An evaluation of the effect of correctional education programs on post-release recidivism and employment: an empirical study in Indiana.

Nally, John ; Lockwood, Susan ; Knutson, Katie 等

Introduction

Correctional education is a crucial part of the correctional system in terms of the prison operation and the educational remedy for prisoner reentry to the community. Since the 1980s, the prison population has dramatically increased and prisoners are likely uneducated and unemployed prior to admission to the prison. Consequently, in recent years, the demand for correctional education has steadily increased but the funding for correctional education has been systematically decreased. In the recent recessionary period, many states have significantly reduced education budgets and/or eliminated education programs in order to resolve budget deficits. Even though the budget for prison education programs is relatively small in the overall budget of the Department of Correction, the public's sentiments and the policy-makers' perceptions of such publicly-perceived "free" education for prison inmates have turned intensely negative. For example, Congress passed a 1994 amendment to exclude prison inmates from receiving federal funding (i.e., the Pell Grant) for post-secondary education programs at correctional facilities. Such "tough on crime" measures do not reduce the overall prison population but aggravate the prison overcrowding problems across the nation.

On the other hand, researchers (Batiuk, et al., 2005; Chappell, 2004; Mercer, 2009; Owens, 2009) find that the recidivism rate among offenders who have participated in post-secondary education programs during incarceration is significantly lower than those offenders who have not. In other words, offenders are likely to be employed after release and less likely to return to the prison if they have a higher education. Additionally, the benefits of correctional education programs, at its core, are frequently measured by the reduction of the recidivism rate among post-release offenders. However, previous studies exclusively focused on the released offenders, without a comparison group, to assess the effect of the correctional education on recidivism. Furthermore, previous studies were largely insufficient in measuring the correlation between correctional education and recidivism due to the lack of post-release employment information among those released offenders. Unlike previous studies, this study has included the offender's employment data, if employed, to adequately evaluate the relationship between correctional education and both post-release employment and recidivism in the study group and the comparison group.

Correctional Education and Impacts

Correctional education has become deeply embedded in American correctional systems due to a flux of uneducated or undereducated inmates. It is a common phenomenon that uneducated or undereducated inmates are likely to return to prison because they are less likely to find a job upon release. For example, Vacca (2004) finds that a notable number of released offenders are unemployed because they do not have sufficient education and professional skills to meet with job demands in a variety of industry sectors. According to U.S. Bureau of Justice Statistics (Harlow, 2003), almost 68 percent of prison inmates have not graduated from high school and most individuals in that group are racial minorities. Undoubtedly, correctional education and professional training programs at correctional facilities have become the primary educational resources for those uneducated or undereducated inmates to strengthen their educational competency and job skills.

Across the nation, a variety of correctional education programs have been utilized in correctional facilities to educate incarcerated inmates. Since a disproportionate number of inmates are considered to be functionally illiterate, a significant portion of educational resources at the state level support those correctional education programs such as Adult Basic Education (ABE) or General Equivalency Diploma (GED) programs. For some, post-secondary college programs have been added to enhance the level of education among incarcerated inmates. As the GED or high school diploma has become insufficient to meet the demands of a variety of fast-evolving, technology-based industry sectors, correctional education programs have been focused on the enhancement of employability for offenders through the development of a variety of certificate-based, skill-oriented programs in the post-secondary education curriculum.

For example, the Education Division of the Indiana Department of Correction (IDOC) has collaborated with seven Indiana institutions of Higher Education to establish up-to-date post-secondary job-oriented certificate programs with the objective of increasing the employability of offenders and reducing the recidivism rate. The IDOC Education Division has carefully allocated federal and state grant funding to meet a high demand of educational needs among Indiana inmates. The identified need to shift the focus from "Liberal Studies" or "General Studies" degrees resulted in more degree programs aligned to Indiana's employment needs through 2016. In addition, several job-oriented certificate programs, such as "Coal Miner Training" and "Certified Bookkeeper" were established using federal funding. This shift in job-specific certifications will be accelerated as Indiana seeks to effectively utilize limited funding to education its inmates.

There are numerous studies of the benefits of correctional education programs in terms of reducing recidivism and decreasing the cost of incarceration (Blackburn, 1981; Burke and Vivian, 2001; Cecil et al., 2000; Fabelo, 2002; Gordon and Weldon, 2003; Hrabowski and Robbi, 2002; Matsuyama and Prell, 2010; Nuttall et al., 2003; Steurer et al, 2001; Taylor, 1992; Vacca, 2004; Ward, 2009). Researchers generally conclude that correctional education has effectively reduced the recidivism rate among released offenders and decreased the over-all cost of incarceration. Specifically, the recidivism rate is significantly decreased if offenders have attained a higher level of education during incarceration.

Even though recidivism has been defined in a variety of measures, the recidivism rate has been commonly used to measure the effectiveness of correction educational programs. However, there is no universal consensus on measuring the success of correction educational programs while employing the recidivism rate as the post-release outcome measure (Batiuk et al., 1997; Fabelo, 2002; Gordon and Weldon, 2003; Jancic, 1998; Nuttall et al., 2003; Stevens and Ward, 1997). The main argument is that recidivism measurement is frequently perceived as arbitrary (Gehring, 2000) or methodologically inadequate (Cecil et al., 2000; Hull et al., 2000; Lewis, 2006). Nevertheless, recidivism is the highly-accepted outcome measure due to mandates from both state and federal funding agencies (Linton, 2007). Even though the success of correctional education may be largely measured by the recidivism rate, it is important to recognize that post-release employment is an important indicator of the success of correctional education. This study has included crucial employment-related information to evaluate the effect of correctional education on post-release recidivism.

