Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

Research Brief

Warning This Web page has been archived on the Web.

Profile of Offenders in Administrative Segregation: A Review of the Literature

2008 No B-39    PDF

Shauna Bottos
Research Branch
Correctional Service Canada

September 2007



The author would like to extend her sincere gratitude to Dr. Christa Gillis and to Dr. Kelley Blanchette for their reviews of this report.


Administrative segregation is a form of confinement that offenders may enter voluntarily or involuntarily. Although it is not meant to be a form of punishment (Solicitor General of Canada, 1998), removal from the general inmate population may have adverse affects on the individual and may also impede reintegration efforts. As a result, there is a concern surrounding the potential overuse of administrative segregation as an offender management tool, particularly for those who request segregation voluntarily (Arbour, 1996). This latter group tends to remain in segregation for considerably longer periods of time than inmates placed in segregation involuntarily (CSC, 2005). A better understanding of the characteristics of offenders in segregation may aid in the development of a screening protocol for identifying offenders at-risk for such placements. This paper provides a review of this literature, as well as the evidence concerning the impact of segregation on offenders.

Most offenders in administrative segregation are placed there involuntarily, typically because they pose a danger to staff, other inmates, or the security of the institution. When offenders request segregation, it is most often because they fear that their personal safety is in jeopardy in the general inmate population (Wichmann & Nafekh, 2001).

A variety of case-specific factors assessed during admission to federal custody distinguish offenders placed in administrative segregation (i.e., voluntary or involuntary) from those who are not. Segregated inmates are most likely to be single, male offenders in their late 20’s to early 30’s, slightly younger than the non-segregated population (Wichman & Nafekh, 2001). Offenders placed in segregation tend to have more extensive criminal histories, more violent offence convictions, lower reintegration potential ratings, and higher security classifications at intake, than inmates who are not subsequently placed in segregation. The former group is also more likely to be assessed as higher risk and need. Throughout their sentence they show the greatest adjustment difficulties while incarcerated.

In contrast to the similarities in the profile of male and female inmates placed in administrative segregation, gender differences become apparent when comparisons are drawn between those admitted voluntarily and those placed there involuntarily. For men, very few characteristics differentiate voluntarily and involuntarily segregated inmates, suggesting that their profile is best captured by that of the administrative segregation population as a whole. For women offenders, several factors distinguish these groups. Women placed in segregation involuntarily are more likely to be Aboriginal offenders, to have started their criminal careers earlier, and to exhibit more continual involvement in criminal activity, than women in voluntary segregation. Involuntarily segregated women are also more likely to have been placed in disciplinary segregation and commit considerably more institutional misconducts than their voluntarily segregated counterparts. This suggests that involuntarily segregated offenders have greater difficulty adjusting to the institutional environment.  Those who experience both involuntary and voluntary segregation, however, appear to be the highest risk and need offenders.

Placement in administrative segregation may have far reaching consequences. There is evidence that segregation may reduce offenders’ ability to be moved to lower security-level classifications and that segregation hinders access to programming (Luciani, 1997). Administrative segregation may also influence early release decisions. Segregated offenders have been found to be significantly less likely to be granted a discretionary release, more likely to be released at their statutory release or warrant expiry date, and to recidivate at higher rates, than non-segregated offenders (Motiuk & Blanchette, 2001; Wichmann & Nafekh, 2001). In contrast to speculations, there is no evidence to suggest that inmates’ psychological functioning is adversely impacted by the segregation experience (Zinger & Wichmann, 1999). Whether these outcomes differ according to voluntary and involuntary admissions has not been examined.

Given the changes in the federal offender population over the years, an update of the profile of offenders in administrative segregation would be useful, including an examination of the reasons offenders enter segregation voluntarily. These will be important steps toward the development of early risk identification protocols and more effective offender management strategies.



In response to concerns surrounding the use of administrative segregation in federal institutions that arose from the Arbour Commission Report in 1996 (Arbour, 1996), the Correctional Service of Canada (CSC) established a Task Force on Administrative Segregation shortly thereafter. Although the primary objectives of this Task Force were to examine a broad range of policy and procedural issues related to administrative segregation, descriptive analyses were undertaken to gain a better understanding of the composition of federal offenders residing in segregation units (Kane, 1997; Motiuk & Blanchette, 1997). Since then, this line of inquiry has been extended to female offenders in segregation (Wichmann & Taylor, 2004) and has also stimulated interest in the psychological effects associated with segregation (Zinger & Wichmann, 1999), and these offenders’ post-release outcomes (Motiuk & Blanchette, 2001).

