Research Reports

Secure Units for Women Offenders:
An Examination of Impacts

2008 N^o R-197

Renée Gobeil,
Kelly Taylor,
&
Jillian Flight

Correctional Service of Canada

October 2007

ACKNOWLEDGMENTS

Thank you to Antonia Sly, Nathalie Gingras, and Stephanie Wiggins for their help in data collection, and to Robin Westmacott for her work in preparation for site visits. Thanks also to Antonia Sly, Nathalie Gingras, Robin Westmacott, and Kim Jordan who assisted with the transcription of the interviews, and to Shannon Gottschall who aided in the coding of qualitative data.

A special thanks to Sara Johnson, Paul Verbrugge, and Kelley Blanchette who assisted in the preliminary development of measurement materials. Paul Verbrugge and Ben Vuong’s guidance in the development of the quantitative databases and in the quantitative data analysis process was also greatly appreciated. Kelley Blanchette also offered extensive editorial guidance.

The authors would like to thank the staff of the Women Offender Sector who provided support throughout this project, as well as staff at the institutions who assisted in the logistics of the site visits and data collection. Finally, special appreciation is expressed to the staff and women from across the country who agreed to provide feedback on their experiences of the openings of the secure units, thereby allowing for the successful completion of this project. We look forward to hearing from you again soon.

EXECUTIVE SUMMARY

Between January 2003 and October 2004, secure units were opened at four regional women’s facilities: Edmonton Institution for Women, Grand Valley Institution for Women, Joliette Institution, and Nova Institution for Women¹. The opening of these units was a result of extensive research and the development of the Intensive Intervention Strategy. The units were intended to address the higher risk and need levels of women classified as maximum security through intensive intervention in a secure environment. Moreover, the openings allowed women previously co-located in men’s institutions to be housed in women’s facilities and, in many cases, to be in their home regions.

The current report presents the results of two studies evaluating the impacts of the opening of the secure units on women classified as maximum security, staff members who work with them, and the other women and staff at each of the regional facilities. For both studies, data were collected both prior to the openings (from participants at the co-located men’s institutions and the regional women’s facilities) and after the openings (from participants at the regional women’s facilities only). In the first study, which focused on the respondents’ perceived impacts of the openings, women and staff responded to a battery of standardized measures and participated in surveys, interviews, and focus groups. For the second study, which provided an alternative perspective of these impacts, Offender Management System records for 506 women offenders were reviewed retrospectively for periods of six months prior to and after the opening of the secure units. This review focused on participation in pro-social activities and on institutional adjustment. Given that all post-test data were collected very shortly after the openings (approximately six months), results represent only the immediate impacts of the openings.

Results of these studies revealed that the opening of the secure units had important immediate impacts. For the women classified as maximum security, involvement in most pro-social activities did not increase after their transfers from the co-located men’s institutions; indeed, many of the women reported experiencing boredom and stress. Additionally, many of the women housed in the secure units indicated that the information and direction received from staff was sometimes inconsistent, leading to frustration, confusion, and tension. Nonetheless, there were no increases in the proportions of women involved in institutional incidents after the units’ opening. There were, however, more women with positive urinalysis results at this time point.

Staff members (both of the secure units and elsewhere in the institutions) were also impacted. They reported increased stress, frustration, and burnout, as well as decreased job satisfaction and morale. Staff members expressed that they would like to receive additional refresher training on topics such as use of force, as well as secure unit training for all institutional staff. Moreover, staff members indicated that they would like greater management support, and that there is divisiveness between the staff assigned to the secure unit and those working elsewhere in the institution. On the other hand, secure unit staff also reported the development of a strong sense of team cohesiveness and support.

Findings suggested that the functioning of the remainder of the women’s facilities (i.e., general population and structured living environments) was also impacted. The women classified as minimum and medium security reported increases in stress, tension, and interpersonal problems, as well as a limitation of their privileges. The women’s movement and access to staff, visits, and recreational activities were all restricted after the openings. Moreover, some reported that the atmosphere of the facilities became more rigid and security-focused.

The collective research results for this multi-method project led to recommendations pertaining to: 1) staffing at the women’s facilities, 2) communication, both between staff and management and between inmates and staff, and 3) re-evaluation of the secure unit initiative once the units have been open for a longer period of time.

Regarding the latter, it is relatively common for there to be a period when difficulties are encountered during implementation of new initiatives, particularly those as complex as the establishment of secure units at the women’s facilities. In addition, and as mentioned earlier, the data in this study reflect only the immediate impacts of the opening of the secure units. A number of changes have occurred since the opening of the units, including structural changes and clarification of relevant policies. A future re-examination would take account of these developments. Such an investigation would also have the added benefit of allowing for the inclusion of the secure unit at Fraser Valley Institution, which was not included in the results reported here.

TABLE OF CONTENTS

• ACKNOWLEDGMENTS
• EXECUTIVE SUMMARY
• LIST OF TABLES
• LIST OF FIGURES
• LIST OF APPENDICES
• INTRODUCTION
  • History of Women's Corrections
  • Research Informing the Intensive Intervention Strategy
  • Intensive Intervention Strategy
  • Current Study
  • Evaluation Strategy
• STUDY 1
  • Method
    • Participants
    • Description of Measures
    • Procedure
    • Analyses
  • Qualitative Results
    • Sample
    • Feedback
      • Operations
      • Staff and Staffing Issues
      • Programs
      • Security
      • Communication
      • Supplementary Feedback
  • Quantitative Results
    • Staff
      • Sample
    • Standardized Measures
      • Social Desirability
      • Job Satisfaction and Stress
      • Safety
  • Offenders
    • Sample
    • Self-Report Data
      • Participation in Pro-Social Activities
    • Standardized Measures
      • Social Desirability
      • Institutional Environment and Safety
      • Interpersonal Problems
• STUDY 2
  • Method
    • Participants
    • Description of Measures
    • Procedure
    • Analyses
  • Results
    • Sample
    • Offender Management System Data
      • Participation in Pro-Social Activities
      • Poor Institutional Adjustment
• DISCUSSION
• Study Limitations
• Summary of Findings
• Recommendations
• Conclusion
• REFERENCES

LIST OF TABLES

• Table 1. Number of Inmate and Staff Participants by Mode of Inquiry and Institution
• Table 2. Staff’s Mean Scores on Standardized Measures
• Table 3. Percentage of Respondents Housed at Each Institution at Pre- and Post-test
• Table 4. Inmates’ Time Spent in Pro-Social Activities
• Table 5. Inmates' Mean Scores on Standardized Measures
• Table 6. Percentage of Women Housed at Each Institution at Pre- and Post-test
• Table 7. Percentage of Offenders Participating in Pro-Social Activities
• Table 8. Percentage of Offenders Displaying Indicators of Poor Institutional Adjustment

LIST OF FIGURES

• Figure 1. Interaction Effect of Time and Unit on Frequency of Staff Use of Force on Inmates
• Figure 2. Interaction Effect of Time and Unit on Self-Reported Involvement in Educational Activities

LIST OF APPENDICES

• Appendix A. Staff and Inmate Interview Protocols
• Appendix B. Focus Group Protocol
• Appendix C. Staff and Inmate Surveys
• Appendix D. Relationships between Social Desirability and Assessment Measures

INTRODUCTION

In recent years, the number of federally sentenced women classified as maximum security has increased, with levels now averaging between 40 and 60 offenders at any given time (Correctional Service Canada, 2003; 2005). Research by Blanchette (1997) has demonstrated that there are statistically reliable differences between women classified as maximum security and those classified as medium and minimum security in several areas; specifically, those classified as maximum security have higher needs and higher risk ratings than do those classified as medium and minimum security. The secure units were designed to assist women assessed as "high" in static risk and dynamic factors through intensive intervention in a secure living environment. This examination attempts to evaluate the opening of the secure units for women offenders at the regional facilities and to examine the impact of their opening on all staff and women directly affected. In order to gain an appreciation of the secure units, this document will begin with a brief history of women's corrections leading up to the development of the Intensive Intervention Strategy for ‘high risk’ women, followed by a summary of research that has informed the development of this strategy.