Methodology

Data and Data Collection

The Education Division of the Indiana Department of Correction (IDOC), in cooperation with the IDOC Research and Planning Division, has continuously modified and updated a dataset of released offenders in order to assess a variety of post-release measures such as recidivism or employment. With assistance from the IDOC Research and Planning Division, the current dataset for assessing post-release recidivism and employment contains several important factors such as offender demographical characteristics (i.e., gender, race, age, and education), legal characteristics of offenders (e.g., legal reason for return to IDOC, or recidivism status), and employment-related characteristics of offenders (e.g., job classification or income).

It is important to mention that the IDOC Education Division has continuously maintained a collaborative relationship with the Indiana Department of Workforce Development (DWD) for employment-related data or information among released offenders. Since 2008, the IDOC Education Division has collaborated with the Indiana Department of Workforce (DWD) to systematically document employment-related information among a cohort of 6,561 offenders who were released from IDOC custody in 2005. Such a collaborative effort between the IDOC Education Division and the Indiana Department of Workforce Development (DWD) has generated crucial information for analyzing the effect of an offender's level of education on post-release employment. In 2011, the DWD updated the post-release employment information for all offenders in this study to include the period of the first quarter of 2008 through the second quarter of 2009.

Study Group and Comparison Group

In order to effectively examine the effect of correctional education on post-release employment and recidivism, the IDOC Education Division has established a study group and a comparison group to evaluate outcome measures (e.g., post-release recidivism). The study group contains 1,077 Indiana offenders who received federal funding from the U.S. Department of Education in the period of 2002-2009 and were released from IDOC custody during that time period. It is important to mention that some offenders received the federal funding to attend correctional education programs but did not complete the program requirements due to early release from IDOC.

On the other hand, the comparison group contains 1,078 Indiana offenders selected from a cohort of 6,561 offenders who were released from IDOC in 2005. (1) The criteria of the selection of the comparison group primarily focuses on the size of sample, race of offender, education level of offender, and whether or not the offender had received the federal funding for his/her education at Indiana correctional facilities. No offender in the comparison group had received the federal funding to attend any correctional education program during incarceration. Most importantly, offenders in the comparison group are randomly selected once they have met with the above-mentioned conditions.

Outcome Measures

By evaluating the outcome measures between the study group and the comparison group, the primary focuses of this study are to examine: (1) the effect of correctional education on offender's recidivism, (2) the effect of correctional education on offender's employment, and (3) the interrelationship of offender's education, employment and recidivism. Nonetheless, the principal dependent measure in this study is to examine the similarity or difference between the study group and the comparison group to further examine the effect of correctional education on post-release recidivism. This study will further examine the post-release employment and recidivism among released offenders with a different level of education.

Findings

Table 1 provides a detailed description of the study group and the comparison group in terms of offender's demographics, education, recidivism, and post-release employment status. Among offenders in the study group (n=1,077), results of this study show that the study group consists of 156 (14.5%) female and 921 (85.5%) male offenders; 688 (63.9%) offenders are Caucasian, 349 (32.4%) offenders are African American, 32 (3.0%) offenders are Hispanic, 3 (0.3%) offenders are Asian or Pacific Islanders, and 5 (0.5%) offenders whose race are unknown; 783 (72.7%) offenders are in the age range of 20-29 years old, 287 (26.6%) offenders are in the age range of 30-39 years old, 4 (0.4%) offenders are in the age range of 40-49 years old, and 3 (0.3%) offenders are 50 years old or older.

Table 1. Descriptive Statistics of the Study Group and the Comparison
Group

Variable Coding Study Group Comparison
 (n=1077) Group
 (n=1078)

Offender Female 156 / 167 / 15.5%
Gender 14.5%

 Male 921 / 911 / 84.5%
 85.5%

Offender Caucasian 688 / 712 / 66.0%
Race 63.9%

 African American 349 / 366 / 34.0%
 32.4%

 Hispanic 32 / 3.0% 0 / 0.0%

 Asian/Pacific 3 / 0.3% 0 / 0.0%
 Islander

 Unknown 5 / 0.5% 0 / 0.0%

Offender 20-29 years old 783 / 184 / 17.1%
Age 72.7%

 30-39 years old 287 / 403 / 37.4%
 26.6%

 40-49 years old 4 / 0.4% 324 / 30.1%

 50 years old or 3 / 0.3% 167 / 15.5%
 above

Offender below high school 25 / 2.3% 232 / 21.5%
Education

 high school or 881 / 846 / 78.5%
 GED 81.8%

 college education 171 / 0 / 0.0%
 15.9%

Legal Reason parole violation 129 / 246 / 33.7%
for return1 40.4%

 probation 98 / 30.6% 179 / 24.5%
 violation

 new commitment 83 / 25.9% 262 / 35.8%

 CTP 10 / 3.1% 44 / 6.0%
 return/violation

Recidivism non-recidivist 757 / 347 / 32.2%
Status offender 70.3%

 recidivist 320 / 731 / 67.8%
 offender 29.7%

Note 1: There were 320 recidivist offender in the study group
and 731 recidivist offenders in the comparison group.