Before reviewing this research, a brief description of the types of segregation used in Canadian correctional facilities and the regulations governing their application is provided. The subsequent overview of the research examining the profiles of offenders in segregation will draw comparisons between administratively segregated inmates and non-segregated inmates and then concentrate upon offenders in administrative segregation voluntarily. Notably, an exhaustive literature search was completed using both academic databases and the internet, and revealed that only Canadian studies, produced by CSC, have profiled the administrative segregation population. As a result, this report draws on this data.


Types of Segregation and Regulations Guiding Use

Beyond the typical separation of incarcerated offenders into minimum-, medium-, and maximum-security facilities, segregation is one of many population management options available to personnel working in penitentiaries. Removal of offenders from the general inmate population, however, is an extreme option that is only to be used under the strict guidelines laid out in the Corrections and Conditional Release Act (CCRA), and only when other less restrictive alternatives have been exhausted or are rendered ineffective (CCRA, 1992). There are three types of segregation: disciplinary segregation and administrative segregation, the latter of which can be voluntary or involuntary.

Disciplinary segregation, unlike administrative segregation, is a sanction that is intended to have punitive value and has a maximum duration of 30 days (CCRA, Section 44 (1-F)). This type of segregation is imposed upon offenders who have been charged with, and found guilty of, a serious disciplinary offence. Not surprisingly then, the conditions of confinement for this group may be harsher than those of administrative segregation as it may also include a loss of certain privileges.

Administrative segregation tends to be the more commonly used separation tactic in institutions compared to segregation for disciplinary purposes (Wichmann & Taylor, 2004). This type of segregation can be either voluntary, where it is requested by the inmate, or involuntary, in which case the inmate is placed in segregation at the discretion of the institutional head (CCRA, Section 31 (3)). In accordance with Section 31 of the CCRA (1992), an inmate may only be placed into administrative segregation if one of the following three criteria holds: 1) the inmate poses a danger to staff, other inmates, or the security of the institution; 2) the inmate has the potential to interfere with an ongoing investigation; or 3) for the inmate’s own safety. As an offender management tool, although the purpose of administrative segregation is to prevent the inmate from associating with the general prison population, it is not meant to be punitive in nature but rather preventative in form (Solicitor General of Canada, 1998). That is, it is employed as a means of preventing altercations, harm, or interference with internal investigations. As such, the law mandates administratively segregated offenders receive the same rights, privileges, and conditions of confinement as the general inmate population, with the exception of those that can only be afforded in the company of other inmates, and those that cannot reasonably be provided because of security concerns or the limitations of the segregation area (CCRA, 1992). Even though the CCRA does not specify the maximum length of time an offender can remain in administrative segregation, they are to be returned to the general inmate population, either in the same or a different facility, at the earliest appropriate time.1

While more offenders in administrative segregation are there involuntarily, there is a subgroup of offenders who request segregation (Motiuk & Blanchette, 1997; Wichman & Nafekh, 2001; Wichmann & Taylor, 2004). Despite this latter group comprising roughly one half of the administratively segregated (predominantly) male offender population (Kane, 1997; Motiuk & Blanchette, 1997), and approximately one quarter of the female offender population (Wichmann & Taylor, 2004), few recent studies have addressed the reasons why these individuals request to be removed from the general prison population. Studies on both male and female offenders have found that the most common reason for requested segregation was to protect their own personal safety (Kane, 1997; Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004). Earlier research suggests the following reasons for seeking protection: conflicts with members in the general prison population over gambling and drug debts, the type of offence for which they were convicted (e.g., sex offences), being suspected of being an informant, personality aberrations, phobias (e.g., fear of homosexual offenders), being the target of sexual aggression, and wanting to escape the crowded and often violent atmosphere of maximum-security (Gendreau, Tellier, & Wormith, 1985; Wormith, Tellier, & Gendreau, 1988). As the federal offender population has changed over the years (Boe, Nafekh, Vuong, Sinclair, & Cousineau, 2003; CSC, 2006a; Motiuk & Vuong, 2005), it will be important that future research provide an update on why some offenders seek segregation voluntarily so that risk factors can be appropriately targeted and mitigated while in custody. In the meantime, research that has begun to examine the characteristics and needs of segregated offenders provides some insight into those factors that may place individuals at risk for segregation.

