Before the Committee on the Judiciary of the United States Senate
Delivered July 21, 2010
“The Second Chance Act: More Evaluations of Effectiveness Needed”
My name is David Muhlhausen. I am Senior Policy Analyst in the Center for Data Analysis at The Heritage Foundation. I thank Chairman Patrick J. Leahy, Ranking Member Jeff Sessions, and the rest of the committee for the opportunity to testify today on the reauthorization of the Second Chance Act. The views I express in this testimony are my own and should not be construed as representing any official position of The Heritage Foundation.
My testimony focuses on the following points:
- More prisoners returning to society means more crime;
- Successful offender reentry is a multifaceted process;
- Recommendations for improving the Second Chance Act; and
- Scientifically rigorous evidence of the effectiveness of prison reentry programs is lacking.
More Ex-Prisoners on the Street, More Crime Congress’s desire to weigh in on the recidivism rates of former prisoners is easy to understand. In 2008 alone, over 735,454 state and federal prisoners were released back into society. However, only 52,348 (7.1 percent) of these former prisoners were released from federal prisons, while the other 683,106 (92.9 percent) were released from state prisons.
Releasing criminals back into society increases crime. Former prisoners have high arrest rates after returning to society. A Justice Department Bureau of Justice Statistics study of 272,111 state prisoners released in 1994 found that two-thirds of prisoners are rearrested within three years. After release, these offenders generate:
- Over 744,000 total arrests,
- 2,871 arrests for murder,
- 2,362 arrests for kidnapping, rape, kidnapping,
- 2,444 arrests for rape,
- 3,151 arrests for other sexual assaults, assaults,
- 21,245 arrests for robbery, and and
- 54,604 arrests for assault.
The highest rearrest rates were for robbers (70.2 percent), burglars (74.0 percent), larcenists (74.6 percent), and motor vehicle thieves (78.8 percent). Prior to their re-imprisonment, these prisoners accounted for 4.1 million arrests, including 550,004 violent crime arrests.
Any reauthorization of the Second Chance Act should fund another Bureau of Justice Statistics study of national prisoner recidivism rates. The results of the last study are 16 years old, so the results may not reflect current recidivism trends.
The high cost that released prisoners impose on society has been empirically demonstrated by Professor Steven Raphael of the University of California, Berkeley and Professor Michael A. Stoll of the University of California, Los Angeles. Professors Raphael and Stoll analyzed the relationship between prisoner releases and state crime rates from 1977 to 1999. Increased prisoner releases were associated with increased violent and property crime rates. A one-person increase in the number of released inmates per 100,000 residents in a state is associated with:
- 0.01 additional murders;
- 0.02 additional rapes;
- 0.18 additional robberies;
- 1.0 additional burglaries; and
- 1.0 additional larceny thefts.
Due to the amount of crime committed by former prisoners, federal, state, and local governments need to operate effective reentry programs. Preventing former prisoners from returning to prison is a worthy goal.
Policymakers on the national, state, and local levels need to understand the complicated nature of the reentry process. The reentry process begins in correctional facilities as inmates prepare for release and continues with their release back to society.
In addition to reentry public policies, other factors that influence successful transition of offenders from prison to community are individual characteristics, family and peer relationships, and community circumstances. Establishing a law-abiding lifestyle after prison involves locating living quarters, obtaining official identification, reconnecting with family, and finding legitimate employment.
The individual characteristics that influence recidivism include demographic characteristics, prison experience, employment history, education level, criminal record, and substance abuse dependence. For example, one long-term longitudinal study of offenders found that attachment to work is associated with reduced recidivism. Unemployed former prisoners and those without high school diplomas are more likely to drop out of reentry programs than those who are employed and have high school diplomas. Also, recidivists tend to have begun their criminal careers at an earlier age and had more serious criminal histories than those who do not recidivate.