In the study group, there are 881 (81.8%) offenders with a high school diploma or GED, 171 (15.9%) offenders with a post-secondary education, and 25 (2.3%) offenders have not completed high school or GED prior to release from IDOC custody. The IDOC Education Division has systematically documented offender's educational endeavors at IDOC correctional facilities. The Education Division has found that, in the period of 2002-2009, 870 (80.8%) offenders in the study group have received the federal funding once, 188 (17.5%) offenders have received the federal funding twice, 17 (1.6%) offenders have received the federal funding three times, and 2 (0.2%) offenders have received the federal funding four times. Meanwhile, 45 offenders who have received the federal funding have not completed the education program at IDOC correctional facilities due to early release.

In the study group, results of this study reveal that 320 (29.7%) offenders are recidivist offenders and 757 (70.3%) offenders are not recidivist offenders. The recidivism rate is 29.7 percent in the study group. Among 320 recidivist offenders in the study group, 129 (40.4%) offenders were returned to IDOC due to parole violation, 98 (30.6%) offenders were returned to IDOC due to probation violation, 83 (25.9%) offenders were returned to IDOC due to committing a new crime, and 10 (3.1%) offenders were returned to IDOC due to a violation of Community Transition Program (CTP). A further examination of recidivist offenders in the study group reveals that, among 320 recidivist offenders, 125 (39.1%) offenders were returned to IDOC within a year after release, 101 (31.6%) offenders were returned to IDOC within 1-2 years after release, 54 (16.9%) offenders were returned to IDOC within 2-3 years after release, and 40 (12.5%) offenders were returned to IDOC after 3 years or more since release.

Table 1 also illustrates offender's characteristics in the comparison group. Results of this study reveal that the comparison group consists of 167 (15.5%) female and 911 (84.5%) male offenders; 712 (66.0%) offenders are Caucasian and 366 (34.0%) offenders are African American; 184 (17.1%) offenders are in the age range of 20-29 years old, 403 (37.4%) offenders are in the age range of 30-39 years old, 324 (30.1%) offenders are in the age range of 40-49 years old, and 167 (15.5%) offenders are 50 years old or older.

In regard to education in the comparison group, this study reveals that there are 232 (21.5%) offenders with an education below high school and 846 (78.5%) offenders with a high school diploma or GED. It is important to note that all offenders in the comparison group in this study have never received federal funding to attend any correctional education program in IDOC correctional facilities.

This study also reveals that 731 (67.8%) offenders in the comparison group are recidivist offenders and 347 (32.2%) offenders are not recidivist offenders. The recidivism rate is 67.8 percent in the comparison group. A further examination of 731 recidivist offenders in the comparison group reveals that 246 (33.7%) offenders were returned to IDOC due to parole violation, 179 (24.5%) offenders were returned to IDOC due to probation violation, 262 (35.8%) offenders were returned to IDOC due to committing a new crime, and 44 (6.0%) offenders were returned to IDOC due to a violation of Community Transition Program (CTP). Furthermore, among 731 recidivist offenders in the comparison group, 548 (75.0%) offenders have returned to IDOC within a year after release, 182 (24.9%) offenders have returned to IDOC within 1-2 years after release, and 1 (0.1%) offender has returned to IDOC within 2-3 years after release. In other words, all recidivist offenders in the comparison group returned to IDOC custody within 3 years since release in 2005.

Table 2 illustrates the employment status among released offenders in the study period of the first quarter of 2008 through the second quarter of 2009 (i.e., 2008Q1-2009Q2). In regard to employment status among released offenders, this study's results reveal that 303 (28.1%) offenders in the study group and 400 (37.1%) offenders in the comparison group have been employed at least one quarter in the period of 2008 Q1-2009 Q2 since release from IDOC. However, a further examination of offender's employment sustainability indicates that more than 60 percent of employed offenders in the comparison group, but only about 38 percent in the study group, have been employed no more than 2 quarters out of 6 quarters in this study period (2008Q1-2009Q2). About 47 percent of employed offenders in the study group have been employed at least 4 quarters (1 year) or more, but only 28 percent of employed offenders in the comparison group have extended their employment at least one year or more in this study period.

Table 2. Employment Status among Released Offenders, 2008Q1 through
2009Q2

Variable Coding Study Group Comparison
 (n=1077) Group
 (n=1078)