Profile of the Segregated Population

Although the early part of the 21st century witnessed a gradual decrease in the voluntary segregation admission rates, since the 2003/2004 fiscal year there has been a reversal of this trend (Head, 2006). The most recent statistics published by the CSC, drawn from the 2004/2005 fiscal year Performance Report, revealed that there were 1,899 admissions or re-admissions to voluntary segregation during this timeframe (CSC, 2005). While this is still considerably less than the number of involuntary admissions (i.e., 5,322), it remains a significant proportion of the incarcerated population. Moreover, the average length of stay for voluntary admissions is nearly double that of those in involuntary segregation (68 vs. 35 days, respectively; CSC, 2005). These issues have rekindled concern over the use of segregation and, in particular, the lack of appropriate alternatives for offenders segregated on a voluntary basis (CSC, 2005). A better understanding of the characteristics of this population will be an important step toward the development of protocols that aid in the early identification of offenders at-risk of seeking segregation. Preliminary research suggests that a variety of case-specific factors assessed during admission to federal custody distinguish offenders who are later placed in administrative segregation from those who are not. Furthermore, there appear to be some important differences between voluntarily and involuntarily segregated offenders, particularly for female inmates.

Profile of Administratively Segregated Offenders

Only three studies to date have examined the characteristics of offenders in administrative segregation (using predominantly male samples) and drawn comparisons between these individuals and non-segregated offenders (Kane, 1997; Motiuk & Blanchette, 1997; Zinger & Wichmann, 1999).  Only one investigation has done so exclusively with women offenders (Wichmann & Taylor, 2004). One large-scale evaluation was completed by the CSC that examined both male and female inmates in segregation (Wichmann & Nafekh, 2001). In general, the literature suggests that segregated offenders are slightly younger than their non-segregated counterparts, with the former group being in their late-twenties to early-thirties and the latter in their mid-thirties (Motiuk & Blanchette, 1997; Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004; Zinger & Wichmann, 1999). Both groups reveal comparable proportions of Aboriginal offenders (i.e., segregated: 13%-16% and non-segregated: 14%-15%; Kane, 1997; Motiuk & Blanchette, 1997), however these proportions tend to be higher when women offenders are examined in isolation (segregated: 26% and non-segregated: 17%; Wichmann & Taylor, 2004). Lastly, one study found that the vast majority of segregated offenders (72%) were single at the time of admission to federal custody (Motiuk & Blanchette, 1997).

In terms of their criminal histories, offenders placed in segregation, for whatever reason (i.e., voluntarily or involuntarily) and regardless of gender, tend to have had more extensive involvement with the criminal justice system, both during their youth and as an adult, than non-segregated inmates (Kane, 1997; Motiuk & Blanchette, 1997; Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004). Women offenders are also more likely to have violent offence histories (Wichmann & Taylor, 2004), and both male and female offenders are more likely to be rated as having lower reintegration potential at intake (Wichmann & Nafekh, 2001).2 Not surprisingly then, segregated offenders are more likely to be designated as maximum-security and non-segregated as minimum-security at the time of admission (Motiuk & Blanchette, 1997).

Another consistent finding across studies and genders is the higher overall risk and need levels of segregated offenders compared to non-segregated offenders at intake assessment (Kane, 1997; Motiuk & Blanchette, 1997; Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004). An in-depth analysis of the areas of greatest need in the studies that included mostly males revealed that those in administrative segregation had identified needs in virtually all domains upon admission to custody (i.e., six out of seven need domains; Kane, 1997; Motiuk & Blanchette, 1997). Specifically, segregated inmates are more likely to have lower levels of educational attainment, unstable work histories, and to be unemployed at the time of arrest. They also tend to have unstable accommodations and financial difficulties, and are likely to be socially isolated, or to associate with substance abusers and criminal acquaintances. Cognitive problems, poor conflict resolution skills, impulse control deficits, hostility, thrill-seeking, and low tolerance to frustration are pervasive among this group, and problems relating to substance abuse often begin at an early age and interfere with most aspects of their lives. Needs in the attitude domain are likely to be extensive and are characterized by negative attitudes toward the criminal justice system and interpersonal relationships. Pro-criminal and non-conforming attitudes are also common among segregated inmates. Although family functioning needs were not found to differ significantly between the groups, offenders at-risk for segregation report more familial dysfunction than those who are not placed in these units (Motiuk & Blanchette, 1997). Another investigation examining the psychological functioning of men in segregation revealed that they also tend to report significantly more mental health problems than those who are not in segregation (Zinger & Wichmann, 1999).

