions covered budgeting, time-management strategies, how to use public transportation, how to set up and use a bank account, and where to get emergency cash or noncash assistance when money is scarce. Finally, drug treatment concentrated on relapse prevention and substance abuse awareness. Those participants who could acknowledge their substance-use issues went to a relapse prevention group, where each person designed a relapse prevention plan and prepared to enter treatment upon release.

 

In addition, the program helped participants build social supports that would be available upon their release from prison. A community coordinator built a network of community-service providers, and participants were connected with the providers before their release. A family counselor and family specialist worked with offenders and their families to address any remaining issues before release. Program participants were also introduced to parole officers and informed about parole supervision requirements so they could avoid parole-related problems.

 

Detailed release plans were created for each program participant, which captured what participants needed to do after they were released in areas of employment, education, community resources, family, and substance abuse. Participants worked on these plans with their case managers, who in turn discussed areas in the release plan with the parole officers who supervised the inmates in the community. The field parole officer could make changes to the plans as an inmate’s circumstances changed after release.

 

Key Personnel

Project Greenlight program staff included corrections counselors, an institutional parole officer, an alcohol and substance abuse treatment team, corrections officers, a family reintegration specialist, a family counselor, a community coordinator, a job developer, an operations director, a project director, a project assistant, and field parole officers.

 

Additional Information: Negative Program Effects

An outcome evaluation (described below in Evaluation Methodology and Outcomes) compared Project Greenlight subjects with a group of inmates who received transitional services through a program that was already in operation at the Queensboro Correctional Facility and with a group of inmates in Upstate New York prisons who received no transitional services. Project Greenlight appears to have increased the rearrest and parole revocation rates of inmates who participated in the program, when compared with inmates who did not participate following release from prison.

Evaluation Outcomes

top border

Study 1

Wilson and colleagues (2005) found that Project Greenlight (GL) participants, who had the most extensive prerelease programming, actually had worse outcomes when compared with Transitional Services Program (TSP) participants, who had some prerelease programming, and Upstate participants, who received no prerelease programming.

 

New Arrests and Revocations

GL participants had significantly more total arrests than TSP and Upstate participants. Forty-four percent of GL participants experienced one or more misdemeanor or felony arrests, compared with 35 percent of TSP participants and 32 percent of the Upstate participants. These differences were statistically significant. The GL group also had high rates of new felony arrests, compared with the other two groups, but the difference was not significant.

 

GL participants had their parole revoked at a significantly higher rate as well. GL participants experienced more parole revocations (29 percent of the group had their parole revoked), compared with the TSP group (25 percent had parole revoked) and the Upstate group (17 percent had their parole revoked).

 

Survival analysis confirmed these initial results: the GL group recidivated more often than the TSP and Upstate groups. Looking at total arrests that occurred during the 12-month period after release, 65.9 percent of GL participants remained arrest free, compared with 75.8 percent of TSP participants and 73.2 percent of Upstate participants. Thus, 10 percent more of the GL group experienced an arrest, compared with the TSP group, after 12 months. The Kaplan–Meier analysis for group comparisons found that the difference between GL and TSP in total arrests was statistically significant, and the difference between GL and Upstate in parole revocations was also significant.

 

Cox regression, which controlled for other variables that might influence outcomes (such as age, education, ethnicity, prior arrests), found that in terms of program effect, the addition of control variables does not mediate the negative effect of the GL program.

 

Finally, all programming of the GL program was delivered by individual case managers. When case managers were added as a proxy for the intervention program into the analysis, the results showed that two case managers in particular appeared to be associated with a significant negative impact on the GL program outcomes.

bottom border

Evaluation Methodology

top border

Study 1

Wilson and colleagues (2005) employed a quasi-experimental design to study the effects of participating in Project Greenlight (GL) on postrelease outcomes.

 

Potential study participants were identified by the New York State Department of Correctional Services (DOCS) MIS/Classification section based on specific eligibility criteria. Eligible inmates were defined as correctional release inmates who were 75 to 105 days from release. Only male inmates could participate in the study, and only inmates who were paroled to one of the five boroughs of New York City were included in the final analysis. Once identified as eligible, inmates were transferred to the Queensboro Correctional Facility in their home community of New York City. Offenders were not eligible to participate if they had a conviction for a sex offense, a high institutional risk score, a current disciplinary sanction, medical reasons, existing warrants, or an ineligible current assignment (e.g., work release, shock incarceration) because Queensboro was a minimum-security facility.

 

Eligible study participants who transferred to Queensboro were assigned to one of two primary study groups: the GL intervention and a comparison group that participated in the DOCS Transitional Service Programming (TSP), an existing transitional services program. During the first 2 weeks, the first 52 participants were immediately assigned to the GL intervention until the program was filled. After program slots were filled, the remaining participants were assigned to the TSP intervention for the next 2 months. After the initial group of GL participants began to leave prison, the primary protocol that was in place involved haphazard sequential assignment to one of the two groups at Queensboro on approximately a two-to-one basis (intervention to comparison). A third group (Upstate) also served as a comparison group and consisted of inmates who were not eligible to participate in GL services and who were ultimately released directly from upstate prisons to New York City without specialized reentry programming.