Employment employed 303 / 28.1% 400 /
Status1 37.1%

(2008Q1 thru unemployed 774 / 71.9% 678 /
2009Q2) 62.9%

Number of one quarter 51 / 16.8% 131 /
Quarter2 32.8%

Been Employed two quarters 64 / 21.1% 110 /
 27.5%

 three quarters 46 / 15.2% 47 / 11.8%

 four quarters 89 / 29.4% 48 / 12.0%

 five quarters 27 / 8.9% 27 / 6.8%

 six quarters 26 / 8.6% 37 / 9.2%

Employment agriculture/mining/etc. 9 / 0.9% 0 / 0.0%
Sector3

(NAICS construction 120 / 12.5% 118 /
classification) 11.3%

 manufacturing 206 / 21.4% 171 /
 16.3%

 wholesale or retail 150 / 15.6% 123 /
 11.7%

 temporary agencies 160 / 16.6% 232 /
 22.2%

 lodging or food service 228 / 23.7% 253 /
 24.2%

 repair & maintenance 23 / 2.4% 34 / 3.3%

 other employments 67 / 6.9% 115 /
 11.0%

Wage 2008Q14 under $1,000 12 / 15.4% 81 / 43.3%

 between $1,000-$1,999 7 / 9.0% 33 / 17.6%

 between $2,000-$2,999 17 / 21.8% 27 / 14.4%

 between $3,000-$3,999 13 / 16.7% 13 / 7.0%

 between $4,000-$4,999 7 / 9.0% 8 / 4.3%

 between $5,000-$5,999 10 / 12.8% 7 / 3.7%

 between $6,000-$6,999 3 / 3.8% 7 / 3.7%

 $7,000 or above 9 / 11.5% 11 / 5.9%

Wage 2008Q2 under $1,000 9 / 10.5% 95 / 44.4%

 between $1,000-$1,999 13 / 15.1% 26 / 12.1%

 between $2,000-$2,999 13 / 15.1% 25 / 11.7%

 between $3,000-$3,999 12 / 14.0% 14 / 6.5%

 between $4,000-$4,999 11 / 12.8% 13 / 6.1%

 between $5,000-$5,999 12 / 14.0% 14 / 6.5%

 between $6,000-$6,999 7 / 8.1% 7 / 3.3%

 $7,000 or above 9 / 10.5% 20 / 9.3%

Wage 2008Q3 under$1,000 44 / 19.8% 72 / 36.9%

 between $1,000-$1,999 38 / 17.1% 40 / 20.5%

 between $2,000-$2,999 36 / 16.2% 18 / 9.2%

 between $3,000-$3,999 24 / 10.8% 11 / 5.6%

 between $4,000-$4,999 19 / 8.6% 16 / 8.2%

 between $5,000-$5,999 18 / 8.1% 14 / 7.2%

 between $6,000-$6,999 14 / 6.3% 6 / 3.1%

 $7,000 or above 29 / 13.1% 18 / 9.2%

Wage 2008Q4 under$1,000 48 / 23.2% 58 / 33.9%

 between $1,000-$1,999 25 / 12.1% 30 / 17.5%

 between $2,000-$2,999 29 / 14.0% 21 / 12.3%

 between $3,000-$3,999 23 / 11.1% 19 / 11.1%

 between $4,000-$4,999 22 / 10.6% 13 / 7.6%

 between $5,000-$5,999 17 / 8.2% 7 / 4.1%

 between $6,000-$6,999 12 / 5.8% 8 / 4.7%

 $7,000 or above 31 / 15.0% 15 / 8.8%

Wage 2009Q1 under$1,000 41 / 22.4% 48 / 35.8%

 between $1,000-$1,999 38 / 20.8% 24 / 17.9%

 between $2,000-$2,999 19 / 10.4% 15 / 11.2%

 between $3,000-$3,999 26 / 14.2% 16 / 11.9%

 between $4,000-$4,999 18 / 9.9% 8 / 6.0%

 between $5,000-$5,999 13 / 7.1% 9 / 6.7%

 between $6,000-$6,999 9 / 4.9% 1 / 0.7%

 $7,000 or above 18 / 9.9% 13 / 9.7%

Wage 2009Q2 under$1,000 36 / 19.3% 58 / 40.6%

 between $1,000-$1,999 27 / 14.4% 24 / 16.8%

 between $2,000-$2,999 26 / 13.9% 16 / 11.2%

 between $3,000-$3,999 25 / 13.4% 10 / 7.0%

 between $4,000-$4,999 21 / 11.2% 11 / 7.7%

 between $5,000-$5,999 15 / 8.0% 4 / 2.8%

 between $6,000-$6,999 16 / 8.6% 6 / 4.2%

 $7,000 or above 21 / 11.2% 14 / 9.8%

Note 1: An employed offender is an offender employed at least one
quarter since release.

Note 2: There are 303 employed offenders in the study group and
400 employed offenders in the comparison group.

Note 3: All employments that offenders had been employed in any
given quarter in the study period of 2008Q1-2009Q2. Meanwhile,
some offenders had multiple employments in any given quarter.

Note 4: There were a different number of employed offenders in
any given quarter in both the study group and the comparison group.

This study also reveals that 303 employed offenders in the study group and 400 employed offenders in the comparison group have been employed in a variety of job sectors in this study period (2008Q1-2009Q2). According to the Indiana Department of Workforce Development (DWD), 303 offenders in the study group, who have been employed at least one quarter in the study period (2008Q1-2009Q2), would likely find a job in the following job sectors (in order): (1) lodging and food services (23.7%), (2) manufacturing (21.4%), (3) temporary help services agencies (16.6%), (4) wholesale or retail (15.6%), (5) construction (12.5%), (6) other employments (6.9%), (7) repair and maintenance (2.4%), and (8) agriculture, mining, and so on (0.9%). On the other hand, 400 offenders in the comparison group, who have been employed at least one quarter in the study period (2008Q1-2009Q2), would likely find a job in the following job sectors (in order): (1) lodging and food services (24.2%), (2) temporary help services agencies (22.2%), (3) manufacturing (16.3%), (4) wholesale or retail (11.7%), (5) construction (11.3%), (6) other employments (11.0%), (7) repair and maintenance (3.3%), and (8) agriculture, mining, and so on (0.0%).

