Search Type Title Author/Investigator Search Advanced Search

Table of Contents | Previous | Next

Chapter 2. New York: Center for Employment Opportunities

Introduction

This chapter describes the Hard-to-Employ evaluation of the Center for Employment Opportunities program (CEO). CEO is a prisoner reentry program that provides employment services to formerly incarcerated people returning to New York City and to people under other types of community supervision. The focus of this evaluation is on the group of clients who were referred to CEO by their parole officer.

This chapter begins by providing information about the complex problems of people who have been recently released from prison and explains why it is urgent for policymakers to understand and support programs that are designed to address their needs. Next, the chapter describes the CEO program’s services, followed by the research design and the procedures used in the evaluation, including the key outcomes and the data sources that will be used to track them. Then the chapter describes the characteristics of the sample population, followed by results from an early analysis of program participation patterns, including the details of each of CEO’s core services and the exposure of the treatment group to these services. The chapter concludes with a description of enhancements that were made to CEO’s program during the study period.

Background and Policy Relevance

Nationally, over 600,000 people are released from state prisons each year. Former prisoners must work hard to become reintegrated into the community, find stable employment and housing, and support their families. Unfortunately, the obstacles they face are substantial. Many have very little income and extensive financial responsibilities, including child support arrears, that have continued to mount during their incarceration. The criminal records of many individuals continue to severely limit their employment options and adversely affect their social outcomes. In addition, many have substance abuse and mental health problems that require ongoing treatment. Not surprisingly, rates of recidivism are extremely high. Recidivism is costly for everyone: the individual, their families, local communities, and the larger society.

The criminal justice system has experienced many changes in recent years, including a tremendous increase in incarceration. Consequently, unprecedented numbers of prisoners are being released each month. In fact, more than four times as many prisoners were released in 2004 as in 1980.1 Many individuals are reincarcerated for parole violations and cycle in and out of prison or jail multiple times for the same original offense. Over one-third of the prison admissions each year consist of people who were out on parole.2 The large number of former prisoners who fail to become reintegrated into society and end up back in prison costs taxpayers billions each year. In 2004, for example, expenditures on corrections were estimated to be over $40 billion.3 Even though spending has increased, former prisoners are no more likely to succeed than they were 30 years ago. Within three years of their release, two-thirds are arrested and more than half return to prison or jail.4

As increasing numbers of prisoners are released to communities each year, concern for public safety is mounting, particularly in inner-city neighborhoods. Many ex-prisoners are returning to communities that already have high rates of crime, unemployment, and poverty.5 In some neighborhoods, more than 25 percent of all men between 18 and 64 are on probation or parole. In these same neighborhoods, more than one of three families live below the poverty line and one of six receive public assistance.6

Prisoner reentry also has direct effects on families and children. More than half of re-entering adults have children under 18, and more than 10 million children in the United States have a parent who was in prison at some point in the child’s life.7 Not surprisingly, long periods of incarceration can be detrimental to family ties and can alienate the recently released parent from his/her children. Aside from the prolonged affects on children’s well-being, diminished family bonds make it all the more unlikely that prisoners will succeed at becoming reintegrated into their communities.8 Many families who were relying on public assistance before a parent was incarcerated suffer increased financial burdens. Research has shown that as many as 44 percent of the caregivers of children with an incarcerated parent report receiving public assistance, such as Temporary Assistance for Needy Families.9 Furthermore, upon their release, nearly one-third of former prisoners expect their families to depend on public assistance.10

While former prisoners face many complex challenges, work seems to be a key ingredient in determining the success or failure of their transition back to society. Studies have shown that when ex-prisoners find and maintain formal employment, their chances of recidivism are reduced and that the better the job, the less likely their chances of recidivism. This finding is particularly true for older former prisoners.11 When ex-prisoners are employed, their housing conditions may be better and their relations with their families and their communities may improve. Many recently released people report feelings of disconnect and alienation from society. Employment can be a key factor in helping these men and women feel more connected to mainstream society and in encouraging them to move away from a criminal trajectory.