An examination of segregated women offenders showed that they tend to have slightly fewer needs than their male counterparts, revealing identified needs on only four of the seven need domains at intake (Wichmann & Taylor, 2004). High need areas included those related to interpersonal functioning (i.e., associates, marital/family), personal/emotional orientation, and substance abuse. Hence, across male and female offenders in segregation, criminal associates, personal/emotional functioning, and substance use are areas that clearly contribute to the profile of offenders at-risk for removal from the general prison population and, therefore, will be important areas to target early in their sentence to reduce the likelihood of being segregated.

In addition to the various factors assessed at intake, the evaluation of offenders’ ability to adjust to institutional life may also distinguish those individuals at a heightened risk of subsequently being placed in segregation. Women offenders in involuntary administrative segregation have been found to be 12 times more likely to have been placed in disciplinary segregation, six times more likely to have been involved in institutional misconducts, and four times more likely to have committed a new offence during their current sentence, than non-segregated inmates (Wichmann & Taylor, 2004). Results of studies comprised of male offenders likewise suggest that disciplinary segregation in previous prison terms is predictive of being placed in administrative segregation (voluntary or involuntary) in the future (Kane, 1997; Motiuk & Blanchette, 1997) and that men in segregation exhibit more adjustment difficulties than their non-segregated counterparts (Wichmann & Nafekh, 2001). These findings suggest that difficulty adjusting to the prison environment may be a potent indicator of later segregation across genders.

Only one investigation has examined the personality characteristics of segregated offenders, an area that may be important for predicting institutional adjustment (Zinger & Wichmann, 1999). Using the widely applied five-factor model of personality developed by Costa and McCrae (1992), Zinger and Wichmann (1999) found segregated offenders to exhibit significantly higher levels of neuroticism, and lower levels of extraversion, openness to experience, agreeableness, and conscientiousness, than non-segregated offenders. Thus, segregated offenders exhibit a tendency toward negative mood states and difficulties coping with stressful situations (i.e., neuroticism). They also tend to be less sociable and assertive (i.e., extraversion), less open to novel experiences and attentive to inner experiences (i.e., openness), less sympathetic towards others, and more egocentric and suspicious of others’ motives (i.e., agreeableness). Finally, their lower levels of conscientiousness indicate that they tend to exhibit disorganized behaviour and impulse control deficits. Since this study was based on a sample of male offenders, whether these findings generalize to women in segregation remains unknown.

In summary, the results of studies comparing administratively segregated and non-segregated offenders suggest that individuals at-risk for segregation tend: to be in their late 20’s or early 30’s; to be single; to have achieved low levels of education; and to present with unstable work histories and several periods of unemployment. They are likely to have had extensive contact with the criminal justice system dating back to their youth, and are particularly apt to be rated as high risk and high need at the time of admission to custody. Identified needs are evident in multiple domains, most notably substance abuse, associates, and personal/emotional orientation. Difficulties adjusting to institutional confinement are prevalent and may be related to their negative, antisocial, impulsive personality style and difficulties coping with stressful events. Together, these findings highlight that several factors are predictive of offenders at-risk for later segregation. However, given that offenders who request segregation may present with a different profile than those placed in such units involuntarily, it is important to examine the voluntary and involuntary segregation populations separately.

Profiles of Voluntarily vs. Involuntarily Segregated Inmates

Despite the growing concern over the recent increase in offenders in voluntary segregation (CSC, 2005), outside of CSC, there is no published research examining the characteristics of these individuals. The present section provides a profile of this population based upon the limited number of studies to date. The profiles differ for male and female offenders and are presented separately.

Voluntarily Segregated Male Offenders

The descriptive analyses completed by the Task Force on Administrative Segregation (Kane, 1997) and Motiuk and Blanchette (1997) were among the first to describe the voluntarily segregated population. In general, their results suggest that these individuals appear to be similar to inmates involuntarily placed in segregation in many respects. In terms of their demographic characteristics, for instance, the authors found that both groups were in their early 30’s. However, they failed to report comparisons between the groups on inmates’ ethnicity and marital status. Though direct comparisons between this earlier and more recent research are difficult, studies published since the late 90’s suggest that the average sentence length, Aboriginal status, and the percentage of men serving life or indeterminate sentences, do not differ significantly according to the type of administrative segregation (Wichmann & Nafekh, 2001). The average sentence length of men in voluntary and involuntary segregation is approximately four years (Wichmann & Nafekh, 2001), and roughly 16% of both groups identify themselves as being Aboriginal. Between six and seven percent of these populations are serving life sentences (Wichmann & Nafekh, 2001). By comparison, the average sentence length of men in the general federal offender population is just under three-and-a-half years (Mullins, 2005) and, identical to the aforementioned percentages, 16% are Aboriginal offenders (PSEP, 2005; Trevethan, Moore, & Rastin, 2002). As of 2005, 22% of the total federal offender population were serving life or indeterminate sentences, 98% of which were male offenders (PSEP, 2005).