Family and peer support is also important to the reentry process. The same long-term longitudinal study also found that marriage was associated with reduced recidivism. Also, former prisoners living with their families are less likely to drop out of reentry programs compared to their counterparts who do not live with their families. However, family conflict can also harm the reentry process, especially in the case of juvenile offenders returning to poor family environments. Just like the family, the influence of peers can influence the reentry process. Association with criminal peers can disrupt positive influences of the family.
Like the family and peer relationships of released offenders, the communities where they settle can provide positive and negative reinforcement. Many prisoners return to neighborhoods characterized by high degrees of social disorganization and crime. Socially disorganized, economically depressed neighborhoods tend to be associated with higher crime rates. Socially disorganized communities regularly lack socialization processes needed to encourage positive behaviors and dissuade negative behaviors.
The Second Chance Act
The Second Chance Act of 2007 expanded the federal government’s role in the provision of reentry services by creating grants for states to implement prisoner reentry programs. The Act authorized up to $330 million for prisoner reentry programs during fiscal years 2009 and 2010. The overwhelming majority of the spending authorization is for the operation of state and local programs.
Federalism Concerns. To address the issue of offender recidivism, the national government should limit itself to handling tasks that fall under its constitutional powers and that state and local governments cannot perform by themselves. First, the federal government should operate “evidence-based” reentry programs for offenders formally incarcerated in the federal correctional system. By “evidence-based” programs, I mean programs that have undergone rigorous scientific evaluations and found to be effective. However, programs based on models previously found to be effective still need to undergo rigorous scientific evaluations. Merely, replicating an “evidence-based” program does not necessarily mean the new program will yield the same results. Second, the federal government should not assume responsibility for funding the routine operations of state and local reentry programs.
The tendency to search for a solution at the national level is misguided and problematic. Offender recidivism is a problem common to all states, but the crimes committed by offenders in the state corrections systems are almost entirely and inherently local in nature and regulated by state criminal law, law enforcement, and courts.
Increasing the national government’s involvement in combating the recidivism of state and local prisoners is detrimental to quintessential federal responsibilities. Using federal agencies and grant programs to provide basic reentry services for state and local prisoners that the states themselves could provide is a misuse of federal resources and a distraction from concerns that are truly the province of the federal government.
A problem that is common to all the states, like offender recidivism, creates an avenue for federal action through the sharing of information and research, including the rigorous analysis of information coming from state and local agencies. Whether it is sharing successful policies and effective innovations or analyzing data, the federal government is well situated to perform this function. The promotion of rigorous research assessing the effectiveness of crime prevention programs is a worthy cause.
Supplement, Not Supplant. Under the Second Chance Act, the state and local grants for Adult and Juvenile Offender State and Local Reentry Demonstration Projects, New and Innovative Programs to Improve Offender Reentry Services, and Prosecution Drug Treatment Alternative to Prison are intended to supplement, not supplant, state and local funding. Supplanting occurs when federal funds are used to replace local funds, such as when federal funds intended for the expansion of reentry programs are instead used to pay for the operation of current programs or service levels. Supplanting has been a wide-spread problem in other Department of Justice grant programs. To ascertain the degree to which supplanting occurs with Second Chance Act grants, Congress should instruct the Office of Inspector General and/or the U.S. Government Accountability Office to conduct audits of grantees.
Performance Monitoring. Performance monitoring through the systematic and recurrent documentation of important features of program performance is crucial to assessing whether programs are operating as intended. When appropriately applied, performance monitoring can provide timely information on program performance to local program administrators and grant-making bureaus.
A Department of Justice Office of Inspector General (OIG) audit of prisoner reentry grants administered by the Office of Justice Programs (OJP) found that the grant-making bureau failed to adequately monitor grants. The OIG audited grants awarded under the Serious and Violent Offender Reentry Initiative (SVORI) and the Prisoner Reentry Initiative (PRI). Grants awarded under the Second Chance Act were still too young in the implementation phase for the OIG to perform an adequate audit.
