Canadian Prison Environments: A Mixed Methods Analysis

Introduction

Better environments (social climates) are thought to be one way to improve prison life for residents and staff, leading to safer correctional facilities and more opportunities for rehabilitation (Auty & Liebling, 2019; Liebling & Arnold, 2004). Building from the organizational psychology literature, the prison environment (PE) has been described by scholars as an interaction of physical, social, and emotional conditions as perceived by residents (Liebling & Arnold, 2004; Moos, 1975). PE is most often measured by surveying inmate (and sometimes staff) perceptions of order, safety, opportunities for programming, and inmate–inmate and staff–inmate relations. Qualitative research is less often used for this purpose, and multiple or mixed methods analysis is also infrequently applied. In the current article, we add to the PE literature by comparing quantitative and qualitative assessments from inmates residing in eight different correctional facilities in the eastern Canadian jurisdiction of Atlantic Canada.

Combining quantitative with qualitative methods strengthens research by expanding the theoretical scope of analysis, providing a more thorough study of the phenomena, and offering a means to triangulate findings. This strategy also improves on the overall validity of the work by providing context otherwise lost with strictly quantitative assessments (Creswell, 2007). Further, direct comparison of diverse institutions allows for a more deliberate viewing of PE differences and documents why the differences may matter to incarcerees. Our comparative research addresses an important critique by Auty and Liebling (2019): “…limitations of existing prison effects research is the failure to adequately distinguish between prisons which may damage and repair” (p.359, emphasis added). Finally, the external validity of prison climate research is enhanced by studies occurring in different jurisdictions across the globe. To our knowledge, this is the first PE survey study conducted in Canada that uses items from a common PE instrument, in this case the Measurement of the Quality of Prison Life (MQPL). Our study will provide a comparison between eight correctional facilities, illustrate that there are differences between facilities, and show that more positive climates can be aspired to by prison officials.

Prison Environments Matter

Proponents of more positive PE argue that good social climates will result in less misconduct, better institutional compliance, more successful rehabilitation, and improved mental health for individuals in correctional institutions. Theoretically, the link between positive climates and behavior change stems from theories that view identity change as dependent, to a large degree, on the local climate supporting such change (Auty & Liebling, 2019; Barquin et al., 2019; Bressington et al., 2011; Liebling & Arnold, 2004). Fair and respectful treatment by staff, feelings of order and safety, and opportunities for change are integral to facilitating transformation within the individual. Thus, in addition to PE research, studies examining the legitimacy of prison regime and staff, as enhanced by the procedurally just treatment of inmates by staff, provide indirect support to the notion that PEs matter, and that day-to-day operation and staff behavior are key (Bottoms, 1999; Hacin, 2018). When individuals view institutional authority as legitimate and believe that they are being treated decently and fairly—are respected, listened to, feel safe and part of a community—then they are more amenable to change.

PE can measure a number of different physical and psychological domains, but generally these overlap with legitimacy and procedural justice constructs, such as physical conditions, access to visits, and (particularly with procedural justice) staff–inmate relations. The point is that studies that assess inmate perceptions of social climate, legitimacy, and procedurally just treatment have found favorable results. These include lower rates of institutional misconduct (Bosma et al., 2020; Tonkin et al., 2012), better mental health (Beijersbergen et al., 2014), a lower likelihood of death by or attempted suicide (Fitzalan Howard & Wakeling, 2020; Liebling, 2011), more prosocial and positive treatment attitudes (Bressington et al., 2011; Day et al., 2012; Franke et al., 2010; Stasch et al. 2018) and less frequent recidivism (Alladin & Hummer, 2018; Auty & Liebling, 2019; Beijersbergen et al., 2016). All such findings are beneficial for incarcerees and staff, creating a safer work and living environment.

A favored approach to the evaluation of PEs is the use of scale measures. Processes here involve having inmates fill out questionnaires that ask them to rate features of institutional life, usually on a 1 to 5 Likert scale. Since the Moos Correctional Institutions Environment Scale was designed in the 1970s, several efforts have been made to improve on its psychometric properties. Researchers in the United States developed the Prison Environment Inventory (Wright, 1985), and the Federal Bureau of Prisons produced the Social Climate scale, or PSCS (Camp & Gaes, 2005). A relatively brief 15-item instrument suitable for use in forensic settings as well as prisons is the Essen Climate Evaluation Schema (Tonkin et al., 2012). Arguably the most popular (and published) PE scale is the MQPL, developed by Liebling and her colleagues in the United Kingdom through appreciative inquiry (i.e., interviewing prisoners on what they felt was important to them). In his review of PE instruments, Tonkin (2016) observed that the psychometric properties of most survey items actually fell short on measures of internal consistency, factor structure, reliability, and responsiveness. In particular, despite their frequent use, researchers rarely scrutinize instruments such as the PSCS and MQPL for psychometric properties.

Recently, researchers from the Netherlands developed the Prison Climate Questionnaire, which has guided a number of the authors’ Life in Custody studies (Bosma et al., 2020). In comparing their instrument to others and assessing the importance of psychometric properties, Bosma and his colleagues contend that the MQPL has generally not been solely concerned with inmate ratings. Researchers using this scale typically integrate a qualitative approach (open-ended interviews, observation) to help understand the dynamics of a prison setting. This point is most salient to the current study as we took such a mixed methods approach.