Finding a steady job is a major challenge for ex-prisoners. Many employers are reluctant to hire someone with a felony record. In fact, employers are much less likely to hire a former prisoner than a member of any other disadvantaged group.12 In a survey of 3,000 employers, two-thirds reported that they would not knowingly hire a former prisoner.13 Most recently released people have other attributes that make them less appealing to potential employers, such as low educational attainment and limited work history. They may also have unstable work habits or competing demands from drug treatment programs, curfews, or other restrictions on their mobility that can make it even more difficult to find and keep full-time employment.

Well-rounded employment services for former prisoners are critical to ensuring better postrelease outcomes. While there are community programs that aim to provide these needed supports, few operate on a large scale and little is known about how effective they really are.

Program Description

CEO in New York City is one of the nation’s largest and most highly regarded employment programs for formerly incarcerated people. Developed by the Vera Institute of Justice in the 1970s, CEO has operated as an independent nonprofit corporation since 1996. It serves an average of 1,800 returning men and women each year. Its paid staff of around 150 work in a variety of positions, including supervising transitional work crews for its Neighborhood Work Project (NWP), providing preemployment training and job development services as part of the Vocational Development Program, and providing executive and administrative support. CEO’s transitional employment program is funded largely by government institutions that hire CEO work crews to perform basic maintenance and other functions. CEO also receives funding from the New York State Division of Parole, local criminal justice agencies, federal Workforce Investment Act (WIA) funds, and private foundations to cover the cost of its vocational development programs.

Participants start the program with a four-day life-skills class focusing on job readiness that covers workplace behavior, job search skills, and decision-making. As part of this component, CEO makes sure that each participant has all the official identification and documents necessary for employment. Participants are then placed at one of 30 to 40 work sites around the city. The work sites are public agencies, such as the City University of New York, which have secured CEO’s services through the New York State Division of Parole. Participants are paid New York State’s minimum wage14 for four days of work per week (they are paid daily, at their work site). These transitional placements are the heart of CEO’s program. Parolees often have a pressing need for cash, and the placements provide them with “daily pay for daily work.” In addition, CEO uses the transitional period to identify issues that are likely to hinder participants’ performance in an unsubsidized job and to work with participants to address these issues. Most participants stay at work sites for two or three months.

Participants spend the fifth day of each week in CEO’s main office, meeting with job coaches to discuss their work performance and prepare for interviews and with job developers to discuss permanent employment opportunities. During the study period (approximately January 2004 through December 2005), there were also some opportunities for short-term, employer-driven training. Participants may spend their fifth day participating in other activities, such as an extensive fatherhood program that helps participants to resolve child support issues (such as reducing current child support orders) and to improve their family relationships.

Research Design

This evaluation rigorously tests the effects of the core components of CEO’s postrelease employment program for parolees. The study provides a test of whether a well-designed post-release program model that provides immediate, transitional work and job placement services in a supportive environment can lead to increased rates of permanent employment and reduced rates of recidivism and reincarceration.

The impacts of CEO’s program will be assessed using a random assignment research design. For purposes of the evaluation, clients whose parole officer referred them to the program, who reported to CEO, and who met the study eligibility criteria (discussed further below) were randomly assigned to one of two groups:

• Neighborhood Work Project (NWP) group (program group). Individuals who were assigned to this group received all of CEO’s program services (described in detail above), including a four-day life-skills class, placement in a transitional job, job coaching, additional services such as the fatherhood program, permanent employment placement services, and all postplacement services.

• Resource Room group (control group). The Resource Room program was designed to provide a basic level of service to individuals who were assigned to the control group and to provide a benchmark against which CEO’s core program could be compared. Individuals assigned to this group participated in a revised version of the job-readiness class (life skills) that lasted one and a half days. Participants were then given access to a resource room equipped with computers (with job search software), phones, voice mail, a printer, a fax machine, and other job search tools, including publications. When clients came into the resource room, a staff person was available, if needed, to assist them with many aspects of job search, including use of the equipment, help writing a résumé, and assistance setting up a voice mail account so that potential employers could leave messages for them.15

One risk in the design is that some members of the Resource Room group may have sought assistance from other employment programs that offer services similar to those provided to the NWP group. This could dilute the impacts of the program comparison, although there are very few other programs offering transitional work to ex-prisoners. The 15-month follow-up survey will help determine whether control group participation in other programs is really an issue, because the survey will obtain information about program participation since random assignment for both research groups.