Few significant differences were found on the criminal history variables under study. Offenders who requested segregation were similar to those placed in segregation involuntarily in terms of their contact with youth court, including whether they had previously been under community supervision or served a sentence under open or secure custody; and they also reported similar experiences with the adult court system (i.e., community supervision, and prior federal terms; Kane, 1997; Motiuk & Blanchette, 1997). However, voluntarily segregated offenders were significantly more likely to have served prior provincial terms as an adult than involuntarily segregated inmates (Kane, 1997; Motiuk & Blanchette, 1997). The only other criminal history variable to differ between the groups was a history of sexual offences, where again, the voluntarily segregated offenders were considerably more likely to have current or previous sex offence convictions than their involuntarily segregated counterparts. Unfortunately, no other offence categories (e.g., drug offences, assault) were examined.

Analogous to their criminal histories, voluntarily and involuntarily segregated inmates were assigned similar custody ratings at intake. Although slightly more offenders in voluntary segregation were designated maximum-security and slightly more in the involuntary group designated minimum-security, these differences were not significant (Motiuk & Blanchette, 1997). Both groups were most likely to be assigned a medium-security rating. Given these findings, it is perhaps not surprising that offenders’ reintegration potential ratings at intake also do not appear to distinguish the groups (Wichmann & Nafekh, 2001).

Offenders who request segregation and those who are placed there involuntarily also exhibit similar risk/need profiles, with both groups tending to be rated as high risk and need (Kane, 1997; Motiuk & Blanchette, 1997). Furthermore, there appear to be no significant differences between the groups on any of the seven need domains assessed at intake. As such, the six need domains described previously for the administrative segregation population as a whole are representative of the needs of this group of voluntarily admitted male offenders (Kane, 1997; Motiuk & Blanchette, 1997). Moreover, mental health and psychological functioning do not distinguish the groups (Motiuk & Blanchette, 1997; Zinger & Wichmann, 1999).

Lastly, institutional adjustment and conditional release outcomes were found to be comparable for voluntarily and involuntarily segregated inmates. Disciplinary segregation during prior prison terms, previous escape attempts, failure on conditional release, and the length of time since their last incarceration were equivalent between the groups (Motiuk & Blanchette, 1997). Thus, studies that have included mostly male offenders suggest that voluntarily and involuntarily segregated individuals are more similar than they are different.

Voluntarily Segregated Female Offenders

At present, few studies have focused on the experience of administrative segregation among female offenders and elucidated a profile of those placed there voluntarily (Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004). One study provided some comparisons between this group and women placed in segregation involuntarily (i.e., Wichmann & Nafekh, 2001), and another between these two groups and a group of women who experienced both voluntary and involuntary segregation (i.e., Wichman & Taylor, 2004). In stark contrast to the few differences noted to distinguish voluntarily and involuntarily segregated male offenders, several demographic, criminal history, risk/need, and institutional variables set apart these groups when women offenders were the population of interest.

Consistent with the studies of male offenders, women in voluntary segregation tend to be in their early 30’s. However, this age is comparable to that of the involuntarily segregated female offender population, as well as women who are placed in both voluntary and involuntary segregation during their sentence (Wichmann & Taylor, 2004). In terms of ethnicity, Aboriginal women may be under-represented in the voluntary admission group, with more tending to be placed in segregation involuntarily (Wichmann & Nafekh, 2001; Wichmann & Taylor, 2004).

Examination of past criminal history reveals that, even though the voluntary group had more prior adult convictions, inmates who experienced involuntary segregation and both voluntary and involuntary segregation, were more likely to have received young offender depositions, and they also tended to exhibit more continual involvement in criminal activity as indicated by fewer or shorter crime-free periods (Wichmann & Taylor, 2004).