While not common, there have been several comparative PE studies of note. In a ground-breaking qualitative study, Sparks et al. (1996) found differences between two UK facilities in their approach to maintaining order in the prison, one being more coercive than the other, but producing greater feelings of prisoner safety. In their research on five English prisons, Liebling and Arnold (2004) found some institutions were rated consistently higher for quality of program, staff relations, and other features. Day et al. (2012) in Australia confirmed a more receptive climate for rehabilitation by staff among inmates in a treatment-oriented facility in contrast to a mainstream institution. Fitzalan Howard and Wakeling (2020) found differences in prisoner PE perceptions between prisons and less self-harm and attempted death by suicide. Finally, in a recent US study of three prisons, Williams et al. (2019) found higher rates of cohesion at a medium security institution and a more program-oriented maximum security institution, when compared to a maximum security prison focused on control. Qualitative data assisted the investigators in identifying several barriers for staff to improve PE, such as limited staff–inmate communication, the negative impact of linear architecture in older prisons, and the impact of overcr 2016 owding on staff–inmate relationships.

The Current Study

Given the authors’ previously conducted qualitative inquiry using in-depth interviews to examine subculture and prison life in Canada's provincial prisons (Ricciardelli, 2014, Weinrath, 2016), this study was intended to build on this past work by combining survey data and qualitative response from inmates housed in Canada's Atlantic provinces. Atlantic Canada includes the provinces of Newfoundland and Labrador, Prince Edward Island, Nova Scotia, and New Brunswick, which comprise 6.4% of Canada's population (Statistics Canada 2021). Residents of Atlantic provinces (too often considered “have not” provinces insofar as they are not as economically strong as the others) have a median annual income of $62,000 compared to the Canadian average of $72,000 (Statistics Canada, 2021). Primary industries are fishing, offshore oil, logging, and tourism. These provinces usually have the lowest crime rates in Canada (Statistics Canada, 2021), although some uptakes in crime counts can also be attributed to the smaller population and greater practice of police reporting (i.e., persons are more likely to report all crime, including petty crime, to the police).

Canadian provincial correctional institutions hold pre-trial detention cases and prisoners sentenced to a maximum of 2 years less a day, although most sentenced prisoners serve 30 days or less, and have been the study of minimal research in Canada. Instead, most researchers focus on federal inmates housed by the Correctional Service Canada in prisons in larger metropolitan and populated jurisdictions. There is a certain irony in this, as provincial correctional institutions house more prisoners overall than federal corrections facilities (Malakieh, 2020). In a comparative sense, Canadian provincial institutions are quite similar to jails in the United States, as both hold pre-trial detention cases (i.e., remanded prisoners to custody who are awaiting trial or sentencing). Sentenced prisoners, however, serve shorter custody periods (in most American states a year or less, in Canada less than 2 years). Atlantic Canada's provincial correctional institutions suffer similar problems to other Canadian institutions: an imbalance of remand to sentenced prisoners (around 50%–50%) and “incompatible” or protective custody (PC) prisoners. These factors limit movement and programming within institutions. There is also an overrepresentation of Indigenous prisoners, although the overrepresentation is not as pronounced as in Canada's central Prairie regions. Most of the institutions in our study were built in the past 20 years, and are multipurpose: that is, they hold minimum/medium/maximum prisoners, as well as pre-trial and sentenced prisoners, they offer limited programming, and have general population units but also specialized units such as low risk or PC.

Methodology

Researchers surveyed eight provincial prisons in Atlantic Canada. Inmates were not typically separated by remand or sentence status; they were mixed on the units. The survey instrument used 51 items adapted from the MQPL and added three open-ended questions about institutional performance.

Procedure

Prior to initiating the study, research ethics approvals were obtained from Ethics Review Boards at three universities. In seven of the eight institutions, correctional staff provided posters advising of the study and its confidentiality. A group administration approach was adopted in the institutions to maximize sample size, meaning that the researchers visited each prison unit, informed inmates about the study, and asked for interested volunteers. Potential participants could then provide informed consent and complete the survey, which was disseminated in hard copy along with pencils and erasers (in one institution administrators approved inmates’ keeping the pencils and erasers).

Participants would complete the survey sometimes on the unit, or, if available, in a room on the unit, or be escorted to another room (e.g., library and program room). Respondents were discouraged from collectively discussing their survey answers. For the few who appeared to struggle with the survey, assistance with the questions was provided by the researchers (i.e., transcribe responses as dictated due to literacy needs), or the aid of another inmate was enlisted. The researchers provided contact information to those interested in study results. The respondents largely welcomed the opportunity to give feedback and were enthusiastic about participating. As well, staff were supportive of the research.