History of Women's Corrections

Women's corrections began in 1835 when the first females were incarcerated in a women’s unit at an all-male institution, Kingston Penitentiary. Almost a century later, in 1934, the Prison for Women was officially opened in Kingston to address the issue of what to do with women under federal sentence in Canada. For the first time, women offenders, regardless of their security status, had a facility of their own. Between its founding and the late 1980s, however, several reports identified numerous shortfalls of Prison for Women and women's corrections in general. These reports enumerated issues such as the geographic dislocation of women from their families, the overly secure environment for low security women, and the lack of sufficient programs and services.

Finally, in 1989, a Task Force on Federally Sentenced Women was established to address these longstanding concerns; it produced the first report to lead to substantial change in women's corrections, the April 1990 report entitled Creating Choices. The recommendation of this report was that the Prison for Women be closed, and that four new regional facilities for women offenders, along with a healing lodge for Aboriginal women offenders, be created. Furthermore, it was suggested that these facilities operate using a ‘community-living’ model. As such, between 1995 and 1997, five new women's facilities, including a healing lodge, began operations and all federally sentenced women in Canada were transferred to these facilities².

Following a series of incidents in 1996, Correctional Service Canada (CSC) determined that, due to their disruptive behaviour, high escape risk, and risk to the public, a small percentage of women offenders required a greater degree of structure and control than the regional facilities could provide. This prompted CSC to move these women out of the new regional facilities. Instead, they were housed at co-located units within a number of men's institutions (the Regional Psychiatric Centre, Saskatchewan Penitentiary, the Regional Reception Center, and Springhill Institution), separate from the male population. This was an interim measure until the most efficient way to support and manage these women could be implemented.

Research Informing the Intensive Intervention Strategy

A number of studies facilitated the development of the Intensive Intervention Strategy, intended to inform the management of women offenders classified as maximum-security. As Blanchette's (1997) research demonstrated significant differences between women offenders classified as minimum and medium security and those classified as maximum security, numerous studies of CSC's incarcerated women were undertaken. In an attempt to better define the issues facing the women classified as maximum security, three studies were conducted: McDonagh (1999), Morin (1999), and Warner (1998).

McDonagh (1999) conducted a qualitative research study to help identify approaches viewed as useful in addressing the issues and needs of women classified as maximum security and facilitating the reduction of their security classifications. In order to examine the perspectives and experiences of non-Aboriginal federally sentenced women who were classified as maximum security, interviews were conducted with inmates and staff. Four main findings informed the development of the Intensive Intervention Strategy. First, distinct programming strategies, in terms of program content and delivery, were necessary given the cognitive capacities of some of these women (e.g., short attention spans, emotional dysregulation / lability). Second, it was found that, for some women, intensive programming was required to address aggressive behaviour. Third, the presence of a predictable and relatively safe environment was suggested to lessen the risk of psychological deterioration and facilitate healing. Finally, it was suggested that there was value in using a multi-disciplinary team approach to manage this group.

A companion report to McDonagh's work presented the opinions, observations, and suggestions of federally sentenced Aboriginal women classified as maximum-security and CSC staff on the procedures, policies, and programs that were in place to help reduce Aboriginal women's security levels (Morin, 1999). Recommendations to address the needs of these women included the development of a specialized treatment program to address suicidal and self-injurious behaviour, and the need for full-time Elder counselling services.

Warner (1998) examined the experiences of women in federal penitentiaries and demonstrated that women with extensive needs may be poorly served by less secure environments. It was found that some women have entrenched and long-term behaviour patterns that may require individualized and intensive learning programs. More specifically, supportive, consistent, and ‘present’ staffing was found to be most conducive to the establishment of supportive relationships, and multi-disciplinary team approaches with mental health expertise were preferred. Finally, the need for specialized mental health programs was emphasized.

Intensive Intervention Strategy

The results of these three studies informed the development of the Intensive Intervention Strategy. The strategy recognised that some women require intensive intervention, but that their risk assessments require that this intervention be provided in a secure environment. For those women with significant mental health needs, Structured Living Environment houses were constructed at each facility. These accommodate women classified as medium and minimum security who require more intensive support. For those women identified as being high risk, the strategy called for the modification and expansion of the existing units of the regional facilities. These expanded units - secure units - house women classified as maximum security.

On January 16, 2003, Solicitor General Wayne Easter presided over the opening ceremony of the first new secure unit at Nova Institution for Women in Truro, Nova Scotia. The opening of the secure unit at Edmonton Institution for Women in Edmonton, Alberta took place in February 2003, followed by the opening of secure units at Joliette Institution (Joliette, Quebec) in April 2003, and at Grand Valley Institution for Women (Kitchener, Ontario) in October 2004. More recently, a fifth secure unit, at Fraser Valley Institution (Abbotsford, British Columbia), was opened in March 2006.

The opening of these units highlights Correctional Service Canada's commitment to a women-centered approach, which recognizes that women have had unique experiences and have a collective history different from men’s. The role of the secure units, as defined by the Secure Unit Operational Plan (CSC, 2003), is to "ensure the safe and humane custody of women while respecting their rights and entitlements under the law; to promote change in their behaviour and coping skills so that they may safely reintegrate to lower level security; and to allow them to integrate where possible and separate when necessary with the main institution" (p. 7). These roles are met through effective management and operation of the units, as well as through programs and services that are provided to the women. The focus of the Intensive Intervention Strategy is the provision of a safe and secure accommodation for the women while emphasizing individually-based intensive staff intervention, programming, and treatment. This strategy is based on an inter-disciplinary model of intervention whereby all staff and individuals working on the unit comprise a team that stresses the importance of dynamic security as opposed to solely static security, meaning fewer static or electronic barriers and more person-to-person interaction.

Current Study

This report aims to provide information on the immediate impacts of the opening of the secure units (i.e., within six months). More specifically, it will investigate five areas:

Evaluation Strategy

The evaluation strategy was developed through examination of relevant literature and consultation with the Service’s Women Offender Sector and Health Services; it included two distinct components. The first component, Study 1, involved gathering quantitative and qualitative feedback from the women and staff at each of the regional women’s institutions. This provided information on the effects of the opening, as perceived by those who were impacted. The second component, Study 2, consisted of comparisons of offenders’ files from prior to and after the opening; this provided an alternative perspective of the impacts of the new units. The results of the two components are integrated in the discussion concluding this report. In both studies, data at the second time point corresponded to very shortly after the opening of the secure units; this study therefore only reflects the immediate impacts of the opening of the secure units.

STUDY 1

Method

Participants

The participants in this evaluation were all those who may have been impacted by the opening of the secure units. Specifically, this included the women living in the secure units and the staff who work there (SU women and SU staff, respectively), as well as the general population women and staff (GP women and GP staff, respectively). In both the general population and the secure unit, staff included primary workers, behavioural counsellors, program officers, assistant team leaders and team leaders. In addition, where possible, feedback was solicited from additional sources, including wardens, deputy wardens, health care staff, and chaplains.

Participants were recruited from four of the regional institutions for federally sentenced women: Edmonton Institution for Women (EIFW), Grand Valley Institution for Women (GVIW), Joliette Institution (Joliette), and Nova Institution for Women (Nova). Fraser Valley Institution, whose secure unit opened only after data collection was concluded (March 2006), was excluded from the study.⁴ Prior to the secure units’ opening, participants were also recruited from a number of men’s institutions where women classified as maximum-security were co-located: the Regional Psychiatric Centre, Saskatchewan Penitentiary, the Regional Reception Centre, and Springhill Institution.

