FORUM on Corrections Research

CORCAN participation and post-release recidivism

Many people who work in corrections believe that an offender who participates in employment programming while incarcerated is likely to have better adjustment in prison and reintegrate successfully into society upon release.

In fact, considerable research suggests that offenders with histories of employment instability are much more likely to both break prison rules and recidivate after release than offenders with stable employment records.

However, an important question remains. Does offender participation in prison work programs have any positive impact on institutional adjustment and post-release recidivism? This article examines this question within the context of the Correctional Service of Canada's correctional industry program (CORCAN). Does continuous CORCAN participation near release impact on offender post-release recidivism ? Research methodology The study sample was composed of 269 federally sentenced male offenders who had worked for CORCAN for at least six uninterrupted months while incarcerated. These offenders were all working for CORCAN within one month before their release and had been released at least one year prior to the study.

The re-admission rates calculated for this study reflect the return of offenders to federal custody for either technical violations of their conditional release or a new offence. New convictions (any or violent) represent offences the offenders were convicted of after release. Sample characteristics The average age of the study sample was 36, but roughly half of the offenders were younger than 31. Most of the offenders were non-native (88.4%), two-thirds of the offenders were single, about three-quarters of the offenders were serving their first federal sentence (two years or longer), and three-quarters of the offenders were serving sentences of less than five years.

Roughly half of the sample had not been released since incarceration. The other half had been returned to custody after having their conditional release revoked. Most of the offenders worked for CORCAN in either fabrication, industries or agriculture, although some offenders did painting, micro-filming, welding and printing work.

Quebec had the greatest regional representation within the sample (35.4%), followed by the Prairie (28.9%) and Ontario (13.7%) regions. Approximately 55% of the offenders were released on statutory release, 24% on day parole and 19% on full parole. The rest of the sample was released at the end of their sentence. Risk level The Statistical Information on Recidivism scale was used to determine the sample's level of risk of re-offending. A thorough review of criminal records placed about half of the sample in the "very poor" or "poor" risk groupings (high risk, with a 63% average probability of arrest within three years of release). The other half of the sample was in the "very good", "good" or "fair" risk groups (low risk, with a 33% average probability of arrest within three years of release).

As expected, about three-quarters of the offenders granted full parole were assessed as low-risk offenders. In contrast, nearly half of the offenders released on statutory release were high-risk offenders. More than half of the offenders released on day parole were low-risk offenders, but slightly more than a third fell into the "very poor" risk category. Release practices, then, appear consistent with assessment of risk. Time of release Offenders who had served four-ninths, or less, of their sentence at release were categorized as "early" releases, while offenders who had served more than four-ninths of their sentence were categorized as "late" releases.

An "early" release within this framework is well below the average time served by offenders prior to release. Offenders are eligible for full parole after serving one-third (three-ninths) of their sentence and for statutory release after serving two-thirds (six-ninths) of their sentence. Two-thirds of the offenders in the sample received an "early" release.

While risk level was found to be unrelated to the timing of offender release, the type of release was significantly related to the amount of time served by the offender. For example, most of the offenders released on full parole (85%) and day parole (97%) received an "early" release. Not surprisingly, nearly half the offenders released on statutory release were released "late" in their sentence.

The offenders released on statutory release who received an "early" release were offenders who were serving remnants of a sentence-they had previously been released, had their conditional release revoked, were returned to federal custody and were then released again on statutory release Post-release outcome The sample's overall return to federal custody rate was 42%, within an average follow-up period of 1.5 years (the follow-up range was from one to three years).

The rate of return to federal custody for offenders released on all forms of day parole was 56.7%, with close to 10% of these returns being for a new offence (see Table 1). Return rates were substantially lower for offenders released on full parole (19.2%). In fact, just one offender released on full parole was returned to federal custody for a new offence.

Table 1

Approximately one-third of the offenders incurred a new conviction after release. However, keep in mind that some offenders were convicted of a new offence that did not result in a return to federal custody (the conviction occurred after their sentence ended and resulted in a fine, probation or a sentence of less than two years). Still, the pattern of results for convictions mirrored the results for return to federal custody.

More important, the rate of return to federal custody (for any reason and for a new offence) for CORCAN participants released on full parole was significantly lower than the national average⁽²⁾ for this type of release (see Table 2). In fact, there was a 27.8% reduction in readmissions to federal custody associated with uninterrupted CORCAN participation near release. A similar pattern emerged for return to custody for any reason and statutory release. Risk, time of release and post-release outcome This study also reconfirmed that risk assessments (based on the offender's criminal history) can predict conditional release outcome. The risk levels of these CORCAN participants were found to be significantly related to return to federal custody (r = -.30, p<.001), return to federal custody for a new offence (r = -.18, p<.01), reconviction (r = -.32, p<.001) and reconviction of a violent offence (r = -.17, p<.001).

Table 2

Nearly all of the offenders released on day parole and full parole were released "early" in their sentence, so we could not draw conclusions as to time of release and outcome for these groups. However, time of release ("early" versus "late") was unrelated to post-release outcome for the offenders released on statutory release. A definite impact The results of this research indicate that uninterrupted participation in CORCAN prison industry programs immediately prior to release may have a positive impact on offender post-release recidivism, particularly for offenders released on full parole.

The fact that offenders released on full parole appear to benefit most from CORCAN participation and that the majority of offenders released on full parole tend to be categorized as "low risk" offenders points to the importance of risk assessment in identifying candidates for discretionary release.

Further, such participation should be encouraged at the early stages of offender sentences for lower risk offenders.



⁽¹⁾Second Floor, 340 Laurier Avenue West, Ottawa, Ontario K1A 0P9.

⁽²⁾The national averages used for comparison were taken from Basic Facts About Corrections in Canada (Ottawa: Solicitor General Canada, 1993). These figures are comparable with other national recidivism averages. See T. Nouwens, L. Motiuk and R. Boe, "So You Want to Know the Recidivism Rate," Forum on Corrections Research, 5,3 (1993): 22-26.