Even though 37.1 percent (n=400) of offenders in the comparison group but only 28.1 percent (n=303) offenders in the study group have been employed one quarter in the study period of 2008Q1-2009Q2 since release from IDOC custody, results of this study reveal that there are a significantly high number of such employed offenders in the comparison group, when contrasted with offenders in the study group, who have a quarterly income below $1,000 dollars. For example, in the first quarter of 2008 (see Table 2, Wage2008Q1), there are 12 (15.4%) of 78 employed offenders in the study group with an income under $1,000; 7 (9.0%) employed offenders with an income between $1,000 and $1,999; 17 (21.8%) employed offenders with an income between $2,000 and $2,999; 13 (16.7%) employed offenders with an income between $3,000 and $3,999; 7 (9.0%) employed offenders with an income between $4,000 and $4,999; 10 (12.8%) employed offenders with an income between $5,000 and $5,999; 3 (3.8%) employed offenders had an income between $6,000 and $6,999; and 9 (11.5%) employed offenders had an income of $7,000 or above. On the other hand, in 2008Q1, there are 81 (43.3%) of 187 employed offenders in the comparison group with an income under $1,000; 33 (17.6%) employed offenders with an income between $1,000 and $1,999; 27 (14.4%) employed offenders with an income between $2,000 and $2,999; 13 (7.0%) employed offenders with an income between $3,000 and $3,999; 8 (4.3%) employed offenders with an income between $4,000 and $4,999; 7 (3.7%) employed offenders with an income between $5,000 and $5,999; 7 (3.7%) employed offenders with an income between $6,000 and $6,999; and 11 (5.9%) employed offenders with an income of $7,000 or higher. In short, this study has clearly indicated that offenders in the comparison group, who have a lower education, if employed, are likely to be employed in a variety of labor-intensive and minimum-wage jobs and job sustainability may be challenging.

Table 3 illustrates bi-variate analyses of recidivism with offender's education in both the study group and the comparison group. The recidivism rate among all released offenders in the study group is 29.7 percent. However, the recidivism rate among offenders who have an education below high school is 32.0 percent, 29.7 percent among offenders who have a high school diploma or GED, and 29.2 percent among offenders who have a college education. On the contrary, the recidivism rate among all released offenders in the comparison group is 67.8 percent. Furthermore, the recidivism rate is 82.3 percent among offenders who have an education below high school but only 63.8 percent among offenders who have a high school diploma or GED in the comparison group. In other words, results of this study reveal that less-educated offenders are likely to become recidivist offenders after release from IDOC custody.

Table 3. Bi-Variate Analysis of Offender's Education with Recidivism
and Employment

Variable Study Comparison
 Group Group

 Below HS High Sch. College Below HS
 (n=25) (n=881) (n=171) (n=232)

Recidivism
Status:

Non-Recidivist 17/68.0% 619/70.3% 121/70.8% 41/17.7%
Offender

Recidivist 8/32.0% 262/29.7% 50/29.2% 191/82.3%
Offender

Employment
Status:

Never been 17/68.0% 642/72.9% 115/67.3% 165/71.1%
employed

Employed 1 2/8.0% 39/4.4% 10/5.8% 27/11.6%
quarter

Employed 2 2/8.0% 55/6.2% 7/4.1% 17/7.3%
quarters

Employed 3 2/8.0% 32/3.6% 12/7.0% 7/3.0%
quarters

Employed 4 1/4.0% 70/7.9% 18/10.5% 6/2.6%
quarters

Employed 5 0/0.0% 23/2.6% 4/2.3% 2/0.9%
quarters

Employed 6 1/4.0% 20/2.3% 5/2.9% 8/3.4%
quarters

Variable

 High Sch.
 (n=846)

Recidivism
Status:

Non-Recidivist 306/36.2%
Offender

Recidivist 504/63.8%
Offender

Employment
Status:

Never been 513/60.6%
employed

Employed 1 104/12.3%
quarter

Employed 2 93/11.0%
quarters

Employed 3 40/4.7%
quarters

Employed 4 42/5.0%
quarters

Employed 5 25/3.0%
quarters

Employed 6 29/3.4%
quarters

Table 3 also provides detailed information on post-release employment among released offenders with different level of education in both the study group and the comparison group. Regardless of offender's level of education, results of this study reveal that offenders in both the study group and the comparison group have a high unemployment rate in the study period (2008Q1-2009Q2). It is important to mention that the recent national recession officially started in December of 2007 and ended in December of 2008. This study has clearly indicated that released offenders have encountered more challenges in finding a job during the recessionary period. Consequently, the unemployment rate among released offenders is significantly higher than that of the general population. This study also finds those offenders with a higher level of education are likely to be employed for a longer period of time.

The results of logistic regression analyses, as Table 4 indicates (see, All Offenders equation), shows the effect of correctional education programs on post-release recidivism is statistically, but negatively, significant, while controlling of other variables. Specifically, offenders who have participated in the correctional education programs during incarceration at Indiana correctional facilities are less likely to be recidivist offenders than those offenders who have not attended any correctional education program during incarceration after release from IDOC custody. Meanwhile, this study's results also show the effect of offender's education on recidivism is statistically, but negatively, significant. In other words, offenders who have a lower education (those offenders have not completed high school, in particular) are likely to be recidivist offenders after release from IDOC custody than those offenders who have had a higher education (e.g., a college degree or high school diploma). However, this study has found that employment bears no significant impact on post-release recidivism in the recessionary period among those released offenders.

Table 4. Logistic Multiple Regression on Recidivism

Variable All Study Group Comparison
(n=2155) Offenders (n=1078) Group
 (n=1077)

Correctional -1.304 ** n/a n/a
Education
Programs

Offender Race .246 * .194 .238

Offender .072 -.069 .130
Gender

Offender Age . 013 .121 ** .008

Offender -.526 ** -.113 -1.038 **
Education

Employment -.161 -.801 ** .193
Status

Constant 2.173 -4.094 ** 1.813 **

Notes: "*" at <.01 and "**" at <.001. Offenders who have
attended correctional education programs are coded as "1" and
offenders who have not participated in any correctional education
program during incarceration are coded as "0." In the logistic
multiple regression analyses, only African American offenders and
Caucasian offenders are included in the present analyses. Offenders
are regarded as "employed," if they have been employed at least
one quarter in the study period (2008Q1-2009Q2).