The OIG found “little to no documentation of grant monitoring activities” for the SVORI grants. For the first two years of the SVORI program, the OJP did not develop any performance measures and could not explain to the OIG why performance measures were not developed during this period.
In particular, OJP did not properly instruct SVORI and PRI grantees on how to define and report recidivism rates. Further, OJP did not request that grantees, including Second Chance Act grantees, report baseline recidivism data. Baseline recidivism data is necessary for performance monitoring to adequately function. Without it, OJP is not able to judge the progress made by the grantees in reducing recidivism.
With performance monitoring, two potential problems for how local administrators respond to performance standards can arise. First, performance monitoring affects who receives provided services because of the incentive for “cream skimming.” Second, local administrators can respond to performance monitoring by “gaming the system.”
Performance management systems can cause local program administrators to select participants based on their perceived likelihood of success on performance measures. This process is called cream skimming. For example, local administrators of Department of Labor job-training grant programs have had robust incentives to select individuals most likely to have positive labor market results, regardless of whether those results were due to job-training participation. Even with the requirement that before-and-after performance measures be collected, performance standards added to the reauthorization of the Second Chance Act may provide local administrators with the incentive to carefully select participants based on anticipated positive outcomes. Local administrators of grant-funded reentry programs can potentially engage in cream skimming by overly selecting individuals believed to have the greatest chance of not reoffending, while discouraging those believed to pose the greatest risk of recidivating. This selection process will allow them to report to OJP a lower recidivism rate than they would have otherwise.
In addition to cream skimming, local administrators can engage in strategic behavior by manipulating whether or not reentry participants are formally enrolled, and thus recorded in the performance monitoring system. “Gaming the system” has occurred with Department of Labor job-training grant programs. Under the Department of Labor performance system, only individuals officially enrolled in job-training were counted towards performance standards. For instance, some local job-training administrators increased reported performance by providing job search assistance without officially registering those engaged in job search assistance. If an unregistered participant gained employment, then the individual would be officially enrolled and counted as a success. Individuals that failed to find work were not officially reported in the performance monitoring system.
Gaming the system can also occur with prisoner reentry grants. Local reentry program administrators may only include reentry participants in official reports to OJP after the former prisoners have managed not to recidivate after a period of time. Thus, the recidivism rate reported to OJP will be understated.
As the OIG noted, the Second Chance Act did not stipulate specific grant monitoring requirements. The reauthorization of the Second Chance Act provides Congress with an opportunity to instruct OJP to improve its monitoring of reentry grants. One corrective step is for Congress to define how recidivism should be measured and instruct OJP to collect baseline and ongoing recidivism data from grantees. Recidivism, reported annually over a three-year period, should be defined as rearrest, reconviction, and reimprisonment for new crimes and revocations. Recidivism rates should also be reported by type of crime (e.g., violent, drug, property, and other crimes).
If a performance monitoring system is created by the reauthorization of the Second Chance Act, then a strategy to prevent local administrators from cream skimming and gaming the system is needed. Requiring the U.S. Government Accountability Office (GAO) or the OIG to audit the validity of performance monitoring systems implemented by OJP may help reduce these problems.
While performance monitoring is important to learning how grantees are implementing their programs, performance monitoring does not provide a rigorous methodology for finding cause-and-effect relationships. Without control groups serving as a counterfactual, performance monitoring is unlikely to provide valid estimates of program impact. Second Chance Act grantees may be tempted to game the performance monitoring system or engage in cream skimming to report improved recidivism statistics. Thus, requiring Second Chance Act grantees to report recidivism rates is not a suitable substitute for experimental evaluations of effectiveness.
Not Enough Evaluation. A major focus of the reauthorization of the Second Chance Act should be gaining objective knowledge about the effectiveness of reentry programs funded by the Act. Reducing recidivism is important, so we need to find out what works.
