Scanned from the original by the Prison Policy Initiative. Please report scanning errors to .

Follow-Up Study of a Sample of Offenders Who Earned High School Equivalency Diplomas (GEDs) While Incarcerated in DOCS

May 2001
State of New York
Department of Correctional Services
Building Number 2
Harriman Office Campus
Albany, New York 12226

Executive Summary

return rate for those who get GEDs inside

Follow-Up Study of a Sample of Offenders Who Earned High School Equivalency Diplomas (GEDs) While Incarcerated in DOCS

At the request of the Department's Division of Program Services, this research report was designed to address the question, "Does earning a High School Equivalency Diploma while incarcerated in DOCS reduce an offender's likelihood of returning to the Department's custody following release?" In particular, the Division of Program Services was interested in determining whether earning a GED while in DOCS has a different impact on offenders under age 21 compared with offenders age 21 and over. The present study examines the return-to-custody rates of a sample of offenders who earned a high school equivalency diploma while incarcerated in DOCS and two comparison groups of offenders who did not earn high school equivalency diplomas while in DOCS.

Education Program Objectives. The Department's academic education program focuses on providing inmates with the attitudes, knowledge, skills and credentials needed to function as contributing adults, both while incarcerated and when released back into the community. The two primary goals for the education program are to ensure that every inmate who leaves the system- 1) possesses a high school diploma or equivalency, and 2) has the skills needed to obtain employment when released from custody, In order to accomplish these goals, the Department mandates education for all inmates until they reach the ninth grade level[1] in reading and math. Additionally, inmates are encouraged to prepare for the Test of General Education Development (GED) which leads to a 2 high school equivalency diploma.[2]

Education Program Description. The Department provides a range of academic education programs for inmates without high school diplomas, through day or evening classes as well as outreach programs. Specifically, the education programs provided by the Department include: 1) Adult Basic Education (ABE) programs in English and Spanish, which focus on reading, writing, and math for inmates who function below the fifth grade level; 2) Pre-GED and GED classes in English and Spanish, which focus on reading, writing, and math for inmates who function above the fifth grade level and preparation of inmates for the GED examination to obtain a high school equivalency diploma; and 3) Bilingual programs which provide ABE and GED instruction in Spanish as well as English-as-a-Second-Language (ESL) for inmates with limited English speaking abilities.

Initial program placement is based on the results of standardized achievement tests administered upon intake as part of the reception/classification process. Achievement tests are subsequently administered (approximately every four months) to inmates participating in academic programs to measure progress and to determine eligibility for placement in more advanced level classes.

A screening test that is similar in both form and content to the GED is usually administered to determine inmate readiness for the actual GED test. The screening procedure enhances the individual's chance of passing the GED by providing a realistic testing experience that in turn maximizes the Department's resources.

The Tests of General Education Development (GED) are developed by the American Council on Education to assess skills, concepts and application of knowledge generally associated with each of the major content areas at the high school level. Policies and conditions under which certificates may be issued and regulations for administering the GED test are set by the New York State Department of Education. Candidates who meet the requirements receive a high school equivalency diploma, which is commonly accepted as a credential where a high school diploma is required.

Background of Research. The relationship between earning a high school diploma and various factors, such as lifetime income, is well-documented in general education research. Criminal justice research has consistently found that offenders typically lack a high school diploma and report only a junior high school level of education. Among new court commitments received by the Department of Correctional Services in 1999, the average reading level, based on standardized tests (Tests of Adult Basic Education (TABE) or Spanish Assessment of Basic Education (SABE))[3] administered during the reception/classification process, was 7 grade. Meanwhile, the average math proficiency based an standardized tests was at the 6 grade level.[4] However, little empirical research has been conducted on the impact of Correctional education programs, such as GED programs, on the recidivism rate of offenders.

Prior Department Research. In view of this lack of research, in 1986 the Department's Assistant Director of Education requested that the Division of Program Planning, Research and Evaluation examine the return-to-custody rate of a sample of offenders who earned a general education diploma (GED) while incarcerated in DOCS.

In December 1986, a research report was issued on the return rate of offenders who earned a GED while incarcerated.[5] The findings of this report were very encouraging. This study found that the return rate of a sample of offenders who earned a GED in 1983 was comparably lower (17.1%) than the Department's overall return-to-custody rate (26.3%).

In 1989, the Department's Director of Education requested that the 1986 study be updated and expanded using the education data that had recently become available on the Department's computerized information system. In response to this request, in July 1989, the Department produced a report[6] that improved upon the 1986 study by expanding the scope and the sample size of the study, as well as by including results from a more comparable control group. Specifically, the 1986 study included only 14 facilities, while the 1989 study included all facilities. Additionally, the 1986 study tracked a sample of 205 offenders while the 1989 study tracked 4,226 offenders. Finally, the 1986 study compared the return-to-custody rate of the sample of offenders who earned a GED while incarcerated to the overall Department return-to-custody rate. In contrast, the 1989 study compared the return rate of offenders who earned a GED while incarcerated to the return rate of offenders who were admitted to DOCS without a high school degree and did not earn a GED while in DOCS.