In the study group, as Table 4 indicates (see, Study Group equation), results of this study, while controlling of other variables, show the effect of employment on recidivism was statistically, but negatively, significant. In other words, this study's results indicate that offenders are less likely to be recidivist offenders if they have been employed after release from IDOC custody. Meanwhile, results of this study also show an offender's age to be statistically and significantly correlated with recidivism. This study's results indicate that older offenders, rather than younger offenders in the study group, are likely to be recidivist offenders. A further examination reveals that such recidivist "older" offenders are likely to be unemployed since release from IDOC. Nevertheless, this study also reveals that the effect of education on recidivism is not statistically significant, while controlling of other factors. Regardless of an offender's level of education in the study group, the recidivism rate is significantly lower than that in the comparison group. Furthermore, an offender's race and gender bear no effect on recidivism.

In the comparison group, as Table 4 indicates (see, Comparison Group equation), results of this study, while controlling of other variables, show recidivism is statistically, but negatively, correlated with an offender's education. Specifically, offenders who have not completed high school are more likely to be recidivist offenders than those offenders who have a high school diploma or GED. This study also reveals that there is no significant difference in terms of the recidivism rate between employed offenders and unemployed offenders in the comparison group. Furthermore, results of this study showed that an offender's gender, race, and age had no significant effect on recidivism among released offenders in the comparison group.

Discussion

The most important finding in this study demonstrates that offenders are less likely to be recidivist offenders if they have participated in correctional education programs in IDOC facilities during incarceration. The recidivism rate among offenders in the comparison group who have not participated in correctional education programs during incarceration reached 67.8 percent in the study period of 2008Q1-2009Q2. All else being equal, this study shows, as statistics illustrate in All Offenders equation in Table 4, an offender who has not attended correctional education programs during incarceration is approximately 3.7 times more likely to become a recidivist offender after release from IDOC custody, while compared with an offender who has participated in correctional education programs during incarceration. (2) In other words, this study has clearly shown that offenders who have participated in correctional education programs during incarceration to enhance their education and/or professional job skills are less likely to return to IDOC custody after release. Specifically, the correctional education programs, as this study's results indicate, become an important contributing factor in reducing the post-release recidivism among those released offenders in Indiana.

The effect of education has also become a significant contributing factor to post-release recidivism among offenders who have not attended correctional education programs prior to release from IDOC custody. All else being equal, this study shows, as statistics illustrate in the Comparison Group equation in Table 4, that an offender who has not completed high school is almost 2.8 times more likely to become a recidivist offender while compared with an offender who has a high school diploma or GED. (3) On the other hand, the recidivism rate among offenders who have participated in correctional education programs at IDOC correctional facilities during incarceration is only 29.7 percent in the same study period. This study also clearly indicates that less-educated offenders are likely to become recidivist offenders after release from IDOC custody.

Additionally, this study reveals that a significant number of released offenders in both the study group and the comparison group have had difficulties in finding a job in this study period (2008Q1 through 2009Q2) which is officially recognized as the recessionary period. Regardless of an offender's education, the unemployment rate among released offenders in the recessionary period is generally expected to be higher than the general population due to a variety of contributing factors such as the scarcity of job opportunity, criminal background, inadequate education, or lack of basic job-related skills. Nevertheless, this study has showed that the employment rate among released offenders in the study group who have participated in correctional education programs at IDOC correctional facilities, has significantly improved from 7.2 percent in Quarter 1 of 2008 to 17.4 percent in Quarter 2 of 2009 in this study period. On the contrary, the employment rate among offenders in the comparison group who have not participated in correctional education programs during incarceration has declined from 17.3 percent in Quarter 1 of 2008 to 13.3 percent in Quarter 2 of 2009 in this study period. This study has revealed that offenders in both the study group and the comparison group are likely to be employed in the labor-intensive and low-wage job sectors such as "temporary help services agencies" or "food services or lodging." Shapiro (2011) finds that 60 percent of newly-created employment in 2010 were low-wage jobs in sectors such as, "temporary help services," "leisure and hospitality," and "retail trade."

One striking finding in this study is that employed offenders in the study group who have participated in correctional education programs during incarceration are likely to have a higher quarterly income than those employed offenders in the comparison group who have not participated in correctional education programs at IDOC facilities prior to release from custody. In other words, educated offenders are likely to have earned a better wage if they are employed after release. Among those less-educated offenders in the comparison group, in particular, results of this study reveal that there is a significantly higher number of "marginally-employed" offenders, who have been employed but whose quarterly income is under $1,000 (or under $334 per month). For example, there are 43.3 percent (n=81) of in a total of 187 offenders in the comparison group who have been employed in the first quarter of 2008 but they have a quarterly income under $1,000. Almost 40 percent of 81 "marginally-employed" offenders have been employed by temporary help services agencies. The recidivism rate among those "marginally-employed" offenders is 65.4 percent. Results of this study clearly indicate that it is extremely difficult for those "marginally-employed" offenders to be financially independent in order to prevent themselves from becoming involved in criminal activity. A similar pattern has been persistent in every quarter of this study period (2008Q1-2009Q2).