The Second Chance Act funds a diverse set of programs across the nation. For this reason, the reauthorized version of the Second Chance Act should fund national, multi-site experimental evaluations of the programs that serve former federal and state prisoners. While evaluating small programs operating in a particular state or city is important, these evaluations do not shed light on the overall effectiveness of typical programs funded under the Second Chance Act. Just because a single program is found to be effective in a particular jurisdiction, or for a certain population, does not necessarily mean that the program is effective in other jurisdictions or among different populations.
Several sections of the Second Chance Act could be improved by the inclusion of congressionally mandated experimental evaluations. Grants administered under each of the following sections from the original legislation should undergo multi-site experimental evaluations:
- Adult and Juvenile Offender State and Local Reentry Demonstration Projects (Title I, Section 101);
- Residential Substance Abuse Treatment for State Offenders Program (Title I, Section 102);
- New and Innovative Programs to Improve Offender Reentry Services (Title I, Section 111);
- Prosecution Drug Treatment Alternative to Prison (Title I, Section 112);
- Family-Based Substance Abuse Treatment Grants (Title I, Section 113);
- Technology Careers Training Demonstration Grants (Title I, Section 115);
- Offender Reentry Substance Abuse and Criminal Justice Collaboration Program (Title II, Section 201);
- Mentoring Grants to Nonprofit Organizations (Title II, Section 211); and
- Responsible Reintegration of Offenders (Title II, Section 212).
In addition to state and local grant programs, the Second Chance Act created the Federal Prisoner Reentry Program (Title II, Section 231) for federal prisoners reentering society should be evaluated for effectiveness using a multi-site experimental evaluation design.
Keys to Successful Evaluation. There are several actions that Congress can take to ensure that the programs it funds are rigorously evaluated for effectiveness. First and foremost, Congress needs to specifically mandate in the laws it passes the experimental evaluation of the programs it authorizes.
The principal reason for the existence of reentry programs, obviously, is to prevent recidivism. Scientifically rigorous impact evaluations are necessary to determine whether these programs actually produce their intended effects. Clearly, there is little merit in the continuation of programs that fail to ameliorate their targeted social problems.
Estimating the impact of programs cannot be made with 100 percent certainty, but with varying degrees of confidence. Thus, all such impact evaluations face formidable control problems that make successful impact estimates difficult. As a general rule, the more rigorous the research methodology, the more confident we can be of the validity of the evaluation’s findings.
Determining the impact of social programs requires comparing the conditions of those who had received assistance with the conditions of an equivalent group that did not experience the intervention. However, evaluations differ by the quality of their methodology to separate the net impact of programs from other factors that may provide the real explanation for differences in outcomes for comparison and intervention groups.
Broadly speaking, there are three types of research designs: experimental designs, quasi-experimental designs, and non-experimental designs. Experimental evaluations that use the random assignment of individuals to the intervention and control groups represent the “gold standard” of evaluation designs. Random assignment helps ensure that the control group is equivalent to the intervention group. Equivalence means that the intervention and control groups have the same composition, predispositions, and experiences. Experimental evaluations are considered to be superior to quasi-experimental and non-experimental evaluations.
Randomized evaluations ensure that pre-program differences between the intervention and control groups do not confound or obscure the true impact of the programs being evaluated. Random assignment allows the evaluator to test for differences between the experimental and control groups that are due to the intervention and not to pre-intervention discrepancies between the groups. By drawing members of the interaction and comparison groups from the same source of eligible participants, these experimental evaluations are superior to other evaluations using weaker designs.
Under quasi-experimental designs, the intervention and comparison groups are formed by a procedure other than random assignment. Quasi-experiments frequently employ methodological and statistical techniques to minimize the differences between intervention and comparison groups that influence the outcomes being measured. This design frequently matches intervention and comparison group members together based on factors thought to influence program impacts.
Similar to quasi-experiments, non-experimental designs use statistical methods to isolate the effects of the intervention by attempting to make the intervention and comparison groups as equivalent as possible. Non-experimental designs often employ multiple regression analysis to isolate the effect of the intervention.