Because of confidentiality agreements with the participating correctional jurisdictions, we cannot report correctional institution populations. The largest facility could hold 370 prisoners, the smallest 56, with the average capacity 167. Study samples ranged from 18 to 87 and averaged 58. Response rates are based on inmate availability, which was far less than the daily total (e.g., they could be in court, working, in medical). The response rates ranged from 59% to 83%, and averaged 70%, within acceptable limits for inmate surveys (Mackenzie, 2006). We entered response rate by institution as a variable in multivariate analysis, which did not influence any of the prisoner ratings.

Survey Measures

We reviewed all MPQL questions with prison administration, which provided approval prior to study commencement. Administrators reported that the items had strong face validity. Researchers chose to use 51 items rather than the 100 + questions representing the total scale, in part due to the perceived utility (why ask more questions than necessary) and to shorten time for study completion. The scale items measures and internal consistency as ranked by Cronbach's alpha coefficients are described. The scale averages and variances are displayed below, after they were created by summing each indictor and then dividing by the number of items to maintain the range from 1 to 5. Higher scores indicate more favorable ratings of prison features.

Personal Development was a summary measure of four questions on inmate perceptions of opportunities to work or undertake self-improvement (i.e., my participation in programs has helped me develop, I am being helped to lead a law-abiding life, I am motivated to be involved, my time here seems like a time to change). A reliability analysis was run, with a relatively weak alpha (α = .63). The variables were summed and divided by four to create a personal development scale (mean = 2.90, SD = .87).

Fairness of the prison regime was based on seven questions on rules and their enforcement (i.e., privileges given and taken away fairly, staff clear in expectations, disciplinary system is unfair [reversed], staff inconsistency with rules [reversed], inmates treated fairly when staff apply rules, easy to obtain information from staff, staff slow to respond to inmate requests [reversed]). The scale alpha was an acceptable .70, with the summed scale displaying a mean of 2.62 (SD = .71).

Staff Relations and the quality of those relations were derived from six questions (e.g., I get on well with staff, relationships with correctional officers [COs] are good, I get support from COs when I need it, I trust COs, Staff display honesty and integrity, I feel I am treated with respect by staff). The standardized alpha was good (.84), and the summated variable derived from these six items produced an average rating of 3.03 (SD = .84).

Prisoner-Support was comprised of inquiries assessing how inmates care about each other (i.e., I get on well with the other prisoners in this prison, prisoners here generally do not help and support each other [reversed], I feel that I am treated with respect by the other prisoners in here, I don’t trust other prisoners in here [reversed], I don’t have friends in this prison I can go to for help [reversed], I don’t trust other inmates [reversed], I am treated with respect by other inmates, Inmates talk about being on the straight and narrow). The seven-item scale managed a weak reliability coefficient (α=.60) and averaged a rating of 3.03 (SD = .61).

Visitation/Family questions concerned visiting access and quality (I am able to receive enough visits, My visitors are treated well, Visits are long enough, I am able to maintain meaningful contact with my family). The internal consistency for this four-item scale was weak (.58) and the rating relatively low (mean = 2.84, SD = .79).

Wellbeing items captured the environmental perception of unit life through general questions such as: The atmosphere on this unit is relaxed and friendly, My experience on this unit is painful (reversed) and I feel tense on this unit (reversed). This three-item indicator achieved a modest reliability alpha (.64) and averaged 3.0 (SD = .94).

Order assesses the PE by using items that address the structures within an institution (e.g., The unit is good at delivering structured routine, This correctional center is well organized, This correctional center is good at maintaining order). The scale showed good internal consistency (α=.71) for three items and attained a low average rating of 2.6 (SD = .96).

Safety is assessed using items that measure fear (e.g., It is very unlikely for an inmate on this unit to be assaulted, I fear for my physical safety [reversed], Some inmates are afraid of other inmates [reversed], I feel safe from being injured, bullied or threatened). These four items showed modest internal consistency (α= .57) with a mean scale response of 2.8 (SD =7.9).

The other method used to assess inmate perception of the environment was qualitative. There were three open-ended questions that were asked in a very general way to allow respondents considerable scope in their responses:

What things do you believe are being done well at this correctional center?

Descriptive and Control Variables

We captured a number of variables in the survey for descriptive purposes and as potential controls in later analysis (Beijersbergen et al., 2016; Bosma et al., 2020; Brunton-Smith & McCarthy, 2016). Variables associated with PE rating and misconduct such as age (in years), gender (male, female), sentence status (remand, sentenced), time served (30 days or less, more than 30 days), residence in a higher risk unit (low-risk unit, general population, PC unit, higher risk) were collected using prisoner self-report on the questionnaires.

Analytical Plan—Quantitative Data

Analytical Plan—Qualitative Data

Recognizing responses to the open-ended items varied in length, from just a few words to more detailed descriptions and examples, an emergent theme analysis that was inductive was employed (Hesse-Biber & Leavy, 2004)—that is, a constructed semigrounded emergent theme analysis (Charmaz, 2014; Glaser & Strauss, 1967). Data were coded into primary emergent themes and then, using axil coding (Corbin & Strauss, 2014), broken down into secondary and tertiary themes. While always controlling for the institution understudy, data from all prisons were combined into one overarching consolidated file to examine similarities and differences across facilities. In working with the combined dataset, we embedded ourselves in the whole data and examined central themes across responses. In presenting the results, quotes are used to illustrate the themes, highlighting participant voices, and centralizing their experiences—but always ensuring the quoted words did not compromise the privacy or identity of any participant. Quotes were edited for grammar and spelling but never compromising meaning or tone.