Description of Measures

This component of the evaluation included both qualitative and quantitative components. The qualitative component involved interviews, focus groups, and open-ended surveys, whereas the quantitative component involved standardized quantitative measures. Of these, the open-ended surveys and standardized quantitative measures were administered on two occasions, the first of which was approximately six months prior to the secure units opening (pre-test), and the second approximately six months to a year after they had opened (post-test). The interviews and focus groups, on the other hand, were administered only once, at post-test (approximately six months to a year after the units’ opening).

Staff and Offender Interviews

Staff and participant interviews (Appendix A) served as an essential source of qualitative data for this evaluation. Semi-structured interviews provided respondents with an opportunity to confidentially express personal views, feelings, and ideas about the secure units. Whenever possible, interviews were conducted with the secure unit women, the primary workers, team leader, and assistant team leader of the secure unit, and the warden and deputy warden of each institution. Any other relevant staff that was interested in participating in a one-on-one interview were also given the opportunity.

Focus Groups

Conducting one-on-one interviews with each of the general population women was not feasible. However, their feedback regarding the secure unit was important, so focus groups were held with the general population inmate committee, or other interested general population inmates, at each facility (Appendix B). Focus Groups involved open discussions facilitated by researchers from National Headquarters with the representatives for the general population women at each institution. These representatives were the voice for the general population women regarding the impact of and feelings toward the secure units.

Staff and Offender Surveys

Open-ended surveys were administered to all interested staff and inmates, both in the secure units and in the general populations (Appendix C). The surveys provided both staff and inmates an opportunity to provide feedback regarding the secure unit, without having to actually participate in a one-on-one interview.

Standardized Quantitative Measures

Staff and inmates were also administered packages containing numerous standardized questionnaires as well as a survey of background information created by the researchers. These questionnaires provided a quantitative measure of impacts of the secure units, thereby complementing the qualitative data.

Staff Package

The Job Satisfaction Scale (Warr, Cook, & Wall, 1979) measures the degree to which a person reports satisfaction with intrinsic and extrinsic features of their job.  The scale consists of 15 items plus one overall summation item, each rated on a 7-point scale ranging from extremely dissatisfied to extremely satisfied. The scale offers a total score, as well as six subscale scores derived from various distributions of the items: Intrinsic Job Satisfaction, Extrinsic Job Satisfaction, Job Itself Intrinsic Satisfaction, Working Conditions Extrinsic Satisfaction, Employee Relations Satisfaction, and a single-item subscale measuring Overall Job Satisfaction.

The Maslach Burnout Inventory (Maslach, Jackson, & Leiter, 1996) was designed to examine how various people in the human services or helping professions view their jobs and the people with whom they work closely. The scale consists of 22 statements of job-related feelings. Each item is scored on a 7-point scale ranging from never to everyday that describes how often respondents feel a particular way about their job. The measure includes three subscales: Emotional Exhaustion, Depersonalization, and Personal Accomplishment.

The Occupational Safety and Health Survey was originally a subscale in the 1996 Correctional Service Canada Staff Survey (Robinson, Lefaive, & Muirhead, 1997). This scale assesses respondents’ perceptions of the ability of CSC policies and procedures to protect staff from harm from inmates, and is composed of four questions rated on a 7-point scale ranging from strongly disagree to strongly agree.

The Personal Safety and Security Scale: Staff Version was adapted from the Prison Social Climate Survey, Staff Version (U.S. Department of Justice, 1993). The scale measures respondents' impressions of the overall safety of the living and working conditions in their institution over the preceding six months. Due to the complex nature of the measure, only nine of its 22 items were included in the study. Four items enquire as to the number of times specific incidents have occurred among inmates (heated arguments, assaults without weapons, assaults with weapons, and sexual assaults). These items assess the frequency and likelihood of specific incidents occurring, including arguments and assaults. Various scoring strategies are used.

The Stress and Coping Among Correctional Officers (Triplett, Mullings, & Scarborough, 1999) measures the degree of stress associated with a variety of sources. The measure does not provide a meaningful overall score; instead, subscales are used to measure various sources of stress. The subscales are: Role Ambiguity, Role Overload, Role Conflict, Career Development, Overwork, and Safety. The measure’s 27 items are scored on a 7-point scale ranging from very strongly disagree to very strongly agree.

Supervisory Support and Stress Scales (Cullen, Link, Wolfe, & Frank, 1985). Two of this tool’s subscales measure workplace support (Supervisory Support and Peer Support), while one measures levels of stress (Work Stress). Workplace support is defined as resources that enable individuals to either resist or overcome potential stressors in their work environment, while levels of work stress are defined by the degree of anxiety or pressure officers feel while on duty. The scale includes 17 questions rated on a 7-point scale ranging from very strongly agree to very strongly disagree.

Finally, the Paulhus Deception Scale (Paulhus, 1998) was also included. This scale assesses the extent to which respondents’ answers reflect an exaggerated or dishonest portrayal of themselves, and is used in interpreting the results of responses to other scales. It includes two subscales measuring the major forms of socially desirable responding: Self-Deceptive Enhancement and Impression Management. Self-Deceptive Enhancement refers to the tendency of respondents to give honest but inflated self-descriptions while Impression Management refers to a tendency to intentionally present oneself favourably in order to impress others. The measure’s 40 items are scored on a 5-point scale ranging from not true to very true.

Offender Package

The Correctional Environment Status Inventory (Wolfus & Stasiak, 1996) measures respondents’ perceptions of the quality of the correctional environment of the institution in which they are incarcerated. It consists of the following six scales: Staff Cohesion, Staff Involvement, Staff Treatment Focus, Clarity and Organization, Offender Relationships, and Offender Treatment Orientation. The inventory’s 66 items are rated on a 5-point scale ranging from completely disagree to completely agree. Though the inventory was developed based on the responses of male offenders, it has also been used with samples of women offenders (Delveaux & Blanchette, 2000; Syed & Blanchette, 2000).

The Inventory of Interpersonal Problems (Horowitz, Rosenberg, Baer, Ureno, & Villasenor, 1988) examines the degree of difficulty and distress associated with various interpersonal domains, including being too controlling or manipulative, being self-centered and resentful, and being too intrusive and having difficulty respecting the boundaries of others. The scale includes 64 items assessing the distress respondents experience in a variety of situations, rated on a 5-point scale ranging from not at all to extremely. These items are combined to produce a total score and eight subscale scores: (1) Domineering / Controlling, (2) Vindictive / Self-Centred, (3) Cold / Distant, (4) Socially Inhibited, (5) Non-Assertive, (6) Overly Accommodating, (7) Self-Sacrificing, and (8) Intrusive / Needy.

The Personal Safety and Security Scale: Inmate Version was adapted for the present study from the Prison Social Climate Survey, Inmate Version (U.S Department of Justice, 1990). This scale measures respondents’ impressions of the overall safety of the housing and working areas in their facility over the preceding six months. The items used in this study are the same as those used in the staff version.

The Self-Control Schedule (Rosenbaum, 1980) is a self-report instrument that assesses individual differences in learned resourcefulness, which refers to the breadth and depth of an individual's self-control repertoire. In the cognitive-behavioural context, self-control refers to an individual's ability to self-regulate internal responses such as emotions, cognitions, or pain, which may otherwise interfere with goal-directed behaviours. The scale includes only a total score, which consists of 36 items scored on a 6-point scale ranging from very uncharacteristic of me to very characteristic of me. Items tap several aspects of learned resourcefulness, including control of emotional and physiological responses, problem solving, delay of gratification, and regulatory self-efficacy. The scale has been used with samples of women offenders (Sly, Taylor, & Blanchette, 2003).

The Paulhus Deception Scale was also included in the offender package. As described earlier in the Staff section, this scale is used to measure social desirability and to assist in the interpretation of the results of other scale responses. The scale has previously been used with samples of women offenders (Irving, Taylor, & Blanchette, 2001).