In both quasi-experimental and non-experimental designs, failure to remove the influence of differences that affect program outcomes may mean that the net impact of the intervention may not be actually due to the program, but caused by the underlying differences between the groups. While quasi-experimental and non-experimental designs use sophisticated techniques, experimental evaluations are still considered able to produce more reliable estimates of program effects. There is evidence that in the realm of criminal justice policy that quasi-experimental and non-experimental evaluations are more likely to find favorable intervention effects and less likely to find harmful intervention effects. Given that experimental evaluations produce the most reliable results, Congress should promote the use of experimental evaluations to assess the effectiveness of federal programs.
Second, these experimental evaluations should be large-scale, multi-site experimental evaluations. When Congress creates programs, especially state and local grant programs, the activities funded are not implemented in a single city or town. Federal grants are intended to be spread out across the nation. For this reason, Congress should require that these programs be evaluated using national, multi-site experimental evaluations. While individual programs funded by federal grants may undergo experimental evaluations, these small-scale, single-site evaluations do not inform policymakers of the general effectiveness of national grant programs. Just because a single program is found to be effective in a particular jurisdiction, or for a certain population, does not necessarily mean that the results are generalizable to the programs operating in other jurisdictions or among different populations.
Third, Congress needs to provide instructions on the types of outcome measures that will be used to assess effectiveness. When assessing the impact of reentry programs, the most important outcome measure is recidivism. Some have questioned the emphasis on recidivism as a measure of effectiveness compared to other measures that assess adjustment or reintegration of former prisoners back into society. While intermediate measures, such as finding employment and housing, are important, these outcomes are not the ultimate goal of reentry programs. If former prisoners continue to commit crimes after going through reentry programs, then the successful effects for intermediate outcomes will still matter little to judging whether the programs are effective. Impact evaluations relying solely on intermediate outcomes tell us little about the effectiveness of reentry programs in promoting public safety. While reentry programs should be assessed on intermediate outcomes, these measures should never serve as substitutes for recidivism outcomes.
Fourth, Congress needs to institute procedures that will encourage government agencies, often possessing entrenched biases against experimental evaluations, to carry out congressionally mandated evaluations. Of the nine prisoner reentry grants created by the Second Chance Act, the Department of Justice is responsible for eight and the Department of Labor is accountable for one (Responsible Reintegration of Offenders).
Simply mandating that an experimental evaluation occur does not necessarily result in the evaluation actually taking place. The Department of Labor has a poor track record for implementing and disseminating experimental evaluations mandated by Congress. For example, the Workforce Investment Act of 1998 mandated a large-scale, multi-site evaluation of the Department of Labor job-training programs. The results of the evaluation were to be finished by September 2005. Despite this mandate and deadline, the Department of Labor under the William J. Clinton and George W. Bush Administrations procrastinated over performing the evaluation. In November 2007, nine years after the passage of the Workforce Investment Act, the Department of Labor finally submitted a request for proposals for the evaluation. According to the U.S. Government Accountability Office, the evaluation will not be completed until June 2015—ten years after its original due date and 17 years after it was mandated by Congress.
While the National Institute of Justice within the Department of Justice has often demonstrated a stronger commitment in conducting evaluations, Congress still needs to take steps to ensure that evaluations are completed in a timely manner. One recommended method is that not later than one year after the reauthorization of the Second Chance Act, and annually thereafter, the Attorney General and Secretary of Labor be required to individually submit a report on the progress that their departments are making in evaluating the programs authorized under the Second Chance Act to the appropriations and judiciary committees of both chambers of Congress. Thirty days after the report is submitted to Congress, it should be made available on the web site of the Departments of Justice and Labor.