Random Assignment and the Sample Intake Process

The MDRC team worked with CEO and the New York State Division of Parole to design a random assignment process that ensured both that the study did not decrease the number of people who received NWP services and that CEO had enough participants to fill its contractual obligations to NWP work site sponsors.

CEO enrolls a new cohort of paroled clients each Friday. In order to accommodate the research study to this routine, each week CEO staff decided in advance how many slots were available for new NWP clients. This information was then entered into a customized database designed by MDRC. When clients arrived on Friday morning, staff used the database to conduct an attendance check, and the system compared the number of attendees to the number of available NWP slots. If there were at least four “excess” clients present, the system randomly assigned between four and 12 clients to the Resource Room group.16 If there were fewer than four excess clients present, random assignment was not conducted that week.

Certain eligibility criteria complicated the process. First, clients from several referral sources arrived together on Friday mornings. For contractual reasons, individuals referred from certain sources (such as a special program at Queensboro Correctional Facility) and Shock Incarceration17 participants had to be placed in NWP and were therefore ineligible for the research.18 Similarly, for both ethical and methodological reasons, individuals who had participated in the NWP program in the past year (“recycles”) were also excluded from the study and assigned to the NWP program. The MDRC database automatically identified these special cases and gave them NWP slots. As discussed above, only “regular parole” clients referred from parole offices around the city were eligible for random assignment to NWP or the Resource Room program.

Only individuals who signed an informed consent form were included in the study sample. Each Friday morning, CEO gave participants a description of the study and random assignment and asked them to sign a consent form agreeing to be part of the study. Regular parole clients were randomly assigned to one of the two program groups, even if they did not agree to be in the study. Nonetheless, the vast majority of clients signed the consent form and agreed to participate in the study. Individuals did not have to consent to the study in order to receive CEO services and meet their parole obligations. Figure 2.1 shows the flow of clients through the random assignment process.

Sample build-up moved more slowly than anticipated because of the need to balance the number of clients at CEO’s work sites and restrictions on the inclusion of certain groups in the study. In addition to slow sample build-up, there were some difficulties managing the expectations of participants assigned to the control group.

Before the study began, CEO placed most eligible clients who came to its offices at a NWP work site. Although CEO, the Division of Parole, and MDRC made every effort to communicate the research design and the possibility of participants not obtaining a work site placement, some parole officers were still unaware of the change at CEO and sent participants there expecting to be placed at a work site. Moreover, CEO is well known in the community and clients often heard about the program from friends or relatives. As a result, some clients assigned to the control group expressed disappointment and frustration when they realized that they would not be placed at a work site. Some parole officers also expressed concern that some of their clients were not obtaining NWP placements from CEO. As would be expected, these situations were uncomfortable for CEO staff. While the study did not reduce the total number of clients who obtained NWP placements, it appeared unfair to particular clients who came to CEO and were not given access to NWP. As random assignment proceeded over time, parole officers and participants became more familiar with the new process at CEO, and staff had less difficulty managing expectations.

Intake for the study ran from January 2004 through October 2005, nearly two years, and involved a total of 977 participants: 568 were assigned to the NWP group and 409 to the Resource Room group. This sample will require a 5 to 8 percentage point difference in outcomes between the two research groups to detect an impact. If the program produces an effect of less than that amount, it may not be statistically significant. It is important to note that even relatively small impacts may be quite policy-relevant. For example, if CEO can generate even a modest reduction in reincarceration, the resulting cost savings to taxpayers could be substantial.

[D]

 

Baseline Data and Key Outcomes

Baseline data were collected from a short baseline information sheet. Some additional baseline data were also obtained from CEO’s internal database, which contained information from a referral form that parole officers are required to fill out when they refer a client to CEO. To be accepted into the program, clients must have this form when they arrive at CEO on Friday mornings.

The study will use several types of follow-up data to assess the impacts of the program:

• Criminal justice administrative data. These data provide information on a range of outcomes, including arrests, parole violations, convictions, and incarceration, for each member of the study sample.