Unlike with male offenders where sexual offences distinguished voluntarily and involuntarily segregated inmates, no offence categories appear to differentiate these groups of women offenders (Wichmann & Taylor, 2004). However, women who experience involuntary segregation tend to be more likely to be serving life sentences than those placed in segregation voluntarily (Wichmann & Taylor, 2004). Although security level classifications have not been reported for segregated women offenders, similar to their male counterparts, reintegration potential ratings at intake do not appear to distinguish voluntarily and involuntarily segregated women (Wichmann & Nafekh, 2001). However, their higher ratings do clearly set them apart from the non-segregated population (Wichmann & Nafekh, 2001).

In terms of their overall risk and need ratings at intake, women placed in either voluntary segregation or involuntary segregation appear to be more similar to one another than those who experience both types of segregation (Wichmann & Taylor, 2004). This latter group had the largest percentage of high risk and high need offenders. However, the areas of greatest need appear to be similar across the groups, and include those pertaining to women’s personal/emotional orientation and substance abuse. Forty-five percent or more of the women exhibited considerable difficulties in these areas. By comparison, the domains of least need may vary according to group membership. For both voluntarily segregated women and those segregated involuntarily, Wichmann and Taylor (2004) found that few women reported high needs in community functioning (i.e., 8% and 9%, respectively), and the former group was also less likely to be needy in the associates or social interaction domain (8%). Although the area of lowest need for women placed in both voluntary and involuntary segregation was in the associates domain, it is noteworthy that no one need area was endorsed by less than 18% of this group. Thus, women who experience both types of segregation clearly appear to be the highest risk and need group.

Considerable differences are also evident between segregated women’s institutional adjustment according to the type of segregation experienced. During the three-year period covered by the Wichmann and Taylor (2004) study, none of the voluntarily segregated women had been placed in disciplinary segregation, whereas offenders in the other two groups had. Similarly, women who had experienced both voluntary and involuntary segregation committed an average of 21 institutional incidents during their current sentence, which was nearly four times as many as the involuntary group, and ten times as many as the voluntary group. Interestingly then, voluntarily segregated women appear to have less difficulty adjusting to the prison environment than women who experience involuntary segregation, either in isolation, or with voluntary segregation, with those placed in both having the greatest difficulties. These findings are in contrast to the poor discriminatory power of institutional adjustment in the studies of segregated men noted above (e.g., Motiuk & Blanchette, 1997).

In summary, several static and dynamic factors distinguish segregated and non-segregated offenders and, in its entirety, the administrative segregation population differs little according to gender. However, when the profiles of the voluntary and involuntary segregation populations are examined separately, clear gender differences emerge. Few characteristics distinguished voluntarily and involuntarily administratively segregated male inmates, thereby suggesting that their profile is best captured by the aggregated administrative segregation population descriptive profile. In contrast, many features differentiated these two groups when female offenders were examined. The preliminary profiles that have emerged will be important to bear in mind in future risk detection and management efforts given the potentially far reaching consequences associated with the segregation experience.

Implications of Administrative Segregation

Reminiscent of the research profiling the administratively segregated population, there remains a dearth of inquiry into the consequences of segregation on offenders and the correctional system. Recognizing the limited evidence on the effects of administrative segregation on offenders, both the Arbour Commission report (1996) and the report published by the Task Force in 1997 speculate that the effects of long-term segregation can be detrimental, both for the individual and for reintegration efforts. These reports were based on early research, however, and the conditions of confinement have changed considerably since that time. Nevertheless, there is some evidence suggesting that segregation may affect individuals’ well-being, their security re-classification status, program participation, discretionary release decisions, and conditional release outcomes. The present section provides an overview of these findings.

As noted previously, administratively segregated (male) offenders appear to report more mental health problems and poorer psychological functioning than non-segregated offenders (Zinger & Wichmann, 1999). However, drawing from the single, well-designed longitudinal study of the effects of segregation over time (i.e., up to a period of 60 days in segregation), there is no evidence to suggest that the psychological functioning of administratively segregated offenders deteriorates over time as a result of being segregated from the general inmate population (Zinger & Wichmann, 1999). Clearly, these findings need to be replicated and it needs to be reaffirmed that the between-group differences on the prevalence of mental health problems are not attributable to placement in administrative segregation (Zinger & Wichmann, 1999). An examination of the effects of segregation on women and Aboriginal offenders should also be considered as these groups have unique needs that may impact upon their ability to adapt to the experience of segregation (Kane, 1997). These factors were not addressed by the aforementioned study.