These modifications to the 1989 report substantially enhanced the reliability of the findings. The inmates released from DOCS who had earned a GED while in custody were expected to have a return rate of 34.8%, but actually returned at a rate of only 34.0%. Meanwhile, inmates released from DOCS who had not earned a GED while in custody were expected to have a return rate of 34.6%, but actually returned at a rate of 39.1 %. Overall, the offenders who earned a GED while incarcerated returned at a much lower rate (34.0%) than those offenders who did not earn a GED while incarcerated (39.1 %). This difference was found to be statistically significant.

Selection of Sample. To get the most recent release cohort possible that had been released for a long enough follow-up period (36 months is the Department standard), inmates released from DOCS' custody in 1996 were selected. All inmates released for the first time in 1996 from DOCS due to parole release, conditional release, or maximum expiration of sentence were selected for inclusion in the study. Next, the inmates were broken into three groups based on their educational background.

Educational Status Number Percent
Obtained GED at DOCS 2,330 14%
Admitted and Released with no Degree 9,419 57%
Admitted with Degree (H.S. Diploma or higher) 4,868 29%
Total 16,617 100%

Follow-Up Period. In accordance with the Department's established recidivism research procedures [7], all inmates were followed for 36 months after their release from custody.

Follow-Up Procedure. The Department's computer files were utilized to determine the number of sampled cases who were returned to DOCS' custody within 36 months of their release. The type of return to DOCS was also examined to determine whether an inmate initially returned as a New Court Commitment or as a Parole Violator.

Overall Findings. Overall, the attainment of a GED had a positive impact on recidivism rates, as measured by return to DOCS' custody. As shown in Table 1, only 32% of offenders who earned their GED while in DOCS' custody returned to custody within 36 months compared with 37% of inmates who did not earn their GED at DOCS. As expected, the return rate of inmates that earned a GED at DOCS was similar to the 32% return rate of inmates admitted with a degree.

Table 1. Educational Group by Return -to -Custody Status All Offenders First Released from DOCS in 1996
*the difference between inmates who earned a GED at DOCS and those with no degree is statistically significant at the .001 level
Educational Group
Return-to-Custody Status
 Total Number of Cases
Did Not Return Returned to Custody
Earned GED at DOCS 68.2% 1,589 31.8%* 741 100.0% 2,330
Admitted with Degree 67.7% 3,297 32.3% 1,571 100.0% 4,868
No Degree 63.4% 5,971 36.6%* 3,448 100.0% 9,419
Total 65.3% 10,857 34.7% 5,760 100.0% 16,617

Findings Among Young Offenders. With respect to young offenders (inmates under age 21 at release), the impact of earning a GED on recidivism rates was even more substantial. Table 2 shows that 40% of young offenders who earned their GED while in DOCS' custody returned to custody within 36 months compared with 54% of young inmates who did not earn their GED at DOCS. Again, the return rate of inmates that earned a GED at DOCS was similar to the 43% return rate of young inmates admitted with a degree.

Table 2. Educational Group by Return-to-Custody Status Under 21 Offenders First Released from DOGS in 1996
*the difference between inmates who earned a GED at DOCS and those with no degree is statistically significant at the .001 level
Educational Group
Return-to-Custody Status
 Total Number of Cases
Did Not Return Returned to Custody
Earned GED at DOCS 59.9% 224 40.1%* 150 100.0% 374
Admitted with Degree 56.8% 105 43.2% 80 100.0% 185
No Degree 46.3% 462 53.7%* 535 100.0% 997
Total 50.8 % 791 49.2% 765 100.0% 1,556

Statistically Significant Difference Between Inmates that Earned GED and Inmates that Did Not Earn GED. In addition to observing the raw difference in the actual return -to-custody rates of the inmates who earned a GED and those who did not, a chi square test was utilized to determine if this difference was statistically significant.

It was found that the difference in return rates between inmates who earned a GED at DOCS and those with no degree was highly statistically significant (p<.001) among all offenders and among young offenders. From a statistical perspective, a difference this large would occur by chance less than 1 time out of 1,000. Therefore, the lower return rate of the GED cases cannot be attributed to chance but to a real difference between the return rates of GED and non-GED cases.

The return rate of the inmates who earned a GED was less than the overall return rate for inmates released for the first time (due to parole, conditional release, maximum expiration) from the Department during 1996. Specifically 32% of inmates that received a GED returned to custody compared with 35% of all inmates (see Table 1). Similarly, 40% of young inmates that received a GED returned to custody compared with 49% of all young inmates (see Table 2).

Type of Return: New Commitment vs. Parole Violator. Inmates can be returned to custody for different reasons. Inmates that commit a new crime after release are returned as New Commitments, with a new sentence. Inmates released on parole can be returned to custody as Parole Violators if they violate the conditions of their parole. Among all offenders who returned to custody within 36 months, those who earned a GED at DOCS returned to custody as a New Commitment at a much lower percentage (34%) than inmates with no degree (41 %) and those admitted with a degree (38%)(see Table 3).