Procedure

During the summer and fall of 2002, prior to the completion and opening of the secure units at each site, researchers travelled to each of four regional women's facilities,⁵ as well as to those men’s facilities in which the women classified as maximum security were co-located, to complete the pre-test phase of data collection. This involved administering the quantitative packages and open-ended survey to all interested women and staff. Participation was voluntary and all participants completing quantitative test batteries signed informed consent forms. Confidentiality was ensured, as the respondents were asked not to identify themselves within the survey documentation.

Between summer of 2003 and fall of 2005, researchers returned to each of the regional women’s institutions to collect post-test data.⁶ At this time the quantitative questionnaire packages and the open-ended surveys were again administered to interested staff and inmates. Focus groups were completed with representatives of the general population women and interviews were conducted with the secure unit women, the primary workers, team leader, and assistant team leader of the secure unit, warden and deputy warden of the institution, and other interested staff. Again, participation was voluntary and confidential, and all interviewees, focus group participants, and quantitative battery respondents signed informed consent forms.

Analyses

Qualitative Data

Content analysis was used to interpret qualitative data. Analyses were conducted separately by time period, mode of inquiry, institution, and for inmates and staff. A step-wise integration (across institutions, across mode of inquiry, across time periods, and for inmates and staff) was used in order to highlight differences and similarities. Though the total number of participants who provided feedback is provided, counts and proportions associated with each theme are not reported due to the impossibility of ascertaining whether any participants provided more than one type of qualitative data.

Quantitative Data

Staff

Since 32 general population and secure unit staff respondents participated in both the pre-test and post-test phases, paired sample and repeated measures analytical techniques were used. These techniques allow for the statistical elimination of variance in scores associated with respondent characteristics and irrelevant to the questions of interest, and therefore result in more precise results. A mixed linear model was used to conduct 2 (time: pre-test, post-test) x 2 (unit: general population, secure unit) analyses of variance.⁷

Inmates

Among the inmates, there were too few respondents who had completed both testing sessions to allow for repeated measures analyses; instead, the total sample of 132 offenders who completed either testing session was used. Of these 132 offenders, 65 participated at pre-test and 67 did so at post-test. All analyses were conducted assuming independence of the pre- and post-test samples.⁸ Data were analyzed using 2 within (time: pre-test, post-test) x 2 between (unit: general population, secure unit) analyses of variance.

Reliability of Measures

The internal consistency coefficients of the standardized measures were calculated at pre- and post-test for both staff and inmates. Any scales for which the conventional threshold value of .60 was not met were eliminated from analyses. As such, for staff, three subscales (the Working Conditions Extrinsic Satisfaction subscale of the Job Satisfaction Scale and the Overwork and Role Overload subscales of the Stress and Coping Among Correctional Officers) were eliminated; for inmates, all standardized scales were retained.

Qualitative Results

This section summarizes the information acquired through interviews, focus groups, and open-ended surveys. The feedback is organized into six areas which aided in the interpretation of the data: Operations, Staffing, Programs, Security, Communication, and Supplementary Feedback. Throughout, a selection of participants’ comments is included in order to illustrate the findings.

Sample

Table 1 presents the number of participants who provided feedback via open-ended surveys and interviews at pre-test and at post-test.

Table 1. Number of Inmate and Staff Participants by Mode of Inquiry and Institution

	Institution
Mode of inquiry	EIFW	GVIW	Joliette	Novaa	Co-loc. Unitsb	Total
Pre-test
Inmates
Open-ended surveys	6	2	13	-	12	33
Staff
Open-ended surveys	39	24	12	-	-	75
Post-test
Inmates
Open-ended surveys	16	10	12	12	-	50
Interviews	7	3	7	4	-	21
Staff
Open-ended surveys	42	12	8	14	-	76
Interviews	4	17	5	6	-	32

Focus groups were also conducted at each institution at post-test. Approximately six to ten women from the general population participated in each focus group. At Nova, the secure unit inmates had their own Inmate Committee distinct from that of the rest of the institution; consequently, two focus groups were conducted at this institution.

Feedback

Operations

The most central theme anticipated by all staff and GP women at pre-test involved concerns that the secure units would negatively impact institutional operations. For instance, staff and women anticipated obstacles such as restricted movement, less visiting time, and less recreation time for the GP women. Furthermore, staff mentioned that additional security requirements, more stringent rules and regulations, loss of privileges, and sharing facilities would be problematic. Post-test results bore out these concerns, as GP inmates did, in fact, report having reduced access to leisure activities and visitation, as well as restricted movement both when an inmate classified as maximum security was being moved and when an institution-wide lockdown was necessitated by an event in the secure unit. Moreover, the environment in the general population was perceived to have become more security-oriented as compared to before the opening. Respondents indicated that this has led to increased tension and stress for GP inmates.

At pre-test, staff also anticipated advantages related to transfers, stating that having the secure unit on site would facilitate the transfer of women classified as minimum or medium security to a maximum security environment if the situation warranted. The GP inmates did not foresee any advantages related to transfers, stating that the secure unit would provide an "easy way out" for staff and result in more GP women being transferred to maximum security. Post-test feedback, however, indicated that women were not reclassified to maximum security at a higher rate than prior to the openings. One institution was an exception; there, both GP and SU women indicated that reclassification to maximum security had taken place for inappropriate reasons, namely due to unconfirmed suspicion of certain behaviours and discrimination based on race and sexual orientation. Conversely, inmates at this institution indicated that those women who did exhibit violent behaviours were not reclassified to maximum security.

A concern mentioned only at post-test was the practice of assigning GP inmates to SU primary workers’ caseloads. As the requirements of the secure unit have kept its staff there most of the time, there has been little time available for contact between GP inmates and their SU primary workers; the affected inmates reported feeling that their needs have not been met.

Respondents’ comments in this area included the following:

Staff and Staffing Issues

At pre-test, staff expressed concern about increased levels of stress, burnout, and the possibility of high staff turnover as a result of the openings. They also anticipated an increase in responsibilities, resulting in work overload. Further, GP staff expressed apprehension regarding acting as backup to the SU staff without supplementary training. Post-test responses bore out these concerns. Both staff and inmates reported that stress, frustration, morale, and burnout have been problems for the SU staff, as well as for the GP staff called upon to work in the secure unit. GP staff indicated that their stress and anxiety are related to the fact that they are being asked to work in the secure unit and they do not feel that their training is sufficient for them to do so. SU staff, in turn, suggested that the dangers associated with their work, the emotional impact of violent incidents, and the burden of working with colleagues who are visibly and vocally unhappy about being assigned to the unit all contribute to their levels of tension.

The inmates also indicated that stress and related feelings have been problematic for staff. Some GP inmates indicated that GP staff who are called upon to work in the secure unit, when exposed to that environment, become frustrated and uneasy. GP inmates reported that when such staff return to GP, they can be impatient and behave in ways which are inconsistent with their previous behaviour.

Another prominent theme, at both pre- and post-test, was concern with regards to understaffing, inadequate training, inexperience, and lack of preparedness. At post-test, staff indicated that insufficient staffing in the secure units has led to overwork, inconsistent direction, reliance upon GP staff to supplement the SU rosters, and, in some cases, excessive or obligatory overtime. Staff recognized that low staffing levels also impacted inmates, as when insufficient staff were available to supervise SU inmates in programs and events with the GP inmates. In addition, GP inmates indicated that low staffing levels result in fewer escorted temporary absences (often because the GP staff member assigned to act as escort was called upon to work in the secure unit), cancelled programs (when program staff were routed to the secure unit), and delays in the completion of paperwork, including correctional plans and paperwork required by the National Parole Board. Finally, SU inmates indicated that the current staffing practices, which rely on GP staff to fill the SU rosters, have led to the dissemination of inconsistent information, as well as inconsistent enforcement of rules and regulations by staff, likely because some GP staff are less familiar with secure unit policies and procedures.