• Earnings and employment data. MDRC is collecting data to show quarterly employment in jobs covered by the unemployment insurance (UI) system in New York State for each sample member. State employment data may be supplemented with information from the National Directory of New Hires, a database maintained by the federal Office of Child Support Enforcement. Data from the New Hires directory would provide information on earnings from employment both within and outside New York State.

• Child support administrative data. These data will include a history of any formal child support payments by sample members who are noncustodial parents.

• Survey data. A 15-month survey is currently being fielded and will be used to measure outcomes that cannot be assessed using administrative data. The survey will obtain data on jobs not covered in the UI records, participation in employment programs other than CEO, family outcomes, and receipt of public assistance and social services.

• Program data. These data provide information on each individual’s participation in the CEO components for which he or she is eligible, including data on NWP work, job coach and job developer appointments, participation in the fatherhood program, and, for control group participants, attendance in the Resource Room.

Characteristics of the Sample

Table 2.1 presents selected baseline characteristics of the research sample. Overall, the characteristics of the CEO study sample are similar to the national population of ex-prisoners. The vast majority of sample members are male (93 percent). Almost all are black or Latino (in national samples, nearly one-third are white, likely reflecting the difference in prison populations in New York City compared with the nation). Most of the sample members are over 30,19 similar to the average age of prisoners being released nationally, which is 34.

Almost half the research sample have at least one child under 18, although most do not live with any of their children. Of those with children under 18, fewer than one-fifth report that they have a formal child support order in place.

The process of obtaining affordable housing is complicated for a returning prisoner, because most do not have income from employment and are not eligible for many other forms of public assistance. In addition, current federally subsidized housing programs give public housing authorities the power to deny housing or terminate the leases of individuals with a history of drug use or criminal behavior. Thus, it is not surprising that only a very small proportion of the sample reside in a house or apartment that belongs to them or is rented in their name. Many live with friends or relatives or in some type of transitional housing.

 

The Enhanced Services for the Hard-to-Employ Demonstration
Table 2.1
Selected Baseline Characteristics, by Research Group
Center for Employment Opportunities

Characteristic			Program Group	Control Group	Total
Gender (%)**	Female		8.3	4.5	6.7
	Male		91.7	95.5	93.3
Age (%)	18 to 24 years		18.4	20.5	19.3
	25 to 30 years		25.5	24.9	25.3
	31 to 40 years		30.0	30.3	30.1
	41 years or older		26.1	24.3	25.3
Average age (years)			33.49	33.38	33.44
Citizenship (%)	Yes		80.2	76.0	78.5
	No		19.8	24.0	21.5
Race/ethnicity (%)	White, non-Hispanic		1.9	2.5	2.1
	Black, non-Hispanic		63.5	63.8	63.6
	Hispanic		31.2	30.5	30.9
	American Indian		0.7	0.5	0.6
	Other		2.7	2.7	2.7
Any children in household (%)	Yes		55.9	55.3	55.7
	No		44.1	44.7	44.3
Number of children under age 18a(%)*	None		54.6	52.1	53.5
	1 child		25.6	24.3	25.1
	2 children		13.0	11.9	12.5
	3 or more children		6.8	11.7	8.9
Average number of children*			0.77	0.90	0.83
Any children under age 18 in household (%)	Yes		15.0	14.6	14.8
	No		85.0	85.4	85.2
Education (%)	High school diploma		10.1	11.3	10.6
	GED		42.8	42.9	42.8
	Technical/associate's/2-year college		3.5	2.7	3.2
	4 years or more of college		0.8	0.6	0.7
	None of the above		42.7	42.5	42.6
High school diploma or GED certificate (%)	Yes		57.3	57.5	57.4
	No		42.7	42.5	42.6
Took courses in prison (%)	Yes		54.4	58.2	56.0
	No		45.6	41.8	44.0
Housing status (%)	Rent, not subsidized or public housing		7.5	9.0	8.1
	Rent, subsidized or public housing		8.3	6.7	7.6
	Lives at own home or apartment		2.2	3.4	2.7
	Lives with friends or relatives		59.7	56.4	58.3
	Transitional housing		10.8	11.6	11.1
	Emergency/temporary housing		2.8	4.5	3.5
	Homeless		1.6	0.7	1.2
	Other		7.2	7.8	7.4
Marital status (%)	Married, living with spouse		9.3	8.1	8.8
	Married, living away from spouse		8.0	7.3	7.7
	Unmarried, living with partner		21.6	21.5	21.6
	Single		61.2	63.1	62.0
Ordered to provide child support to a child under age 18 (%)	Yes		19.6	20.0	19.8
	No		80.4	80.0	80.2
Mandated to report to CEO (%)*	Yes		12.8	16.8	14.5
	No		87.2	83.2	85.5
Ever employed (%)	Yes		80.5	81.2	80.8
	No		19.5	18.8	19.2
Ever employed for 6 consecutive months by one employer (%)	Yes		59.8	63.9	61.5
	No		40.2	36.1	38.5
Received paychecks for at least 6 consecutive months from one employer (%)	Yes		57.0	60.5	58.4
	No		43.0	39.5	41.6
Among those who were ever employed:	Employed for 6 consecutive months by one employer (%)	Yes	75.1	79.7	77.0
		No	24.9	20.3	23.0
Sample size			568	409	977