Table 3. Educational Group by Return Type All Offenders Returned to DOCS' Custody within 36 Months
*the difference between inmates who earned a GED at DOCS and those with no degree is statistically significant at the .001 level
Educational Group
Return-Type
 Total Returned
Returned as New Commitment Returned as Parole Violator
Earned GED at DOCS* 33.9% 251 66.1% 490 100.0% 741
Admitted with Degree 38.1% 599 61.9% 972 100.0% 1,571
No Degree* 40.9% 1,409 59.1% 2,039 100.0% 3,448
Total 39.2% 2,259 60.8% 3,501 100.0% 5,760

Among young offenders, those who earned a GED at DOCS returned to Department custody as New Commitments at a much lower percentage (33%) than inmates with no degree (46%). However, as expected, they returned as New Commitments at a similar rate as those admitted with a degree (31 %)(see Table 4).

Table 4. Educational Group by Return Type Under 21 Offenders Returned to DOCS' Custody within 36 Months
*the difference between inmates who earned a GED at DOCS and those with no degree is statistically significant at the .001 level
Educational Group
Return-Type
 Total Returned
Returned as New Commitment Returned as Parole Violator
Earned GED at DOCS* 33.3% 50 66.7% 100 100.0% 150
Admitted with Degree 31.3% 25 66.8% 55 100.0% 80
No Degree* 46.0% 246 54.0% 289 100.0% 535
Total 42.0% 321 58.0% 444 100.0% 765

Conclusion. The major finding of this study can be summarized by the following statements:

Among inmates first released from DOCS' custody in 1996 due to parole release, conditional release, or maximum expiration of sentence, those who earned a GED while incarcerated returned to custody at a statistically significantly lower rate than offenders who did not earn a GED while incarcerated. The relationship between GED attainment and return-to-custody is particularly strong among offenders who were under age 21 at release.
In examining this finding, it may be argued that those inmates who successfully earned a GED were more motivated or competent than those who did not participate in or did not complete a GED program and that this factor is related to their future success on parole. Therefore, it could be contended that these individuals would be expected to do well on parole as a result of their motivation and not just achievement of a GED.

However, this research was designed to analyze the relationship of GED attainment and recidivism without attempting to attribute any observed differences completely to the impact of the GED program. Therefore, the lower return rate of the sample of offenders who earned a GED may be jointly attributed to the offenders' motivation, capabilities, participation in other programs, and participation in the GED program. Of course, the possible existence of this self-selection bias should not lead to the conclusion that the provision of GED programs to these motivated offenders is unnecessary or uneconomical. On the contrary, it may be argued that it is appropriate correctional policy to offer such individuals opportunities to maximize their potential for successful reintegration into the community.

This study's finding, of a statistically significant difference between the return rates of offenders who earned a GED while incarcerated and those who did not earn a GED, confirms the positive relationship between these two factors which was suggested by the previous studies conducted by the Department. In fact, the prior study conducted by the Department (in 1989) found a very similar difference in return rates among the release cohorts examined, In the 1989 study, it was found that inmates who had earned a GED while at DOCS returned to custody at a rate of 34% compared to a rate of 39% among inmates who did not earn a GED while at DOCS. In the present study, 32% of inmates who earned a GED returned to custody within 36 months compared with 37% of inmates who did not earn a GED. The fact that both of these studies (which are based on relatively large sample sizes) arrived at a similar conclusion regarding the relationship between earning a GED and returning to Department custody seems to further validate this finding.

Additionally, this study's finding that the relationship between earning a GED and rate of return-to-custody is especially strong for young offenders validates the importance of the Department's emphasis on preparing inmates under the age of 21 to pass the GED exam. The finding of statistical significance along with the substantial follow-up period involved (36 months) enables the reader to consider these results with some degree of confidence. This study clearly indicates that those offenders who earned a GED while incarcerated were less likely to return to the Department's custody than a comparable sample of offenders who did not earn a GED.

Prepared by:
E. Michele Staley
Program Research Specialist III
Program Planning, Research and Evaluation
May 2001

1. Education for inmates became mandated until they reach the 9th grade level on May 1, 200 1. Previously, inmates were mandated to be in education only until they reached the 8th grade level.

2. Inmates under the age of 21 are mandated by Department Directive (#4804) to be enrolled in an academic education program until the attainment of a GED.

3. The TABE and SABE tests are produced by CTB McGraw-Hill.

4. See Characteristics of New Court Commitments: 1999. New York State Department of Correctional Services, 2000, pp. 117-118.

5. Follow-Up Sample of a Sample of Offenders who Earned High School
Equivalency Diplomas WhileIncarcerated. New York State Department of Correctional Services (DOCS), December 1986.

6. Follow-Up Study of a Sample of Offenders who Earned High School Equivalency Diplomas While Incarcerated. New York State Department of Correctional Services (DOCS), July 1989.

7. See 1996 Releases: Three Year Post-Release Follow-Up. New York State Department of Correctional Services, July 2000.