On a more positive note, some staff indicated at pre-test that they looked forward to the opportunity to learn new skills, and to interact with new co-workers. These themes arose at post-test also, though infrequently. Specifically, some staff indicated that they enjoyed the challenge of working with a more demanding inmate population, as well as the opportunity to use a wider repertoire of skills. Many staff also indicated that SU staff have become a very cohesive team. Regrettably, staff also mentioned divisiveness between SU and GP staff, confirming a pre-test concern.

Relevant comments included:

Programs

All pre-test respondents anticipated that the secure units would provide an increase in appropriate programming and services available to women classified as maximum security; both staff and women viewed this outcome as an improvement over the programming offered to women in the co-located units. Post-test respondents, however, indicated that there were few beneficial programs or activities available. The one exception was Dialectical Behaviour Therapy, whose importance and utility were frequently mentioned.

For post-test respondents, there seem to be two components to this issue. First, respondents indicated that there have been too few programs and other opportunities for SU women. Both inmates and staff indicated that there were few programs, educational or employment opportunities, and other activities available to the inmates, and that they are consequently frustrated, bored, and in some cases, unmotivated.⁹ Of particular concern has been SU women’s access to spiritual and religious care. Secondly, the activities which were available were not always considered ideal. Specifically, the employment opportunities, mainly cleaning, have not provided women with useful or transferable skills. Similarly, the programs and activities available were often not correctional programs. For example, when asked to list available programs, SU inmates more often mentioned such activities as hobby craft or horticulture than they did correctional programs.

Respondents’ comments in this area included the following:

Security

At pre-test, the majority of the GP and SU staff voiced concerns regarding security issues related to being unprepared, lack of planning, and limited resources. They stated that the institutions did not have the right mindset or security to handle the SU women. SU staff argued that too few precautions had been put in place regarding the physical structure of the units and that there were "a number of deficiencies in the building itself".

These structural concerns were also mentioned at post-test. Both SU and GP staff indicated that components of the physical construction of the secure units posed security risks, including "blind spots" in the way the units are constructed, the lack of food slots in cell doors, insufficient security cameras, and doors which offer free egress. There was frustration among staff that their comments regarding these safety issues had not been attended to; in some cases, this has led to resentment of management, who were perceived as responsible for this lack of follow up.

Data suggested fewer complaints regarding procedural shortcomings. A number of GP and SU staff members, however, mentioned the necessity of providing refresher training in areas such as use of force and arrest and control in order to ensure staff and inmate safety. Moreover, some concern was expressed by GP and SU inmates and staff regarding the movement level system¹⁰ and the management protocol policies¹¹ used to further classify women housed in the secure unit. Some staff members indicated that the progression to level four is too rapid, and that the reduction in security precautions associated with this decrease in movement levels is therefore also too rapid. Some staff and inmates (particularly at one institution) also suggested that work is necessary to improve the policies relating to management protocol inmates in order to ensure the safety of both staff and inmates.

Among post-test respondents, another finding was that the prevalence of incidents in the secure unit was higher than had been anticipated. Such incidents have, to date, included physical and verbal assaults (both inmate-staff and inmate-inmate), self-injury, a suicide by hanging, and a hostage-taking. It should be noted that staff at two institutions stated that there have been insufficient / untimely debriefings associated with these incidents, as well as insufficient access to Critical Incident Stress Management aid.

Some staff members felt that these incidents were partially attributable to inadequate access to specific tools, such as restraints and pepper spray. Others mentioned that they are discouraged by management from acting on their own initiative when accessing and utilizing tools such as pepper spray and handcuffs, particularly as it applies to transportation of an inmate.

Comments relevant to security included:

Communication

Problems with communication were identified at both pre- and post-test. At pre-test, staff expressed a clear desire for improved lines of communication and increased support for staff decisions and knowledge. They emphasized the importance of addressing concerns raised by staff, arguing that the effective management of SU women would depend on such actions. At post-test, however, this issue continued to be of considerable concern, especially at EIFW and GVIW. A number of staff members related that their input and concerns were disregarded by management, both within the secure unit and at an institutional level.

Communication between inmates and staff was also a concern at both testing periods. The SU women who provided feedback at pre-test felt that they had not been adequately advised and informed of what the secure unit openings would entail or what to expect. Some of the women at the co-located units expressed confusion regarding the opening, citing beliefs that once transferred they would be locked up for 23 hours each day and be isolated for extended periods of time. At post-test, these concerns had shifted somewhat; respondents were primarily concerned with insufficient communication around rules and regulations. Specifically, SU women indicated that staff enforcement of rules and policies was sometimes inconsistent (perhaps due to insufficient training of GP staff, as previously discussed). Moreover, a number of staff members indicated that expectations were not consistently communicated to SU inmates, as concessions and exceptions were often made in an effort to placate inmates. They indicated that this has resulted in the SU inmates receiving reinforcement of negative behaviour, and consequently learning that these behaviours are appropriate.

Relevant comments include:

Supplementary Feedback

Overall, the inmates’ pre-test expectations regarding the secure units were negative. The GP women anticipated that the presence of the SU women would create nervousness and emotional strain, ultimately having a bad influence on the GP women. They expected that the environment created by the secure unit would result in a lack of respect and an increase in violence among GP women. The SU women also saw the opening of the secure units as being a negative change because of the necessity of adjusting to a new environment. Some women explained that they were adjusted to their current setting and would prefer to maintain the status quo.

At post-test, respondents’ impressions of the units had not improved much. In fact, a number of GP and SU inmates and staff mentioned that the secure units should be fully self-contained or be grouped together in a distinct facility. It was suggested that this separation would reduce or eliminate the impacts of the secure unit on the GP staff and inmates (e.g., being asked to work in the secure unit, reduced access to leisure and visitation, restricted movement). Others suggested that rather than separating the entire secure unit from the main institution, only management protocol inmates should be located elsewhere. The possibility of a Secure Handling Unit for women offenders was also suggested.¹²

Post-test respondents’ feedback, however, also revealed an unexpected positive finding: GP inmates expressed a desire to have the SU women more strongly integrated into the institutional community, and to interact with them where possible. Moreover, GP inmates at one institution were especially concerned about the wellbeing of the SU inmates. They felt that in addition to being isolated, the SU women may be subject to overly restrictive or invasive measures - most notably, confinement using a restraint chair which they were shown on their tour of the unit.

One area where no clear conclusion emerged was with regards to SU women having more motivation to decrease their security classifications as a result of their proximity to GP. Though both GP and SU staff mentioned this as a possibility at pre-test, the majority of post-test respondents were unsure whether such an effect had occurred. A number of SU inmates indicated that having more opportunities to participate in activities with the GP women might increase their motivation in this regard.

Respondents’ comments in this domain included:

Quantitative Results

This section summarizes information garnered from the battery of quantitative instruments. The results are presented separately for staff and women. In each case, a summary of the sample is presented first, followed by the results of analyses of the standardized measures.

Staff

Sample

As previously mentioned, staff participants represented a matched pre-post sample. In other words, all staff participants included in the study completed the quantitative measures both at pre-test and at post-test. The results therefore address changes over time for the same participants.

Institutional Information

Of the 32 respondents, the greatest proportion worked at EIFW (50%); the remainder worked at GVIW (9%), Joliette (16%), and Nova (25%). On average, at pre-test, respondents had been working at their institution for about two years (M = 2.24, SD = 2.10), with a range of 3 months to 7.1 years. The average length of time worked did not differ by institution.

Most respondents were working as primary workers (65% and 55% at pre- and post-test, respectively), though a considerable proportion were team leaders or assistant team leaders (10% and 16%). Programs staff (6% and 10%), health care staff (6% and 3%) and staff from other categories (13% and 16%), including parole officers, community reintegration officers, and case management officers, also participated. Changes in proportions from pre- to post-test represent staff members who changed jobs.