According to the Bureau of Labor Statistics (U.S. Department of Labor, 2011 & 2010), there are 10.4 million adults in 2009 and 8.9 million adults in 2008 among the so-called "working poor." U.S. Department of Labor (2011) defines the "working poor" as persons who spent at least 27 weeks in the labor force (that is, working or looking for work) but whose incomes still fell below the official poverty level. Furthermore, minorities such as African Americans and Hispanics are almost twice as likely as their Caucasian counterparts to be among the working poor. The labor statistics have also revealed that attaining a higher education will enhance the chance to obtain full-time employment which will directly diminish the proportion of the working poor in the labor force. The U.S. Department of Labor (2011, p. 2) stated, "Of all the people in the labor force for 27 weeks or more in 2009, those with less than a high school diploma had a higher working-poor rate (20.3 percent) than did high school graduates with no college (8.8 percent). Workers with an associate's degree and those with a bachelor's degree or higher had the lowest working-poor rate: 4.7 percent and 2.1 percent, respectively." The "marginally-employed" offenders in this study exhibit similar characteristics to those of the "working poor", having low-wage, temporary jobs with an income below the poverty level. Consequently, the "marginally-employed" offenders are likely to become recidivist offenders after release due to the fact that they lack the financial resources to sustain themselves in the community.

This study also finds that the unemployment rate among released offenders is significantly higher than the general population. Undoubtedly, released offenders, with criminal background, are likely to encounter increased barriers in seeking a job during a recessionary period. Furthermore, those offenders are likely to be released under legally-mandated conditions of probation or parole which required them to seek and maintain gainful employment or be enrolled in a course of study or vocational training. According to the Bureau of Labor Statistics (BLS), there were approximately 2.1 million "marginally attached" workers in the first quarter of 2009 (U.S. Department of Labor, 2009). The U.S. Department of Labor (2009) defines a "marginally attached" worker as "someone who is currently not in the labor force but he/she wants full-time work and has actively looked for a job sometime in the past 12 months." A "marginally-attached" worker is not considered to be either employed or unemployed, so he is not included in the "official" unemployment number that is released by the US government every month (Kodrzycki, 2000). Such "marginally-attached" offenders present a unique challenge to analyze the effect of correctional education programs on post-release employment because it is extremely difficult to obtain any crucial information about whether or not such "marginally-attached" released offenders are actively looking for work or remain unemployed due to their educational deficiency. A further study on such "marginally-attached" released offenders is needed in the near future.

Conclusion

This study's results indicate that correctional education may serve as an important mechanism in reducing the recidivism rate among released offenders, which, in turn, will significantly reduce incarceration costs that are associated with recidivist offenders. Furthermore, this study finds that the unemployment rate among released offenders in the recessionary period is significantly higher than the unemployment rate among the general population. At this moment of financial crisis in both state and federal agencies, limited resources are available to provide educational programs to offenders in an attempt to enhance their opportunities to find employment upon release. Even though this study has clearly indicated that the effect of correctional education on recidivism is significant, a longitudinal study is needed to accurately assess the effect of correctional education on post-release employment among released offenders.

Reference

Batiuk, M. E., Lahm, K. F., McKeever, M., Wilcox, N., & Wilcox, P. (2005). Disentangling the effects of correctional education. Criminology & Criminal Justice, vol. 5, no., 1, pp. 55-74.

Batiuk, M. E., Moke, P., & Rountree, P. (1997). Crime and rehabilitation: Correctional education as an agent of change--A research note. Justice Quarterly, vol. 14, no. 1, pp. 167-180.

Blackburn, F. S. (1981). The relationship between recidivism and participation in a community college program for incarcerated offenders. Journal of Correctional Education, vol. 32, no. 3, pp. 23-25.

Burke, L. O., & Vivian, J. E. (2001). The effect of college programming on recidivism rates at the Hampden County House of Correction: A 5-year study. Journal of Correctional Edcuation, vol. 52, no. 4, pp. 160-162.

Cecil, D. K., Drapkin, D. A., MacKenzie, D. L., & Hickman, L. J. (2000). The effectiveness of adult basic education and life-skills programs in reducing recidivism: A review and assessment of the research. Journal of Correctional Education, vol. 51, no. 2, pp. 207-226.

Chappell, C. A. (2004). Post-secondary correctional education and recidivism: A meta-analysis of research conducted 1990-1999. Journal of Correctional Education, vol. 55, no. 2, pp. 148-169.

Fabelo, T. (2002). The impact of prison education on community reintegration of inmates: The Texas case. Journal of Correctional Education, vol. 53, no. 3, pp. 106-110.

Gordon, H. R., & Weldon, B. (2003). The impact of career and technical education programs on adult offenders: Learning behind bars. Journal of Correctional Education, vol. 54, no. 4, pp. 200-209.

Harlow, C. (2003). Education and correctional programs. Bureau of Justice Statistics, U. S. Department of Justice.

Hrabowski, F. A., & Robbi, J. (2002). The benefits of correctional education. Journal of Correctional Education, vol. 53, no. 3, pp. 96-99.

Hull, K., Forrester, S., Brown, J., Jobe, D., & McCullen, C. (2000). Analysis of recidivism rates for participants of the academic/vocational/transition education programs offered by the Virginia Department of Correctional Education. Journal of Correctional Education, vol. 51, no. 2, pp. 256-261.

Jancic, M. (1998). Does correctional education have an effect on recidivism? Journal of Correctional Education, vol. 49, no. 4, pp. 152-161.

Kodrzycki, Y. K. (2000). Discouraged and other marginally attached workers: evidence on their role in the labor market. New England Economic Review, May/June, pp. 35-40.