Results

Means Comparison

Overall, prisoner ratings were generally either in the middle or below 3.0, indicating generally lower overall ratings of the PE (Table 1). Custody was not perceived as a positive experience. The highest grand mean rating was an average of 3.08, close to the “not sure” rating of “3.” Five of the eight ratings were below 3, indicating a fairly consistent negative response by Atlantic housed prisoners for personal development, fairness, order, family visitation, and safety. Staff relations, prisoner support, and wellbeing were rated at around 3, which although more positive, was not a strong endorsement of prison conditions in Atlantic Canada.

OPEN IN VIEWER

Table 1.

Prison Environment Mean Scores for Personal Development, Fairness, Staff Relations, Prisoner Support, Order, Family/Visitation, Wellbeing and Safety.

		Correctional institutions
Scale indicators	Grand mean	1	2	3	4	5	6	7	8
Personal development	2.90	3.19	3.19	2.98	2.88	3.07	2.75	2.88	2.64
	.87	.82	.92	.87	.88	.85	.86	.89	.87
Fairness	2.62	2.92	2.80	2.79	2.73	2.74	2.65	2.34	2.62
	.71	.72	.70	.78	.75	.74	.63	.65	.71
Staff relations	3.08	3.28	3.32	3.15	3.21	3.11	2.85	3.05	2.73
	.84	.82	.90	.84	.60	.75	.80	.89	.84
Prisoner support	3.04	3.04	2.99	3.12	3.10	3.11	3.0.2	2.95	3.04
	.61	.42	.68	.59	.71	.63	.53	.73	.61
Order	2.65	3.20	2.97	2.82	2.89	2.85	2.77	2.10	2.29
	.96	.94	1.04	.89	.93	.86	.92	.83	.96
Family/visitation	2.84	3.01	3.06	2.95	3.08	2.92	2.73	2.63	2.84
	.79	.90	.70	.86	.63	.68	.82	.82	.79
Wellbeing	3.00	3.33	3.15	3.17	3.07	3.28	3.15	2.61	2.65
	.94	.88	1.02	.95	.97	.83	.84	.97	.94
Safety	2.81	3.11	3.13	2.91	2.81	2.80	2.86	2.62	2.56
	.79	.62	.82	.72	.89	.68	.70	.96	.79
N	461

Note. Standard deviations are italicized.

There were persistent and clear differences between Atlantic institution ratings of environment. Institutions 1 and 2 were consistently either above a 3 average, or at approximately 3. Institutions 3, 4, and 5 varied more in scores, but were often below 3. Prisons 6, 7, and 8 were consistently lower than others, mostly below 3, with particularly low scores for institution 8.

One of the ways to think about differences is to compare the mean institutional score to the grand mean and convert it to a percentage (Table 1). For example, the grand mean for personal development is 2.90, while correctional institution 1 averaged 3.19. The difference is .29, which is (.29/2.90), an improvement of 10%.

Using that metric, our first two prisons are often 10% or higher than the average (eight of 16 outcomes), and in only two cases are the differences lower than 5%. Not surprisingly, the middle group of correctional institutions tends to score close to the average, with only seven out of 24 cases showing greater than a 5% drop from the overall mean. Finally, the last three institutions are frequently lower by 5% to 21% (13 out of 24 comparisons).

Multivariate Analysis

To augment the means comparisons and control for other influences, an OLS regression was run (Table 2). To alleviate some of the limitations due to sample size, prisons were grouped into high (1 and 2), moderate (3, 4, 5) and low (6, 7, 8) performing. The moderate group was used as the reference category.

OPEN IN VIEWER

Table 2.

Prison Environment Scale Scores by ^b Age, Sentenced, Time in Custody, Lower Risk, Protective Custody, High and Low Performing Prisons ^c .

	Personal develop	Fairness	Staff relations	Prisoner support	Order	Family/visits	Wellbeing	Safety
Constant	3.275	3.04	3.077	3.756	3.257	2.849	3.616	3.482
	(.248)	(.200)	(.229)	(.175)	(.259)	(.228)	(.263)	(.222)
Age	−.004	.001	.011***	−.008**	.006	.003	−.009*	−.010***
	(.004)	(.003)	(.003)	(.003)	(.004)	(.003)	(.004)	(.003)
Sentenced	.152 a	.206***	.275***	.161**	.180*	.057	.356***	.216**
	(.084)	(.067)	(.078)	(.059)	(.088)	(.077)	(.089)	(.075)
Time in custody	−.234*	−.277***	−.380***	−.222**	−.418***	−.088	−.177	−.201*
	(.106)	(.085)	(.098)	(.075)	(.111)	(.097)	(.112)	(.095)
Lower risk	.262**	.214**	.169 a	−.073	.320**	.115	.241*	.077
	(.101)	(.081)	(.093)	(.071)	(.105)	(.093	(.107)	(.090)
Protective Custody	.293**	.152 a	.391***	−.015	−.026	.155	.048	−.055
	(.102)	(.081)	(.094)	(.072)	(.106)	(.093)	(.108)	(.091)
Higher performing prisons	.298*	.154	.314**	−.128	.238 a	.150	.018	.221*
	(.126)	(.101)	(.117)	(.089)	(.132)	(.116)	(.134)	(.113)
Lower performing prisons	−.125	−.226***	−.121	−.124 a	−.433***	−.174	−.357***	−.175*
	(.097)	(.077)	(.089)	(.068)	(.101)	(.089)	(.102)	(.086)
Adjusted R2	.06	.09	.12	.04	.14	.03	.09	.07
N = 461

