Does the day of release matter to offender outcomes

Research Highlights: Offenders released early in the week have better outcomes, but this may be due to differences in risk profile.

Previous research in the Correctional Service of Canada shows a differential impact of release type on post-release success. However, little is known about the timing of release and its influence on outcomes. This study examined the impact of day of release and related this to type of release for offenders in CSC.

The study included 22,229 men (22% Indigenous) and 1,625 women (30% Indigenous) offenders released between April 1, 2012 and March 31, 2017. For most analyses, days of release were dichotomized into early week (Monday to Wednesday) and late week (Thursday and Friday). Analyses focusing on early discretionary release (EDR), an administrative option allowing release up to five days earlier for non-discretionary release, were limited to offenders on statutory release: 11,998 men and 546 women. Release suspensions and returns to custody (with or without an offence) were examined.

• Regardless of gender, one-third of offenders were released on Thursday; the lowest proportion of offenders was released on Friday (15% for men and 9% for women).
• A greater proportion of men were suspended (49% versus 38%) and returned to custody (37% versus 32%) when released later in the week compared to early in the week. This pattern held constant across Indigenous ancestry, security classification at release, region of supervision, and age at release and was consistent when time at risk was held constant using fixed follow-up periods of one, three, and six months. The time immediately following release (one-month fixed follow-up) is logically the period when the day of the week of release would have the most impact on outcomes (see table).
• Similarly, women released late in the week were more likely to be suspended than those released early in the week (40% versus 33%); younger women released later versus early in the week had higher suspension rates.
• For both men and women, those released early in the week remained in the community longer before returning to custody. There was no difference in time to first suspension.
• Non-discretionary releases were more likely to occur later in the week; these offenders had higher rates of suspensions and returns to custody.

• A similar pattern was evident in the use of EDR. The majority of offenders with an approved EDR were released 2-3 days prior to their scheduled statutory release date. EDR approvals were more likely to result in releases later in the week (89% versus 11% for men and 79% versus 21% for women). Those released later in the week had an increased rate of suspension; women released later on EDR also had increased rates of return to custody.

Regardless of gender or ancestry, offenders released earlier in the week had better outcomes in the short and longer term but this could be due to their lower risk profile. However, the results may also indicate that offenders released earlier in the week are better able to access parole officers and services before the week-end. The effect is consistent enough to suggest that the practice of releasing offenders earlier in the week to connect with community supports should be promoted.

Please e-mail the Research Branch or contact us by phone at (613) 995-3975.

You can also visit the Research Publications section for a full list of reports and one-page summaries.

Prepared by: S. Farrell MacDonald & K. Wardrop