In addition to an individual’s mental state, segregation may further impact upon an offender’s ability to receive lower security-level classifications during the course of his or her sentence. Indeed, one study found serving time in segregation to be the strongest predictor of inmates’ security reclassification status (Luciani, 1997). Nevertheless, it is possible that the characteristics of this population, such as their poorer institutional adjustment and higher risk levels, may in fact be what are impacting lower security re-classifications.

Another factor that could impact offenders’ security classification, as well as have repercussions for early release decisions, is program participation (Luciani, 1997). Although the rights and privileges to which offenders in segregation are entitled include access to programming, the very nature of the segregation environment can impose limits on such opportunities while in custody. Furthermore, since many offenders can remain in segregation for significant periods of time, most notably those in voluntary segregation, their ability to address the goals of their correctional plan may be greatly impeded. Research has shown that program completion is inextricably linked to an offender’s ability to move from higher to lower levels of security classification (Luciani, 1997), and that participation in correctional interventions influences the likelihood that parole will be granted (National Parole Board, 2002).

Placement in administrative segregation may also have a direct impact on parole decision-making, with several recent studies of male (Motiuk & Blanchette, 2001; Wichmann & Nafekh, 2001) and female offenders (Wichmann & Taylor, 2004) reporting that offenders in administrative segregation were significantly less likely to be granted a discretionary release (i.e., day or full parole) and more likely be released on statutory release or at their warrant expiry date than non-segregated offenders. Disentangling the cause-and-effect relationships between administrative segregation and security re-classification levels, and between administrative segregation and release type, remains for future research. Similarly, the higher recidivism rates noted among segregated individuals compared to offenders not placed in segregation (Motiuk & Blanchette, 2001; Wichmann & Taylor, 2004), may be an artefact of segregated offenders’ higher risk and need ratings. As such, a cause-and-effect relationship should not be inferred.


The present literature review sought to describe the administratively segregated population, with a particular interest in those offenders who reside in segregation units voluntarily. Although some noteworthy contributions to describing the profile of the segregated population have been made, several areas remain in need of addressing. Given the changing federal offender profile, for instance, an updated profile of the administrative segregation population could be useful. In accordance with CSC’s strategic priority to enhance interventions targeting Aboriginal offenders (CSC, 2006b), comparisons should be made between Aboriginal and non-Aboriginal offenders. Male and female inmates should also be compared in light of the trends toward, and proven effectiveness of, gender-informed programming (Blanchette, 2001). The increased predominance of offenders with mental health needs in the changing federal offender population is also a notable area of interest in future profiling research given the Service’s intention to focus greater attention on these offenders (CSC, 2008; CSC, 2006b). As a related responsivity consideration, the personality profile of the segregated population could also be explored as personality may impact offenders’ ability and predisposition to seek and/or cope with segregation (Zinger & Wichmann, 1999). Only one study to date has examined the personality characteristics of offenders in administrative segregation. The assessment of segregated (i.e., both voluntary and involuntary) offenders’ personality can highlight their strengths and weaknesses (e.g., impulse control vs. disinhibition) and, in accordance with the responsivity principle of effective correctional programming (Andrews & Bonta, 2003), has the potential to guide the types of programs and/or treatment strategies selected. Lastly, the reasons offenders voluntarily enter segregation may provide further insight into this group’s unique needs.

Incorporating these suggestions into future research on segregation would provide information that could be used to develop a protocol aimed at identifying offenders at-risk for placement in administrative segregation. The advent of such a tool would be useful for identifying this group early in their sentence so that their risk could be minimized by ensuring that their areas of need are targeted through appropriate interventions. This would allow correctional personnel to take a proactive approach to offender management, as well as help CSC meet its strategic priorities of enhanced capacities to address the mental health needs of offenders and safe and secure institutions (CSC, 2006b). For offenders already residing in segregation units, knowledge of their characteristics would provide insight into the types of programs that are needed in these units but often unavailable. This would foster inmates’ reintegration into the general offender population and, ultimately, their safe transition into the community.


Andrews, D.A., & Bonta, J. (2003). The psychology of criminal conduct (3rd ed.). Cincinnati, OH: Anderson Publishing Company.

Arbour, L. (1996). Commission of inquiry into certain events at the Prison for Women in Kingston. Ottawa, ON: Public Works and Government Services of Canada.

Blanchette, K. (2001). Effective correctional practice with women offenders. In L.L. Motiuk & R.C. Serin (Eds.), Compendium 2000 on effective correctional programming (pp. 160-173). Ottawa, ON: Correctional Service of Canada.