Note. Standard errors are in parentheses.

a

p < .05 one-tailed, for two-tailed *p < .05, **p < .01, ***p < .001.

b

Sentenced (1 = sentenced, 0 = remand), time in custody (1= > 30 days, 0 = 30 days or less), low risk (in low-risk unit = 1, no =0), protective custody (PC) (1 = in PC unit, 0 = no), reference category for low risk or PC unit placement is general population placement.

c

Higher performing (Yes = 1, No =0), low performing (Yes = 1, No =0), reference category is middle performing.

The strength of the equations was weak to modest; the variance explained ranging from .03 to .14. The most variance explained was for the concept of order (R²=.14), with fairness and wellbeing tied for second (R²=.09).

The higher performing correctional institutions showed effects mostly in the predicted direction. Compared to the middling prison group, higher scores are evident in six of the eight environmental domains, with four of those statistically significant. In the summary that follows, all p values are two-tailed unless noted otherwise. Personal development (b = .298, p < .05) and staff relations (b = .314, p < .01) showed the strongest and most stable impacts for higher performing prisons. Perceptions of order (b = .238, p < .05, one-tailed) and safety (b = .221, p < .05) likewise favored higher prisons, although not as strongly. Prisoner support showed a negative but nonsignificant effect (b = −.128 ns), while fairness and family/visitation (b = .150 ns) had smaller nonsignificant effects, and wellbeing was almost nil.

Lower performing prisons more consistently showed effects in the predicted negative direction compared to the middle group, with five of the eight effects statistically significant at p < .05 one-tailed or lower. The strongest negative effects were for order (b = −.433, p < .001) and wellbeing (b = −.357, p < .001). More moderate effects were seen for fairness (b = −.226, p < .001), prisoner support (−.124, p < .05 one-tailed), and safety (b = .175, p < .05).

For our controls, net of the effect of other variables, age did not have consistent effects across environmental measures, although effects for staff relations, prisoner support, safety, and wellbeing are in the expected direction. Older prisoners are less involved in conflict and better able to cope, which, as such, is not surprising that they report better staff relations, prisoner support, and higher wellbeing and feelings of safety. Sentenced offenders, who are compared to pre-trial detention cases, tend to report more positive perceptions of the PE; and although these effects were only moderate, they were statistically significant in seven of the eight domains, with a particular strong impact on wellbeing (b = .356, p < .001). In the case of time served, familiarity apparently does not breed high ratings, as inmates spending more time in prison tended to perceive the environment more negatively, with both moderate and stronger inverse effects observed. Inmates housed in lower-risk units had ratings of effects that were statistically significant for personal development, fairness, and staff, and were in the predicted direction in seven of the eight domains, displaying modest to small relations, order and wellbeing. Finally, prisoners in PC units generally did not show much difference in their PE assessments than other participants, but did show higher, statistically significant ratings for personal development, fairness, and staff relations.

What Things do You Think are Being Done Well at This Correctional Centre?

Across institutions, 293 participants responded to the item; however, 138 participants provided negative or ambiguous feedback. Thus, the analysis is based on the 155 inmates who responded to the item as asked (i.e., positively). Positive findings did vary across institutions, with female inmates being most positive about their PE. Positive feedback, such as “all good” or “almost everything” was provided by a small number of respondents in nearly every institution.

In six of the eight facilities, there were some reports of being satisfied with food in terms of taste and portion (“food in this jail is very good and tasty!”). Only in one institution was access to physical exercise, which was described as “fair” and the “gym” was “ok,” reported positively.

As well, there were some respondents who were happy with programming and schooling offered (e.g., “school programming”; “new programs are being introduced all the time along with groups of volunteers coming in to talk”) and available resources (e.g., “church visits, General Equivalency Degree, good library”). Others noted that “order is maintained quite well” within the institution, specifically the “routine” was upheld (e.g., “keeping order and everything under control”). In select institutions, particular units were described as quite positive (e.g., “Unit seven is almost like rehab, you have a guard that does programs with the inmates”, “Unit seven is the best unit. Gives you the opportunity to reflect, things should be more like unit seven”).