SOURCE: CEO Baseline Information Form. NOTES: Random assignment began on January 9, 2004, and ended on October 21, 2005. In order to assess differences in characteristics across research groups, chi-square tests were used for categorical variables and analysis of variance (ANOVA) tests were used for continuous variables. Levels for statistically significant differences between the program and control groups are indicated as: *** = 1 percent; ** = 5 percent; * = 10 percent. a This category is missing a total of 16 sample members who stated that they have children but did not provide the children's ages.

 

Only about half the sample have completed a high school diploma or General Educational Development (GED) certificate and a very small proportion have any postsecondary education. Most do have at least some employment history. More than 80 percent report that they have worked. The majority of those who have worked report that they worked for a single employer for six consecutive months.

As expected with a random assignment research design, there were few differences in background characteristics between the two research groups. Moreover, the background characteristics of the CEO sample are similar to those of the larger population of ex-prisoners, making the findings of this study all the more important to policymakers.

Many experts believe that, the sooner after release a former prisoner receives needed supports, the more likely he or she is to have a successful transition. Indeed, the CEO model is designed for people who have just been released from prison. Criminal justice data (not shown), however, suggest that many of the participants in the study were not referred to CEO directly after their release but, instead, many months later. One can imagine any number of factors that might influence the point at which a parolee is actually referred to CEO. Perhaps participants unsuccessfully explored other avenues to finding employment before their parole officer referred them to CEO.

In contrast, several other subsets of the CEO population who are not part of the study almost always come to the program immediately after their release (in one program, orientation at CEO occurs the day after prisoners are released, and they begin the program the following Monday). The results for the regular parole population are obviously quite important and policy-relevant but, at this point, it is not clear how well they will represent the results for other groups served by CEO. The MDRC team will continue to investigate how the regular parole population compares with other CEO clients.

Early Findings from the Assessment

In July and August 2004, MDRC, assisted by the Urban Institute, conducted an early assessment of the CEO evaluation. This assessment was conducted seven to eight months after the start of random assignment to ensure, early on, that random assignment and the study design model were being implemented as planned. The assessment showed a high level of participation in CEO’s core program components among program group members. The data also indicated a clear difference in service receipt at CEO between the two research groups: As expected, members of the control group did not receive any of CEO’s core services.20 Such clear differences suggest that the evaluation will be a reliable test of CEO’s program.

Participation and Service Receipt

As part of the early assessment, participation rates in CEO’s main components were examined using NIGEL, CEO’s Management Information System. These results were updated as additional sample and follow-up data were available and have not changed dramatically since the assessment. The results, presented in Table 2.2, show rates of participation in CEO’s core program activities between the date of random assignment and December 31, 2005, for an early cohort of program group members who entered the program during the first year of the evaluation, between January 2004 and December 2004. By limiting the analysis to this early sample, it is possible to track participation patterns for a minimum period of one year after random assignment.

• Life skills. The first step in the CEO program is completion of the life-skills class. Nearly three-quarters of the NWP group completed the class, while only about 40 percent of the control group completed the shorter version that was designed for them.