 

The sample initially was to include 2,016 potentially eligible participants, but the study was terminated early (the study lasted 1 year rather than the intended 3). The final count was 805 potentially eligible participants; of those, 735 inmates were included in the final analysis. GL participants (n=344) were 57.6 percent non-Hispanic black, 37.2 percent Hispanic, and 5.3 percent non-Hispanic white/other, with an average age of 33.6 years. TSP participants (n=278) were 55.8 percent non-Hispanic black, 36.7 percent Hispanic, and 7.6 percent non-Hispanic white/other, with an average age of 33.4 years. The Upstate control group (n= 113) was 53.1 percent non-Hispanic black, 40.7 percent Hispanic, and 6.2 percent non-Hispanic white/other, with an average age of 32.5 years. There were no significant differences among the three groups on demographics. There were also no significant differences on criminal history. The GL group was slightly more likely to have more prior arrests and convictions than TSP participants, and TSP participants were slightly more likely to have more arrests and convictions than the Upstate participants but none of the group differences rose to statistical significance.

 

Data for the study came from various sources. Background information, including demographic and institutional data, on study participants was obtained from the DOCS. The New York State Division of Criminal Justice Services (DCJS) provided information on criminal history in New York State. The primary outcome measure of interest was recidivism, including arrests and parole violations that occurred in the 12-month period after returning to the community. Information on new offenses came from a comprehensive statewide database kept by DCJS. The Division of Parole provided violation information and data on drug tests conducted during parole supervision.

 

The final analysis included only study participants who were under parole supervision in one of New York City’s five boroughs. Study participants paroled outside of New York City were not included. The study used survival analysis to examine postrelease failure, and control variables were introduced through Cox proportional hazard models. Life tables, which provide an accurate method of summarizing time-to-event data, were also used for each group to examine time-to-failure and survival curves. Terminal (or failure) events were defined as arrests, felony arrests, and parole revocations. Time intervals were based on weeks at risk. The cumulative proportion surviving is the percentage of the group that has not been arrested at the time interval specified (12 months). Tests of significance are shown for the survival trends using the Kaplan–Meier method.

bottom border

Cost

top border
There is no cost information available for this program.
bottom border

Evidence-Base (Studies Reviewed)

top border
These sources were used in the development of the program profile:

Study 1
Wilson, James A., Yury Cheryachukin, Robert C. Davis, Jean Dauphinee, Robert Hope, and Kajal Gehi. 2005. Smoothing the Path From Prison to Home: An Evaluation of the Project Greenlight Transitional Services Demonstration Program. New York, N.Y.: Vera Institute of Justice.
http://www.vera.org/sites/default/files/resources/downloads/Greenlight_Technical_Report_FINAL.pdf
bottom border

Additional References

top border
These sources were used in the development of the program profile:

Brown, Brenner, Robin Campbell, James A. Wilson, Yury Cheryachukin, Robert C. Davis, Jean Dauphinee, Robert Hope, and Kajal Gehi. 2006. Smoothing the Path from Prison to Home: A Roundtable Discussion on the Lessons of Project Greenlight. New York, N.Y.: The Vera Institute of Justice.
https://www.ncjrs.gov/pdffiles1/nij/grants/213714.pdf

Ritter, Nancy. 2006. “No Shortcuts to Successful Reentry: The Failings of Project Greenlight.” Corrections Today 68(7):94–97.

Wilson, James A. 2007. “Habilitation or Harm: Project Greenlight and the Potential Consequences of Correctional Programming.” NIJ Journal 257:2–7.
http://www.nij.gov/journals/257/habilitation-or-harm.html

Wilson, James A., and Robert C. Davis. 2006. “Good Intentions Meet Hard Realities: An Evaluation of the Project Greenlight Reentry Program.” Criminology & Public Policy 5(2):303–38.

Wilson, James A. and Christine Zozula. 2011. “Reconsidering the Project Greenlight Intervention: Why Thinking About Risk Matters.” NIJ Journal 268:10–15.
https://www.ncjrs.gov/pdffiles1/nij/235890.pdf

Wilson, James A., and Christine Zozula. 2012. “Risk, Recidivism, and (Re)habilitation: Another look at Project Greenlight.” Prison Journal 92(2):203–30.
bottom border


Program Snapshot

Age: 18+

Gender: Male

Race/Ethnicity: Black, Hispanic, White, Other

Geography: Urban

Setting (Delivery): Correctional, Other Community Setting

Program Type: Academic Skills Enhancement, Alcohol and Drug Therapy/Treatment, Aftercare/Reentry, Cognitive Behavioral Treatment, Group Therapy, Individual Therapy, Vocational/Job Training, Wraparound/Case Management

Targeted Population: Prisoners

Current Program Status: Not Active

Listed by Other Directories: National Reentry Resource Center