At pre-test, nearly all respondents (91%) worked in GP; at post-test, this proportion was reduced by a third (59%). Unsurprisingly, staff members who participated at pre-test had worked in their unit for a shorter length of time than had those who participated at post-test. Length of time worked also differed significantly by unit, with GP respondents working in their unit for an average of three years (M = 3.09, SD = 2.02), as compared to SU respondents who had been working in their unit for less than a year (M = 0.79, SD = 0.43). Again, this is not surprising given that the secure units had been open for less than a year at post-test.

Demographic Information

Of the 32 staff respondents, 78% were women and 22% were men. This sample was composed primarily of Caucasian respondents (87%). The rest of the sample was divided amongst Aboriginal (10%) and Black (3%) staff; no one identified as being from another visible minority.

Respondents’ age ranged from 23.6 to 66.2 years, with an average of 38.7 years (SD = 11.12). On average, the respondents had completed a significant amount of post-secondary education. The majority of the sample were college or university graduates (24% and 58% respectively), while the remainder had either graduated from high school (6%) or completed a portion of a college diploma (6%) or university degree (6%).

Analyses revealed that there were no differences by testing session or unit in terms of gender, ethnicity, or education. Though age was also found not to differ from pre- to post-test or by unit, a test comparing the age of those who did and who did not change units from pre- to post-test approached significance, t(22) = 2.00, p < .10; participants who changed units were an average of 8.80 years younger than their counterparts who did not.

Standardized Measures

Social Desirability

Before assessing the relationships of each of time and unit with the battery of standardized measures, correlations were computed between scores on these measures and scores on the Paulhus Deception Scale, a measure of social desirability.¹³ These correlations are presented in Appendix D. At pre-test, scores on the self-deceptive enhancement scale were not correlated to scores on any of the standardized measures. Scores on the impression management subscale, on the other hand, were positively correlated with the Employee Relations Satisfaction subscale of the Job Satisfaction Survey (r = .32) and negatively correlated with three subscales of the Stress and Coping Among Correctional Officers scale (r = -.30 to -.32). The directions of these correlations demonstrate that staff had a tendency to positively exaggerate their satisfaction with employee relations, as well as to downplay their levels of stress relating to role ambiguity, role conflict, and career development.

This inclination towards portraying a positive impression was considerably more pronounced at post-test. After the opening of the secure units, scores on nearly half of the standardized measures’ subscales (7 of 17) were significantly related to self-deceptive enhancement, and more than half of the subscales (10 of 17) were correlated with impression management. First, self-deceptive enhancement was positively associated with both the Overall Job Satisfaction subscale of the Job Satisfaction Survey (r = .33) and the Personal Accomplishment subscale of theMaslach Burnout Inventory (r = .38). These correlations demonstrate that staff try to see their job satisfaction and personal accomplishments more positively than is warranted. Conversely, self-deceptive enhancement was negatively correlated with two subscales of the Maslach Burnout Inventory (r = .-40 to -.46), two subscales of the Stress and Coping Among Correctional Officers scale (r = .-44 to -.46), and the Work Stress subscale of the Supervisory Support and Stress Scales (r = .-31). These correlations suggest that staff attempt to view their burnout due to emotional exhaustion and depersonalization, their stress related to role ambiguity and career development, and their work stress as less pronounced than warranted.

Turning to impression management, positive correlations were seen with three subscales of the Job Satisfaction Survey (r = .32 to .35), the Personal Accomplishment subscale of theMaslach Burnout Inventory (r = .35), and the Supervisory Support subscale of the Supervisory Support and Stress Scales (r = .35). Negative associations were found with the remaining two subscales of the Maslach Burnout Inventory (r = -.47 to -.63) and three subscales of the Stress and Coping Among Correctional Officers scale (r = -.41 to -.52). These findings demonstrate that staff tend to exaggerate their satisfaction, sense of being supported, and sense of accomplishment while minimizing their stress.

In general, social desirability was associated with a tendency to respond positively, both by exaggerating the positive and by minimizing the negative. These findings indicate that a certain amount of caution must be exercised in interpreting the relationships of unit and time with the battery of standardized measures, especially in relation to analyses including post-test results. Given the pattern of findings, this care should be particularly marked when results demonstrate positive impacts. Given that negative impacts would run counter to the pattern of exaggeration demonstrated in these analyses, they may be interpreted less cautiously.

The means and standard deviations of scores on the standardized measures are presented in Table 2. These were calculated both by time and by unit.

Table 2. Staff’s Mean Scores on Standardized Measures

Measure	Mean (SD) by Time		Mean (SD) by Unit
	Pre-test (n = 6 to 32)	Post-test (n = 6 to 32)	GP (n = 10 to 47)	SU (n = 4 to 16)
Job Satisfaction Survey
Overall Job Satisfaction	5.26 (1.00)	4.66 (1.43)	5.19 (1.15)	4.25 (1.34)
Intrinsic Job Satisfaction	32.77 (6.79)	29.88 (7.14)	32.54 (6.90)	27.63 (6.43)
Extrinsic Job Satisfaction	36.73 (5.42)	34.68 (6.92)	36.29 (6.28)	34.00 (6.08)
Job Itself Intrinsic Satisfaction	20.48 (4.37)	18.50 (4.43)	20.43 (4.21)	16.69 (4.17)
Employee Relations Satisfaction	24.21 (6.12)	21.81 (6.58)	23.86 (6.67)	20.50 (5.10)
Total Score a+	74.90 (12.01)	68.90 (13.85)	73.95 (12.85)	65.88 (12.82)
Maslach Burnout Inventory
Emotional Exhaustion a	21.34 (9.16)	25.59 (11.82)	23.11 (10.84)	24.88 (10.83)
Depersonalization b+	8.45 (4.86)	9.41 (6.84)	7.98 (5.67)	11.69 (6.04)
Personal Accomplishment	33.23 (8.32)	32.72 (7.08)	33.30 (7.38)	32.00 (8.56)
Occupational Safety and Health Survey
Total Score a+ b+	18.39 (5.57)	14.47 (6.54)	17.60 (6.11)	13.25 (5.98)
Personal Safety and Security Scale
Number heated arguments	29.93 (37.84)	17.59 (31.38)	24.46 (35.94)	21.00 (32.87)
Number assaults without weapons	2.32 (1.52)	3.81 (2.90)	2.68 (1.93)	4.21 (3.29)
Number assaults with weapons a+	1.00 (0.63)	1.67 (1.22)	1.30 (1.06)	1.60 (1.14)
Number sexual assaults	1.56 (0.73)	0.71 (0.49)	3.33 (1.71)	2.50 (2.12)
Frequency inmates have weapons	3.25 (1.71)	3.72 (1.46)	3.33 (1.71)	2.50 (2.12)
Frequency staff use force on inmates c	1.80 (0.79)	2.90 (1.07)	2.44 (1.08)	3.29 (1.07)
Frequency inmates use force on staff a+	1.90 (0.72)	3.25 (1.04)	2.40 (1.10)	2.80 (1.14)
Likelihood of inmate being assaulted	1.84 (0.78)	2.23 (0.79)	1.90 (0.76)	2.44 (0.81)
Likelihood of staff being assaulted	1.47 (0.84)	1.91 (0.78)	1.56 (0.85)	2.06 (0.68)
Stress and Coping Among Correctional Officers
Role Ambiguity	12.61 (5.09)	13.31 (4.91)	12.36 (4.93)	14.75 (4.81)
Role Conflict	17.19 (5.88)	18.88 (6.10)	17.98 (6.10)	18.25 (5.92)
Career Development	18.27 (8.47)	20.83 (7.90)	18.11 (8.47)	23.50 (6.15)
Safety b	21.80 (5.97)	23.97 (7.05)	21.11 (6.36)	28.13 (4.01)
Supervisory Support and Stress Scales
Supervisory Support	27.94 (5.20)	21.48 (7.14)	26.63 (6.19)	25.00 (7.16)
Peer Support	22.61 (5.65)	22.23 (5.81)	22.15 (5.93)	23.19 (5.04)
Work Stress	23.68 (5.91)	26.42 (7.19)	24.87 (6.41)	25.56 (7.59)

Job Satisfaction and Stress

The pre- to post-test difference on the Job Satisfaction Scale approached significance, demonstrating that respondents were less satisfied with their jobs at post-test than at pre-test. Examination of responses on the Supervisory Support and Stress Scales, however, indicated that there were no differences in the staff respondents’ perceptions of supervisory support, peer support, or work stress from pre- to post-test. However, respondents’ scores on the Emotional Exhaustion subscale of the Maslach Burnout Inventory were significantly higher at post-test than at pre-test, indicating that after the secure units opened, staff experienced more emotional exhaustion-related burnout than previously. Notably, this finding persisted in spite of staff’s tendency to minimize their emotional exhaustion-related burnout after the unit’s opening.