Some participants felt that communication between staff and inmates was “good” (e.g., the relationship between guards and inmates is good, get along but still stern for rules”). Quite noteworthy, in a few institutions, participants spoke positively about staff, writing that staff were excellent at “accommodating our needs,” were “great,” and provided fair treatment. Here, a respondent wrote “I feel that actual care for the inmates on a human level is here.” Other positive feedback tied to staff ranged from how staff provide “assault responses,” described as effective, helped prisoners (e.g., “Things done well are CO's helping inmates, captains very professional and also lend professional help to inmates on rules”), and treated them well (e.g., “In general they seem to try”). It is noteworthy that in one institution of very low ranking, three inmates voiced concerns for staff safety and treatment (e.g., “Safety is a concern with staff, more so for their safety than with the inmates”).

Very rarely, but across most institutions, a few provided positive comments about healthcare (e.g., “health care is getting better,” “addiction and mental health counseling”). In select institutions, participants noted learning about mental health, and others stated the nurses were very caring.

Overall, what incarcerees valued in PEs was consistent across institutions, although the degree of positive reporting varied depending on where an inmate was housed. Staff relations, programming, and routine were most consistently described positively. Less regularly, healthcare, physical activity, and food received positive feedback as noted below.

What are Your Most Serious Concerns at This Correctional Centre?

In total, 359 participants provided responses voicing their most serious concern about the institutional environment. In seven institutions, those providing feedback expressed concern about the quality and quantity of food provided (e.g., “Food trays are small and you are always hungry at night,” “Most terrible food served in the province”). Another concern voiced was about available canteen, specifically the cost of canteen and the available selections.

In all eight institutions, participants overwhelmingly reported concerns about poor treatment from staff (including “Bullying,” “Unfair treatment,” and “The guards lied to me”) a lack of communication between staff and inmates, and inconsistency from staff. Others wrote about harassment from staff directed to prisoners (e.g., “Extreme favoritism to specific inmates”; “Guards are disrespectful”) and between staff (e.g., “When a new guard comes here to work they are polite to us at first…. I’ve seen a group of bad guards pick at a good guard for being nice to inmates”). Some reported feeling judged by staff as well as taunted. In three institutions, poor treatment from staff included “racism,” the racism was directed at either English-speaking versus French-speaking prisoners or tied to ethnic identity.

Survey respondents also voiced concerns about the amount of time they spent “Locked up” in their cells, due to either staff shortages (here, participants voiced concerns about staff retention, with some noting that the institution was “Short staff” and lacked the “Ability to maintain staff”) or being locked up for another inmate's negative behavior. In one institution, participants voiced concerns about time spent in lock-up or solitary (e.g., “The most serious one is how they will keep you in solitary confinement without a good reason”). In two institutions, respondents actually voiced being physically assaulted by staff (e.g., “I took a bad beating from correctional officers that left me disabled for life…my bones were broken, fingers dislocated, crushed bones”). Participants voiced a need to “Get rid of the bad apples” among the correctional staff. Finally, in select institutions, those providing feedback were concerned that their requests were left unmet by staff (e.g., “Being able to get the stuff that I’ve requested,” “No response to most requests submitted on offenders request form”).

Of note, in four institutions, concerns were voiced about violence on the units and fear for their safety. Specifically, respondents talked about being housed with “Incompatibles” or “Inmates in for breaching are mixed with people in for serious violence and killing.” Some vocalized concerns about “Getting stabbed, eyes poked out” and the process around intervening in assaults.

In seven of the eight institutions under study, there was concern about healthcare (e.g., “Not providing proper health care,” “The wait times are insane”) and specifically psychiatric help (e.g., “Not enough psychiatric help. Not enough social workers time.”). Respondents were concerned about healthcare provision (e.g., “Getting medications properly and addiction help. Issues regarding healthcare also included how they felt treated by staff (e.g., “The psychologist—real asshole talks down to everyone and is very arrogant”) and the access to services (e.g., “Medical, and the time it takes to see a doctor here to be on the proper medication”). In one particular institution, participants described mental healthcare as in crisis (e.g., “Mental health is a crisis here!!! Help us inmates”) and specifically about the lack of prescription medication available (e.g., “They took me off all my meds to lock me up with no rehabilitation”; “Being cut off meds that are prescribed by other doctors”). In select institutions, there was also concern about dental care (e.g., “Right now my tooth is in extreme pain and there is nothing I can do but wait months”).

Overwhelmingly, participants wrote about inadequate programming in the institutions (e.g., “More programs”), including programs for self-betterment (e.g., “Lack of opportunity for programs to better myself for the outside,” “No rehabilitation has been offered to me. No programs have been started for me despite requests.”), as well as work (e.g., “No opportunities to make the best of this time, e.g.,—work to make money for release”). Here, they also noted that, without programming, they were not getting the “Help” they required to “Get ready for the streets.” Unique to one institution is the number of respondents who were anxious about re-entry and the lack of provision of “Required tools to adequately succeed in our reintegration back into society again.” Here, the problem was about being “Released receiving no or next to no help for their issues, that is, addictions, anger management, etc.” or “Being released in the middle of nowhere.” Some felt the lack of activities available created boredom and opportunities for poor behavior (e.g., “low amount of activities,” “there is not much to do so boredom often raises the chance for trouble”). They voiced that they did not have support and programming that helped them navigate why they offended, instead feeling they were “Warehoused” or in “Simply a cage for human animals.”