• NWP work sites. After they complete the life-skills class, NWP participants are placed in transitional employment. As discussed above, placement at an NWP work site is the core of CEO’s program. Nearly all NWP group members who completed the life-skills class worked in NWP during the follow-up period, and, overall, two-thirds of the program group worked in NWP. Nearly half the program group worked at least one month at an NWP work site, with most of those working between two and three months. (Of course, working longer at an NWP site is not always better; an individual may be deemed job-ready and placed in a permanent job after a relatively short period in NWP.)

• Job coaching. The role of the CEO job coach is to assess and promote job readiness for each NWP participant. Job coaches teach the life-skills classes, track participants’ attendance and performance at NWP work sites, communicate with parole officers, and, where appropriate, issue warnings and suspensions. An important aspect of the job coach’s responsibilities is to identify and address any issues needed to make an individual ready for regular employment. Job coaches do this by meeting weekly with participants and by communicating with work site supervisors. Ultimately, it is the job coach who decides when a participant is deemed “job-ready” and begins to work with a job developer to seek a permanent job. For most of the study period, job coaches were also responsible for postplacement follow-up to promote job retention. As expected, the data from NIGEL show that a large majority of the participants who worked at an NWP work site attended at least one meeting with a job coach during the one-year follow-up period (81.6 percent — not shown in table); overall, 56.5 percent of the NWP group met with a job coach.

• Job development. Once an NWP participant is deemed “job-ready,” a CEO job developer conducts an initial assessment of the participant and begins the process of matching the participant with a permanent job. Job developers build relationships with employers in order to identify job openings and match participants with them. Job developers typically prepare participants for a particular interview and, in some cases, will accompany the participant to an interview. Job development is a particularly valuable service, because CEO has identified dozens of employers who are willing to hire individuals with a criminal record. Nearly all NWP group members who worked at an NWP site (nearly two-thirds of the program group overall) met with a job developer during the one-year period.

• Support services. CEO offers participants a number of other supports to assist them to find permanent employment, some directly and some through referrals to other organizations. These supports include assistance with clothing, including boots for use at NWP work sites; food allowances; assistance finding housing; transportation assistance; and other similar services.

 

The Enhanced Services for the Hard-to-Employ Demonstration
Table 2.2
Program Participation and Neighborhood Work Project (NWP) Employment Center for Employment Opportunities

Outcome (%)		NWP Group
Completed life skills		72.7
Received any job coaching servicesa		56.5
Received any job developer services		65.8
Ever worked in NWP		66.1
Weeks worked in NWPb	Never worked	33.9
	Less than 1 week	3.8
	1-4 weeks	15.4
	5-12 weeks	27.9
	13-24 weeks	15.5
	More than 24 weeks	3.6
Days between date of random assignment and NWP	0-7 days	67.8
	8-14 days	24.0
	More than 14 days	8.2
Sample size		303

SOURCE: MDRC calculations from CEO's Network for Information Gathering Evaluation and Learning (NIGEL) system. NOTES: This table reflects NWP employment through December 31, 2005. There were four control group members (1 percent) who worked in NWP during the follow-up period. NWP outcomes for these control group members are not shown in the table. The sample in this table is limited to program group clients who were randomly assigned between January 2004 and December 2004, allowing at least a 12-month follow-up period for everyone. Results in this table are weighted by month of random assignment. a The percentage who received any job coaching services may be underestimated due to inaccurate data entry. MDRC is working with CEO to obtain an accurate estimate of this outcome. b It is important to note that weeks worked may not be consecutive but may include a total of weeks worked after an individual's date of random assignment. This variable is created by taking total days worked in NWP and dividing by 4, because participants work four days per week in NWP and attend job coaching or other CEO services on the fifth day.

Recent Program Enhancements

CEO implemented a few “enhancements” to the program model midway through the study period.21 In upcoming reports, MDRC will examine whether there are any differences in the program’s effects for those who came to CEO after the enhancements were in place, compared with those who started the program before the enhancements.

• Bonuses. As part of the job retention component, CEO began to offer bonuses to participants who hold a non-NWP job for a certain period of time. These bonuses, called “Rapid Rewards,” are a series of noncash rewards (such as transportation passes) that CEO gives when participants present their pay stubs at various incremental time periods. Participants can earn over $500 in Rapid Rewards during the first 12 months of their employment.