Fifth, congressionally mandated evaluations, upon completion, must be submitted to the appropriations and judiciary committees of both chambers of Congress in a timely manner. Thirty days after the any evaluation is submitted to Congress, the evaluation should be made available on the respective web sites of the Departments of Justice and Labor. Requiring that Congress and the public be informed of evaluation results is important because government agencies are quick to release positive results, but sometimes they are reluctant to release negative results. For example, a cost-benefit analysis of Job Corps that was finalized in 2003 found that the benefits of Job Corps do not outweigh the cost of the program, but the Department of Labor withheld it from the general public until 2006. An evaluation of Head Start that reported underwhelming results has also experienced unusual delays in being released by Department of Health and Human Services. While the evaluations conducted by the National Institute of Justice do not have the same history of delays, Congress still needs to be vigilant in ensuring that evaluation results are disseminated in a timely manner.
Prisoner Reentry Evaluations
There is considerable debate over the effectiveness of corrections and reentry programs. Some have concluded that several types of programs are effective, while others have cast doubt on the ability of these programs to reduce recidivism. Prisoner reentry programs operated by secular and faith-based organizations offer a wide range of services. However, there are not enough scientifically rigorous evaluations of secular and faith-based prisoner reentry programs to make generalizations about the overall effectiveness of these programs. While I was unable to identify any experimental or rigorous quasi-experimental evaluations of faith-based programs, I did identify five evaluations of secular programs: two used experimental methods, two used quasi-experimental methods, and one used a combination of experimental and quasi-experimental methods.
CEO Prisoner Reentry Program.  The Center for Employment Opportunities (CEO) Prisoner Reentry Program is an employment-based program that places recently released prisoners immediately in transitional jobs, usually in nonprofit or government agencies. While working their transitional jobs, participants receive assistance in finding permanent, unsubsidized employment.
An experimental evaluation found that CEO Prisoner Reentry Program participants did not have statistically different arrest rates two years after release from prison. After two years, the intervention group had an arrest rate of 37.7 percent, compared to the 41.8 percent arrest rate for the control group—a statistically indistinguishable difference of 4.1 percent. A statistically indistinguishable difference means that the difference between the intervention and control groups cannot be attributed to the program. However, CEO had more success at lowering conviction rates. After two years, the intervention group had a conviction rate of 30.5 percent, compared to the 38.3 percent conviction rate for the control group—a statistically significant difference of 7.7 percent. This difference in convictions is explained by the fact that the intervention group was less likely to be convicted of misdemeanors and not felonies.
After two years, the intervention group was less likely to be incarcerated in jail or prison. The intervention group had a reincarceration rate of 49.5 percent, compared to the 55.4 percent reincarceration rate for the control group—a statistically significant difference of 5.9 percent.
The program appears to be ineffective at moving participants into unsubsidized employment. During the course of the two-year evaluation, 59.6 percent of intervention participants found unsubsidized employment, compared to 62.8 percent for the control group—a statistically indistinguishable difference of 2.7 percent.
Washington State Work Release. During the early 1990s, 218 eligible prisoners were randomly assigned to serve out their sentences or enter work release facilities in Seattle, Washington. Participates were required to be involved in gainful employment or job training while participating in the program. Work release participants were obligated to remain in their work release facilities unless they were engaged in approved work and other activities.
One year after random assignment, work release participants had a recidivism rate of 22 percent compared to the recidivism rate of 30 percent of the non-work release participants. However, this difference of 8 percent was statistically insignificant, meaning that the difference cannot be attributed to participating in the work release program. Further, a cost-effectiveness analysis demonstrated “basically no differences in costs between work releases and inmates completing their full terms in prison.”
Boston Reentry Initiative.  The Boston Reentry Initiative is an interagency initiative designed to help move violent adult offenders released from jail back to their neighborhoods. Through multiple agencies, BRI uses mentoring, social service assistance, vocational training, and education to help offenders reintegrate into society. Rather than selecting participants most amenable to rehabilitation, BRI officials selected what they considered to be the “highest risk offenders” for treatment.