In comparing GP and SU staff, differences were also found. Though there were no differences on the Job Satisfaction Scale or the Supervisory Support and Stress Scales, differences approaching significance were noted on the Depersonalization subscale of the Maslach Burnout Inventory. Specifically, the difference on this subscale suggested that SU staff experienced more burnout related to depersonalization, or sense of detachment from one’s body and experiences, than did GP staff.

Safety

There were differences in perceived level of safety found both across time (pre-post) and by site (GP versus SU). First, scores on the Occupational Safety and Health Survey decreased over time, suggesting that after the secure units opened, staff had lower confidence in CSC’s policies and procedures vis-à-vis their ability to prevent staff from being harmed by inmates. According to the Personal Safety and Security Scale, the perceived frequency with which assaults with weapons and incidents of inmate use of force on staff occur also both approached significance. Moreover, scores on the Safety subscale of the Stress and Coping among Correctional Officers Scale were significantly higher for SU respondents than for GP respondents, indicating the SU staff experienced more stress associated with their perceived safety.

The difference in scores by unit on the Occupational Safety and Health Survey also approached significance, suggesting that relative to GP respondents, SU respondents were less certain of CSC’s ability to prevent harm from inmates. However, no differences were detected by unit on the Stress and Coping among Correctional Officers Scale.

Finally, an interaction effect of time and unit was detected with regards to the frequency with which staff use force with inmates. As can be seen in Figure 1, though both GP and SU staff perceived such use of force to have increased in frequency from pre- to post-test, the perceived increase was greater for SU respondents.

Figure 1. Interaction Effect of Time and Unit on Frequency of Staff Use of Force on Inmates

Offenders

Sample

Institutional Information

The proportions of the total sample of 132 women housed at each institution at pre-test and post-test is shown in Table 3. At each phase of testing, the largest proportion of inmate respondents was living at Joliette. In the post-test and total samples, the proportions of respondents from EIFW, GVIW, and Nova were approximately equal; at pre-test, however, EIFW and GVIW were underrepresented.

Table 3. Percentage of Respondents Housed at Each Institution at Pre- and Post-test

Institution	Pre-test	Post-test	Total
Edmonton Institution for Women	10%	26%	18%
Grand Valley Institution for Women	9%	22%	15%
Joliette Institution	30%	32%	31%
Nova Institution for Women	20%	20%	11%
Regional Psychiatric Centre	3%	-	2%
Saskatchewan Penitentiary	15%	-	7%
Regional Reception Centre	9%	-	5%
Springhill Institution	4%	-	2%

Note. The Regional Psychiatric Centre, Saskatchewan Penitentiary, the Regional Reception Centre, and Springhill Institution are included only at pre-test as these are the men’s institutions where women offenders were co-located prior to the secure units’ opening.

On average, inmates had been housed at their institution for about a year and a half (M = 1.70, SD = 2.14), with a range of 1 month to 10.2 years. Of the women, 89% were housed in the general population. The remaining respondents were from the maximum security and secure units. On average, respondents had been housed in their units for over three quarters of a year (M = 0.81, SD = 1.02), with a range of 1 month to 5 years. These time periods did not differ by testing period, institution, or unit.

Demographic Information

Of those participants who provided information with regards to ethnicity, 68% were Caucasian, 20% were Aboriginal, 8% were Black, and 4% were members of another visible minority. The women’s average age was 36.9 years (SD = 10.5), with a range of 19.9 to 64.0 years. Half of the women had not completed high school (50%), while about a quarter had obtained a high school diploma (24%). Smaller proportions of the respondents had begun post-secondary education (15%) or had received a post-secondary diploma or degree (11%). Finally, over half (56%) of the women had previously been incarcerated. There were no differences in these proportions and averages according to testing session or unit.

Self-Report Data

Participation in Pro-Social Activities

Participation in pro-social activities was assessed by self-report (as part of the background survey included in the standardized battery). Average hourly participation in these activities is presented in Table 4 both by time and unit.

Table 4. Inmates’ Time Spent in Pro-Social Activities

	Mean (SD) by Time		Mean (SD) by Unit
Variable	Pre-test (n = 27 to 50)	Post-test (n = 37 to 48)	GP (n = 58 to 87)	SU (n = 6 to 11)
Time spent in work (hrs/wk) a, b	29.48 (23.06)	20.50 (19.53)	26.26 (22.71)	12.55 (10.48)
Time spent in educational activities (hrs/wk) c	13.06 (15.33)	17.60 (43.15)	17.49 (36.43)	4.79 (5.80)
Time spent in programs (hrs/wk)	15.97 (12.10)	14.29 (16.11)	15.70 (14.99)	13.57 (11.07)
Time spent in spiritual or religious activities (hrs/wk)	8.89 (10.58)	12.29 (31.48)	11.11 (6.25)	8.50 (9.30)
Time spent in activities with staff (hrs/wk)	2.37 (3.95)	3.24 (7.24)	2.89 (6.24)	4.33 (6.28)
Time spent in recreational activities (hrs/wk)	19.04 (12.80)	18.96 (25.27)	19.50 (21.05)	20.64 (14.60)

Work and Education

Differences were detected in the self-reported amount of time spent weekly in both work-related activities and educational activities. Women’s involvement in work and vocational activities was lower at post-test than it was at pre-test. Additionally, SU women spent about half as much time engaging in work-related activities compared to their GP counterparts.

With regards to self-reported time spent in educational activities, a significant interaction effect of time and unit was detected; the main effects were therefore not considered. As can be seen in Figure 2, while the time spent in such activities increased from pre- to post-test for GP women, it decreased slightly for SU women. This indicates that though GP women participated in a greater amount of educational activities at each testing period, the gap between the GP and SU women increased from pre- to post-test.

Figure 2. Interaction Effect of Time and Unit on Self-Reported Involvement in Educational Activities

Programs and Activities

No differences were found by time or unit on weekly amount of time spent in programs (e.g., substance abuse, cognitive and living skills), time spent in spiritual or religious activities, time spent in activities with staff, or on time spent in recreational activities. These results indicate that both pre- and post-test respondents, and GP and SU respondents, engaged in programs and other pro-social activities for similar amounts of time each week.

Standardized Measures

Social Desirability

Again, associations between the standardized measures and the Paulhus Deception Scale were calculated in order to ascertain the extent to which social desirability played a role in women’s responding. Fewer such associations were found for the women than had been found for staff. At pre-test, three subscales, as well as the total score, of the Inventory of Interpersonal Problems were found to be negatively associated with self-deceptive enhancement (r = -.31 to -.39), demonstrating that women see themselves as having fewer interpersonal problems than is really the case. At this time point, none of the measures were associated with impression management.

At post-test, two of the same subscales of the Inventory of Interpersonal Problems continued to be negatively correlated with self-deceptive enhancement (r = -.30), though these correlations were of lesser magnitude than at pre-test. After the units’ opening, the total score of the Self-Control Schedule was positively associated with impression management (r = .39), demonstrating that women had a tendency to exaggerate their ability to self-regulate their emotions and cognitions.