Lewis, J. (2006). Correctional education: Why it is only "promising." Journal of Correctional Education, vol. 57, no. 4, pp. 286-296.

Linton, J. (2007). United States Department of Education Update. Journal of Correctional Education, vol. 58, no. 2, pp. 95-97.

Matsuyama, K, & Prell, L. (2010). Education, employment and offender reentry. Correction Today, vol. 72, no. 4, pp. 90-91.

Mercer, K. (2009). The importance of funding post-secondary correctional educational programs. Community College Review, vol. 37, no. 2, pp. 153-164.

Nuttall, J., Hollmen, L., & Staley, E. M. (2003). The effect of earning a GED on recidivism rates. Journal of Correctional Education, vol. 54, no. 3, pp. 90-94.

Owens, C. D. (2009). Social symbols, stigma, and the labor market experiences of former prisoners. Journal of Correctional Education, vol. 60, no. 4, pp. 316-342.

Shapiro, L. (2011). 60% of new jobs in 2010 were in low-paying industries. The Huffington Post (on 01/13/11). (http://huffingtonpost.com/)

Steurer, S.J., Smith, L., & Tracy, A. (2001). Three state recidivism study. Lantham, MD: Correctional Education Association

Stevens, D., & Ward, C. (1997). College education and recidivism: Educating criminals is meritorious. Journal of Correctional Education, vol. 48, no. 3, pp. 106-111.

Taylor, J. M. (1992). Post-secondary correctional education: An evaluation of effectiveness and efficiency. Journal of Correctional Education, vol. 43, no. 3, pp. 132-141.

U.S. Department of Labor (2011). A profile of the working poor, 2009. Report 1027. March 2011. (http://www.bls.gov/cps/cpswp2009.pdf).

U.S. Department of Labor (2010). A profile of the working poor, 2008. Report 1022. March 2010. (http://www.bls.gov/cps/cpswp2008.pdf).

U.S. Department of Labor (2009). Ranks of Discouraged Workers and Others Marginally Attached to the Labor Force Rise During Recession. U.S. Bureau of Labor Statistics, U.S. Summary 09-04/April 2009. (http://www.bls.gov/opub/ils/pdf/opbils74.pdf).

Vacca, J. S. (2004). Educated prisoners are less likely to return to prison. Journal of Correctional Education, vol. 55, no. 4, pp. 297-305.

Ward, S. A. (2009). Career and technical education in the United States prisons: What have we learned? Journal of Correctional Education, vol. 60, no. 3, pp. 191-200.

Footnote

(1.) In 2008 the Education Division of the Indiana Department of Correction (IDOC) in 2008 collaborated with the IDOC Research & Planning Division to establish a database, which has contained more than 43 percent (n=6,561) of 15,184 offenders who were released throughout 2005, to conduct a follow-up study of the cohort of 6,561 released offenders in terms of post-release recidivism or employment.

(2.) The odds ratio is calculated by taking an antilog of the coefficient presented in the logistic estimates, as indicated in the All Offenders equation in Table 4. The odds ratio presents the situation in which an offender is identical in all respects except for the value on the variable of interest (i.e., participation in correctional education programs). In this case, the odds ratio (3.7:1) for an offender, who has not attended any correctional education programs during incarceration, to be a recidivist offender after release relative to an offender, who has participated in the correctional education programs, is taking an antilog of [e.sup.-1.304].

(3.) The odds ratio is calculated by taking an antilog of the coefficient presented in the logistic estimates, as indicated in the Comparison Group equation in Table 4. The odds ratio presents the situation in which an offender is identical in all respects except for the value on the variable of interest (i.e., offender's education). In this case, the odds ratio (2.8:1) for an offender, who has not completed high school, to be a recidivist offender after release relative to an offender, who has a high school diploma or GED, is taking an antilog of [e.sup.-1.038].

John Nally, Ed.D.

Susan Lockwood, Ed.D.

Katie Knutson, M.P.A.

Taiping Ho, Ph.D.

Biographical sketch

JOHN NALLY is the Director of Education for the Indiana Department of Correction. Along with being a past President of the Council of State and Federal Directors of Correctional Education, he has served on the Executive Board of the Correctional Education Association and was a member of the Reentry Roundtable on Education at the John Jay College of Criminal Justice. He has a Bachelor of Science and Master of Science degree from Indiana State University and a doctoral degree from Oakland City University.

SUSAN LOCKWOOD is the Director of Juvenile Education for the Indiana Department of Correction and a College of Education Faculty Member of the University of Phoenix, Indianapolis Campus. She is the President of the Council of State and Federal Directors of Correctional Education and actively involved with the Correctional Education Association. She completed her Bachelor's and Master's degrees at Ball State University, and her doctoral degree at Oakland City University.

KATIE KNUTSON is the Director of Research and Evaluation at the Indiana Department of Child Services. Prior to her current position, she served as a Senior Research Analyst for the Indiana Department of Correction. She received her Bachelor of Arts degree and Masters in Public Administration from Ball State University. Her studies explore correctional education programs and their impact on state correctional systems.

TAIPING HO is a full professor in the Department of Criminal Justice and Criminology at Ball State University in Muncie, Indiana. He has published numerous research articles on a variety of topics including police recruitment, criminal defendants with cognitive disorders, and police use of deadly force. He is an experienced police officer, has worked as a program specialist in correctional institutions, and is a CASA (Court Appointed Special Advocate) volunteer. He completed his doctoral studies at Florida State University.