While the evaluation of BRI did not use an experimental design, the propensity score analysis used in this quasi-experimental evaluation makes this evaluation more scientifically rigorous than most other quasi-experimental designs. Further, BRI’s focus on targeting high-risk offenders may bias the results of the evaluation to understate the program’s ability to reduce recidivism. Compared to the comparison group, BRI participants experienced statistically significant reductions of 30 percent in overall and violent arrest rates.
While the BRI evaluation found positive results, this program and others found to be effective need to be replicated and rigorously evaluated in other settings before policymakers and academics can conclude that these interventions are effective. In particular, BRI should undergo an experimental evaluation. The criminal justice programs that have been deemed “effective” and serve as “model” programs have often been those implemented under optimal conditions. These programs have been comprised of highly trained professionals operating under ideal conditions. In addition, the conditions under which these programs operate are carefully monitored to make certain that the participants receive the intended level of treatment. In the real world, program conditions are almost always less than optimal.
Serious and Violent Offender Reentry Initiative. Created in 2003, the Serious and Violent Offender Reentry Initiative (SVORI) was an interagency reentry pilot program that coordinated the activities of the Departments of Education, Housing and Urban Development, Justice, and Labor. Before and after release, program participants were provided education and training, family services, health services, and other transition services.
Much like the BRI quasi-experimental evaluation, an evaluation of SVORI used a propensity score analysis to estimate the impact of the program on participants. The evaluation assessed the impact of SVORI participation at 12 adult and 4 juvenile sites on official measures of recidivism.
For adult males, participation in a SVORA program did not lead to lower arrest rates three months to 24 months after release, compared to non-participants. Reincarceration rates of adult male participants were statistically indistinguishable from the reincarceration rates of non-participants three months to 24 months after release.
More success was found with adult female SVORI participants. While the rearrest rates of adult females were not different during the first six months after release, participants were less likely to be arrested nine months to 24 months after release.A similar pattern held for reincarceration rates. Reincarceration rates of adult female participants were statistically indistinguishable from the reincarceration rates of non-participants three months to 9 months after release. However, participants had statistically lower reincarceration rates 12 months to 24 months after release.
Project Greenlight.  Project Greenlight, a short-term, prison-based reentry program operating in New York City, applied cognitive-behavioral skills training to prisoners eight weeks before their release. The program mainly endeavored to increase “post-release outcomes by (1) incorporating an intensive multimodal treatment regimen during incarceration and (2) providing links to families, community-based service providers, and parole officers after release (although there was no actual community follow-up).” The cognitive-behavioral skills training approach used by Project Greenlight is labeled as a “What Works” or “evidence-based” model based on the results of previous research.
An evaluation found that Project Greenlight “did not reduce recidivism and may actually have increased it.” The evaluation used a mixed-design that combined a quasi-experiment design for the first five months of assigning inmates to the program with random assignment design during the last six months. Project Greenlight participants were compared to a group of inmates that did not receive any pre-release transition services and to a group that received alternative transition services.
Compared to the inmate group that received the alternative transition services, Project Greenlight participants saw their chances of arrest after one year increase by 41 percent. Project Greenlight participants did not have statistically different arrests rates compared to inmates receiving no services.
Policymakers on the national, state, and local levels need to be concerned about prisoner reentry. To address the issue of offender recidivism, the federal government should operate reentry programs for offenders formally incarcerated in the federal correctional system. Further, the federal government should not assume responsibility for funding the routine operations of state and local reentry programs.
Prisoner reentry programs need to be rigorously evaluated to determine their effectiveness at reducing recidivism. I believe the need for more evaluations transcends political party lines. Both Democrats and Republicans should agree on this issue.
Policymakers should not implement prisoner reentry programs because advocates of federal funding believe these programs are effective. There has to be a solid base of scientific knowledge demonstrating that these programs are effective. Thus, Congress needs to do more to ensure that the reentry programs it funds are rigorously evaluated.