• Passport to success. Work site supervisors are required to report daily about an individual’s performance on the job for that day. Job coaches then use these reports as a way of keeping apprised about a client’s performance and job readiness. For much of the study period, these daily reports were made on paper forms that were difficult for job coaches to keep track of for multiple participants across multiple job sites. CEO has now implemented the “passport to success” system, which requires every participant to carry a “passport” each day at the work site. The work site supervisors use the passport to report on participants’ performance for that day, and job coaches review the passports at the end of each week when participants meet with them.

• New employee work sites. These are special NWP work sites designed for new participants. They are more intensive than the regular NWP work sites and provide more job coaching assistance and specialized individual attention. Certain supervisors and job coaches are specially trained to work with individuals at these work sites.

Conclusions

Research has shown that stable employment can be a crucial component of successful reentry for former prisoners. Unfortunately, many ex-prisoners have attributes that make them less appealing to employers. Aside from the direct effect of being an ex-offender, many have low levels of education and previous work experience. Moreover, because prisons are overcrowded and expenses for corrections are mounting, most prisoners do not have opportunities to participate in meaningful job training programs while they are incarcerated. As a result, community-based programs have become an even more important resource for providing ex-prisoners with crucial employment-related services and job training and placement assistance. While there are community programs that provide supportive services to ex-prisoners, very few offer an immediate paid transitional job and permanent job placement services. CEO is one of the few established reentry programs operating on a large scale that offers these types of services.

Evidence from past evaluations of reentry programs showed that few, if any, strategies were effective at reducing recidivism and increasing employment among ex-prisoners. This created a widespread view that this population could not be helped. In recent years, however, experts have tried to change that view and have used research to show that some reentry strategies do show promise. To help policymakers understand better which strategies are most effective, experts and criminal justice officials agree that there is an urgent need to rigorously evaluate the most promising reentry strategies.

The Hard-to-Employ evaluation of CEO will be the first random assignment evaluation of a transitional employment program for ex-prisoners in many years. The findings of this study will provide critical answers to many questions left unanswered by previous studies. Future publications from this study will evaluate whether a strong transitional employment model can increase employment and break the cycle of reincarceration among former prisoners.

---

1 NGA Center for Best Practices (2005). (back)

2 Travis, Solomon, and Waul (2001). (back)

3 NGA Center for Best Practices (2005). (back)

4 U.S. Department of Justice (2004). (back)

5 Roman and Travis (2004). (back)

6 Re-entry Policy Council (2006). (back)

7 Hirsch et al. (2002). (back)

8 Visher, LaVigne, and Travis (2004). (back)

9 Travis and Waul (2004). (back)

10 Re-entry Policy Council (2006). (back)

11 Harer (1994); Uggen (2000). (back)

12 Holzer (1996); Holzer, Raphael, and Stoll (2002). (back)

13 Holzer, Raphael, and Stoll (2002). (back)

14 When the study began in 2004, the minimum wage was $5.15 per hour. In 2005, it increased to $6.00 per hour. The current minimum wage in 2007 is $7.15 per hour. (back)

15 Control group members who worked diligently in the Resource Room for three months but were unable to find employment on their own were offered CEO’s job placement services. (back)

16 For programmatic and logistical reasons, it was decided that new Resource Room classes should comprise four to 12 clients. (back)

17 Shock Incarceration is New York State’s boot camp program. (back)

18 Toward the end of the sample intake period, some Queensboro referrals were accepted into the study. (back)

19 Note that the population in this study is considerably older than CEO’s other participants, many of whom are between the ages of 18 and 25. (back)

20 The assessment showed that no one in the Resource Room group worked in an NWP position; however, recent data collected on NWP participation showed that there were four control group clients who worked in NWP at some point during their follow-up period. It is unlikely that four sample members will change the overall results of the study. Nonetheless, these individuals will be flagged for the impact analysis, and MDRC will examine whether their outcomes have any effect on CEO’s impacts. (back)

21 CEO also began to operate a Young Adult Program during the study period that provided specialized services specifically designed to meet the needs of younger participants. Only a small proportion of clients in the study received these services because of timing as well as age restrictions. (back)

 

Table of Contents | Previous | Next