In three prisons, a need for more activities was noted, drawing attention to a need for more yard time (e.g., “No yard time in this unit except bullpen which is concrete and a 2 foot step down and up—I can't do this easily. I'd like to be on grass instead of concrete”). In one institution, a need for a “gymnasium” was noted, and in another, participants spoke about the lack of rec time, with a particular emphasis on outdoor rec (e.g., “No outside rec”). The latter, no outdoor recreation, is concerning, given participants’ feedback in one particularly low-ranking institution, about not having access to fresh air or the ability to go outside for months at a time (e.g., “Outside fresh air”; “No access to fresh air between months October–May”).

In two institutions, the overall cleanliness of the facility was a complaint (e.g., “Cleanliness”; “Black mold on the walls of our unit”), clothing (e.g., “2 outfits for 1 week if laundry actually comes”; “Laundry and clothing, they don’t give us enough or stuff for winter”), and self due to lack of hygiene products (e.g., “Bad supply of hygiene products”). Multiple inmates noted that the quality of the mattresses and the practice of double-bunking were serious problems.

In six institutions, those surveyed made a request for more contact with family (e.g., “Visits”; “Maintaining contact with family”) and fixing the phone system as well as making it more affordable which would make it easier for inmates to connect with family (“No being able to make calls to people that can't afford phone calls!”). Some made a request for “Help to make amends with family and keep positive relationships”).

What Things Would You do to Improve Things in This Correctional Centre?

In total, 345 participants had suggestions for improving their PE, requesting in all eight institutions more programming, including treatment and re-entry support programs (e.g., “We need treatment programs for addictions and mental health and more help with planning successful release plans…”). Respondents wanted anything to help “better themselves” and feel “valued,” but did request specific programs around managing addiction (e.g., “Get back to N. A. Meetings”) and anger (e.g., “Have more programs ex. A.A / A. Management”). Participants genuinely felt programming was necessary (e.g., “I think there should be more programs to attend to help inmates,” “I would work on programs for change. Even if one person got help it would be well worth it!”). A subset of the respondents asked for programs designed specifically for assisting with re-entry (e.g., “help give inmates a reason to want to change in life and… help on a release plan”). Here one spoke about how “it would be good to utilize rehabilitation for people who are willing to participate, maybe use unit 2 as a rehabilitation centre, or detox center for inmates who want to help themselves”.

Some desired more schooling (“Put new programs. Make GED available to everyone fairly”) and activities (e.g., “more activities to better ourselves”). There were those who also desired more equal access to programs (e.g., “Offer more programs to PC inmates”) and longer programs (“Increased duration of programs such as AA and implementation of new programs such as NA”). In three institutions, participants requested work (e.g., “more work for working unit), or work opportunities, specifically respondents wanted opportunities to learn more applied skills that they could use during re-entry to help prepare for community living.

Regarding material needs and substances, in eight institutions, there were requests for more food and healthier food (e.g., “A proper diet. Ours is carbs and sugar based,” “Food awful,” “Better food”). Often, here, respondents also asked for more canteen options and less cost (e.g., “lower canteen prices”) as well as more methods to deposit money into prisoner accounts.

At one low-ranking institution, a need for more cleanliness was expressed—and to deal with the “rodents” in the institution (e.g., “mouse traps”). In five of the eight prisons, participants requested new (e.g., “new beds as old ones are completely done”) and better shower beds or mattresses facilities. Some inmates did note they would like to have new clothing daily and more access to legal information. Also requested were “Personal TVs in cells,” being able to wear their “own attire,” pillows and linens, and “Internet access.”

In seven institutions, participants asked for more healthcare services (e.g., “have social workers, mental health and addiction services”) and better medical attention (e.g., “Being visited by or seen by our regular doctor”) and treatment (e.g., “More mental health and addictions help”). Others requested dental care (e.g., “Dental”), and medications (e.g., “Get meds under control”).

In five facilities, increasing opportunities to connect with family was mentioned (e.g., “I would love for mothers to be able to see their children/open visits”). Included here are those who asked for the opportunity to connect with an incarcerated partner and those who hoped for opportunities to connect without barriers (e.g., “I would like to see visits to be better for the ones who have kids. It's way too hard to have them come and see you thru a window. Should be able to hug them”). Some requested either contact visits in person, visits more frequent then once a week (e.g., “Increase visits, maybe get contact visits for people who have children”), or a better phone system that eases the effort required to call people in the community (e.g., “The phone system is horrible. You can't leave messages and it costs too much”). The lack of affordability of the phone system, especially long-distance calls, was noted across multiple institutions (e.g., “because some people don't have money like everyone else so some have to go without calling home”).

In six institutions, those surveyed requested more recreation time (e.g., “At least an hour for gym”; “I would like two gym spots”). They voiced that “Increased recreation would improve inmate wellbeing as well as physical and emotional health,” some clearly stated the need for more “Fresh air.” To assist with recreation time, more equipment (e.g., “A basketball net,” “Weight room,” “more cardio (currently none)”) was suggested. Perhaps an easier request to implement, participants in one institution requested that “yard time” be less early in the morning (e.g., “Giving yard different times each day would awesome. I feel that we are offered yard too early in the morning, so that no one goes to yard, it makes it easier on the staff”).