Boe, R., Nafekh, M., Vuong, B., Sinclair, R., & Cousineau, C. (2003). The changing profile of the federal inmate population: 1997 and 2002. Research Report No. R-132. Ottawa, ON: Correctional Service of Canada.

Correctional Service of Canada. (2005). Performance report for the period ending March 31, 2005. Ottawa, ON: Correctional Service of Canada.

Correctional Service of Canada. (2006a). The changing federal offender population: Profiles and forecasts. Retrieved May 25, 2007, from

Correctional Service of Canada. (2006b). 2006-2007 report on plans and priorities. Ottawa, ON: Correctional Service of Canada.

Correctional Service of Canada. (2008). Transforming corrections to strengthen public safety. Special issue: Moving towards transformation. Let’s Talk, 33. Retrieved July 17, 2008, from

Corrections and Conditional Release Act. (1992).

Costa, P.T. Jr., & McCrae, R.R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.

Gendreau, P., Tellier, M.C., & Wormith, J.S. (1985). Protective custody: The emerging crisis within our prisons. Federal Probation, 44, 55-63.

Head, D. (2006, May). 2005-2006 year-end review. Executive Development Symposium, Correctional Service of Canada. Retrieved June 4, 2007, from http://infonet/pa/documents/2005-06_YearEndReview_e.ppt#1

Kane, D. (1997). Commitment to legal compliance, fair decisions and effective results. Ottawa, ON: Correctional Service of Canada.

Luciani, F. (1997). The security classification of offenders: Development of security reclassification protocols. Ottawa, ON: Correctional Service of Canada.

Motiuk, L.L., & Blanchette, K. (1997). Case characteristics of segregated offenders in federal corrections. Research Report No. R-57. Ottawa, ON: Research Branch, Correctional Service of Canada.

Motiuk, L.L., & Blanchette, K. (2001). Characteristics of administratively segregated offenders in federal corrections. Canadian Journal of Criminology, 43, 131-143.

Motiuk, L.L., & Serin, R.C. (1998). Situating risk assessment in the reintegration potential framework. Forum on Corrections Research, 10, 19-22.

Motiuk, L.L., & Vuong, B. (2005). Homicide, sex, robbery and drug offenders in federal corrections: An end-of-2004 review. Ottawa, ON: Research Branch, Correctional Service of Canada.

Mullins, P. (2005). Aboriginal male offenders in federal custody: Risks, needs, and between-group comparisons. Issues of Substance Canadian Centre on Substance Abuse National Conference 2005. Retrieved July 20, 2007, from

National Parole Board. (2002). National parole board policy manual. Ottawa, ON: National Parole Board.

Public Safety and Emergency Preparedness Canada. (2005). Corrections and conditional release statistical overview. Ottawa, ON: Public Safety and Emergency Preparedness Canada.

Solicitor General of Canada (1998). CCRA 5 year review: Administrative segregation. Ottawa, ON: Public Safety Canada.

Trevethan, S.D., Moore, J.P., & Rastin, C.J. (2002). A profile of Aboriginal offenders in federal facilities and in the community. Forum on Corrections Research, 14, 17-19.

Wichmann, C., & Nafekh, M. (2001). Moderating segregation as a means to reintegration. Forum on Corrections Research, 13, 31-33.

Wichmann, C., & Taylor, K. (2004). Federally sentenced women in administrative segregation: A descriptive analysis. Research Report No. R-158. Ottawa, ON: Research Branch, Correctional Service of Canada.

Wormith, J.S., Tellier, M.C., & Gendreau, P. (1988). Characteristics of protective custody offenders in a provincial correctional centre. Canadian Journal of Criminology, 30, 39-58.

Zinger, I., & Wichmann, C. (1999). The psychological effects of 60 days in administrative segregation. Research Report No. R-85. Ottawa, ON: Research Branch, Correctional Service of Canada.

1 For further information on administrative segregation in federal corrections, please refer to Commissioner’s Directive 709: Administrative Segregation, which can be obtained at:

2 Reintegration potential refers to the likelihood that an offender will be successfully reintegrated back into society, and is rated as low, moderate, or high. Male offenders’ reintegration potential is based on a combination of ratings designated by the Custody Rating Scale (CRS), the Statistical Information on Recidivism (SIR) Scale, and the static/dynamic component of the offender intake assessment process.  For women offenders, reintegration potential is based on scores on the CRS and the static/dynamic risk levels ascertained at intake.