In seven prisons, participants wanted improvement in how inmates were treated by staff (e.g., “Try to get staff and inmates on the same page”), also suggesting a need for purging “The management of anyone with a ‘they’re only inmates, who cares’ attitude” and offering “Better training to the correctional officers.” Some desired equal treatment for all inmates, including more one-on-one interactions between prisoners and staff (e.g., “Move officer to inmate interaction”) and more communication, both more efficient and timely communication. Others felt that staff treated them poorly, which included that they were “not listened to” and that staff were “rude” and “disrespectful.” Concerns ranged from inconsistent implementation of rules and poor following of a schedule (e.g., “This place needs a scheduled routine followed,” “Rules either need to be enforced or eradicated”), the need for more staff working (e.g., “Have staff so ranges are open,” “More staff,” “Hire more staff”) or more staff training (e.g., “More trained staff and fully trained”).

In three institutions, participants desired less time locked-up, particularly due to staff shortages (e.g., “Less lockups due to staff such as ‘short staff’”) or unfair treatment (e.g., “When one inmate is locked for bringing something in they are not supposed to and the rest of us at times get locked in, on account of it. Find it is not fair for those who haven't done anything.”).

In two facilities, there were requests tied to improving safety. For instance a participant, echoing others, wrote “Place dangerous offenders away from say petty thieves. Keep sex offenders on a range of their own.” There was particular concern about the placement of inmates with diverse criminality on the same unit, personal safety, and about “Putting criminals together in one large room … when they are thrown into a survival situation their instincts are to get worse rather than better.”

In two institutions, respondents called for their more culturally specific needs be met, for instance to “Erect a tepee or Wig-wam at our Native sweat sites” or a “Healing” unit or “church.” At another prison, participants requested “second chances” for temporary absences (e.g., all kinds, like give second chances for TAs and programs “cause now if you screw up they hold it over you forever”).

Discussion and Conclusion

This study's findings highlight differences in inmate perceptions of the PE between institutions, with some facilities clearly rated higher by residents, and others clearly rated lower. Overall, however, ratings were on the low side (lower than 3.0 on 5.0 scale, the midpoint), with grand means illustrative of this weak performance. Of course, there was no consensus among incarcerees; a sizable minority voiced favorable positions toward staff and the general environment. But, the majority viewed prison life in a negative light. On a more optimistic note, there were clear differences between institutions, some sizable, that identified institutions where custody time was felt to be better spent. Efforts at better prison management can produce more positive results at the institutional level.

The qualitative data generally bore out our quantitative ratings. Institutions with more positive ratings produced more positive commentary and less critical statements. The use of open-ended questions provided important feedback on areas not covered in the survey, such as food quality and healthcare, as well as mental health and addiction services. Similar to Williams et al. (2019), qualitative data were found to be more helpful in identifying areas of concern that were priorities to address.

Inmates overwhelmingly wanted greater access to programming or work during the day. Opportunities to improve themselves or simply engage in constructive activity were the most important needs expressed. Greater access to family members was a concern at most institutions, particularly access without barriers (i.e., contact visits). Although it is true that most provincial prisoners serve short sentences, there are still many who serve over 30 days, and family contact can help manage the shock of incarceration (Pelvin, 2019).

Perhaps not surprisingly, food was a serious concern, including portions provided and the cost of canteen. Inmates appeared better educated on the importance of good nutrition and felt the lack of good food and small portions were punitive measures. In addition, the lack of quality healthcare also evoked strong reactions. Despite some inmate reports of good relations with correctional officers, most offenders believed that there needed to be better training, more efforts at staff retention, and a general improvement in their relationships with staff. While stated in different ways, procedurally just treatment was important to prisoners.

There were a pair of institutions where participants reported serious safety concerns, and these were among the lower-rated institutions. We viewed this as a serious issue, as criminalized people should not depend on luck in being placed in a prison that is free from threats or harm. At a minimum, Atlantic institutions should provide consistent safety for individuals housed there. ¹

There were limitations in this study. Despite a high participation rate, a selection effect might mean more negative or more positive inmates might have been overrepresented. The research also showed generally modest effects at the institutional level on PE indicators. That is, considerable variance went unexplained, meaning other factors impact prisoner perception of the prison climate. One-on-one surveys and face-to-face interviews might have produced more accurate and in-depth data than our group administration method.

In conclusion, we believe that this study has made several contributions to the literature. We compared multiple prisons and found substantive differences, contributing to this more limited sub-area of PE research. Study findings from Canada on PE using the MQPL provide another source of international data for correctional scholars researching this area. This investigation also shows the advantage of a mixed methods approach to prison research, not only to improve on the validity and trustworthiness of both quantitative and qualitative findings, but also to improve on the identification of policy directions for corrections officials. Finally, the study recommendations here are based on direct feedback from incarcerated individuals, an essential outcome for any PE study.