Notably, none of the subscales of the Correctional Environment Status Inventory were associated with social desirability at either testing period, demonstrating that data stemming from this measure can be considered particularly candid. The data relating to several subscales of the Inventory of Interpersonal Problems, as well as to the Self-Control Schedule, must be interpreted with more caution, though the number and magnitude of significant correlations is not great.

Means and standard deviations for the standardized measures for all inmate respondents are presented in Table 5 both by time and by unit.

Table 5. Inmates' Average Scores on Standardized Measures

Measure	Mean (SD) by Time		Mean (SD) by Unit
	Pre-test (n = 35 to 61)	Post-test (n = 32 to 58)	GP (n = 57 to 100)	SU (n = 8 to 12)
Correctional Environment Status Inventory
Staff Cohesion a	24.00 (4.54)	27.22 (5.30)	25.33 (5.20)	28.50 (5.26)
Staff Involvement	40.51 (13.99)	42.95 (15.91)	42.78 (15.51)	35.00 (9.10)
Staff Treatment Focus	35.98 (11.94)	38.59 (13.77)	37.86 (12.90)	32.10 (10.86)
Clarity and Organization	26.73 (5.46)	27.23 (6.62)	26.98 (6.38)	25.50 (6.33)
Offender Relationships b	39.31 (7.59)	39.69 (10.05)	40.27 (8.99)	33.82 (6.13)
Offender Treatment Orientation	39.63 (5.35)	41.30 (5.89)	40.55 (5.71)	38.50 (6.33)
Inventory of Interpersonal Problems
Domineering / Controlling	6.28 (5.07)	7.22 (5.90)	7.03 (5.69)	4.38 (4.41)
Vindictive / Self-Centred	8.96 (4.66)	9.81 (6.76)	9.64 (5.97)	5.75 (4.92)
Cold / Distant	8.37 (6.44)	10.52 (7.76)	9.59 (7.38)	6.38 (6.21)
Socially Disinhibited	9.71 (6.33)	11.69 (6.69)	10.73 (6.64)	10.50 (5.71)
Non-Assertive a	10.42 (7.42)	13.12 (7.77)	11.48 (7.58)	14.00 (8.19)
Overly Accommodating a	9.58 (7.19)	12.18 (6.88)	10.99 (7.06)	10.25 (7.54)
Self-Sacrificing	10.40 (6.96)	13.12 (7.37)	11.89 (7.23)	9.38 (6.30)
Intrusive / Needy a	6.48 (5.73)	8.86 (6.15)	7.97 (6.06)	5.38 (6.28)
Total Score	70.04 (37.62)	88.65 (44.42)	79.25 (43.05)	66.00 (33.96)
Personal Safety and Security Scale
Number heated arguments	26.52 (35.55)	40.90 (115.58)	34.12 (89.23)	25.92 (35.12)
Number assaults without weapons	6.29 (16.43)	6.09 (15.08)	6.60 (16.98)	4.20 (2.49)
Number assaults with weapons	1.86 (8.51)	2.20 (8.28)	2.22 (8.98)	0.89 (1.62)
Number sexual assaults	0.26 (0.77)	1.21 (4.07)	0.84 (3.16)	0.10 (0.32)
Frequency inmates have weapons b	1.63 (1.55)	1.73 (1.79)	1.56 (1.62)	3.18 (1.54)
Frequency staff use force on inmates b	1.80 (1.47)	1.78 (1.40)	1.60 (1.34)	3.09 (1.44)
Frequency inmates use force on staff b	1.19 (1.30)	1.20 (1.15)	1.08 (1.16)	2.60 (1.07)
Likelihood of inmate being assaulted	1.38 (0.98)	1.48 (1.19)	1.33 (1.09)	2.00 (0.77)
Likelihood of staff being assaulted b	1.09 (0.90)	1.06 (1.04)	0.97 (0.91)	1.83 (1.03)
Self-Control Schedule
Total Score	28.14 (31.98)	32.14 (335.6)	30.54 (32.01)	24.57 (40.69)

Institutional Environment and Safety

There were a number of differences in the perceptions of the institutional environment held by the GP and the SU women. For four of the items of the Personal Safety and Security Scale, scores were significantly higher for SU women than their GP counterparts, indicating that SU women perceive it to be more likely that inmates have weapons on them or in their quarters, that staff be assaulted, that inmates use force on staff, and that staff use force on inmates. Additionally, GP respondents’ scores were significantly higher than were SU respondents’ scores on the Offender Relationships subscale of the Correctional Environment Status Inventory. This difference reveals that GP women perceived staff to have better relationships with inmates than did SU women.

In terms of pre- to post-test differences, a significant difference was found in the Staff Cohesion subscale of the Correctional Environment Status Inventory. This demonstrates that the inmate respondents perceived the institutional staff to be a more cohesive group at post- than at pre-test.

Interpersonal Problems

No differences were detected on the measures of interpersonal problems (Inventory of Interpersonal Problems, Self-Control Schedule) in comparing GP and SU women’s responses. However, three of the Inventory of Interpersonal Problems’ eight subscales, as well as the total score, differed significantly from pre- to post-test. More specifically, mean scores increased from pre- to post-test on the Non-Assertive subscale, the Overly Accommodating subscale, and the Intrusive / Needy subscale, demonstrating that the women had more interpersonal problems of various types at post-test than at pre-test.

STUDY 2

Method

Participants

Participants were all women incarcerated at the four regional women’s institutions (excluding Fraser Valley Institution) and at the four co-located institutions, for periods of six months prior to and after the opening of the secure units. The 60 women (34 pre-test and 26 post-test) for whom an official security level decision had not yet been finalized were excluded from the sample, leaving a total of 506 participants for whom data were available. Of these, there were 253 in each of the pre-test (prior to the openings) and post-test (after the openings) samples. Some women (n= 91) were included in both the pre- and post-test samples as they remained incarcerated throughout both time periods.

Description of Measures

This component of the evaluation was purely quantitative, and comprised reviews of inmate files maintained in the Offender Management System, the Service’s automated offender information database. This data sampling strategy, unlike that in Study 1, did not depend on voluntary participation, and encompassed the total federally sentenced woman (inmate) population at the relevant times. Moreover, this data source provided a second perspective on the women’s participation in pro-social activities and institutional adjustment.

Pro-Social Activities

The files of the women offenders were verified for participation in pro-social activities. This included participation in visits generally, and for private family visits specifically, as well as participation in employment and educational activities and programs. Participation in employment and educational activities was assessed globally. Participation in programs was assessed both globally (i.e., any program) and in a number of specific areas: substance abuse, violence prevention, cognitive and living skills, women-specific programs, psychology, and ‘other’ programs (e.g., personal development, HIV / AIDS).¹⁴

Poor Institutional Adjustment

Women’s records were also examined for participation in institutional incidents and other behaviours indicative of institutional adjustment problems. First, women’s participation in institutional incidents (as an instigator or an associate) was examined, both globally and specifically: assault incidents (including inmate fights, assaults on inmates, and assaults on staff), self-harm incidents (suicide, attempted suicide, and self-inflicted injury), substance abuse incidents (under the influence), disturbance incidents (minor disturbance, fire, damage to government property, cell extraction), and disciplinary incidents (possession or transportation of contraband, disciplinary problems, intelligence, possession of unauthorized item, other incidents) were all noted. Second, each woman’s file was examined for indications of positive drug test results. Third, each woman’s involuntary admissions to administrative segregation were examined. Admissions to segregations were assessed globally as well as specifically: admissions due to being in danger and due to jeopardizing the safety of the institution or another individual.¹⁵

Procedure

Electronic file records were examined retrospectively for periods of six months prior to and after the opening of the secure units. The last date of each six month period coincided with the last date of on-site data collection at each institution, ensuring comparability of findings from the two studies. For those offenders who had been admitted or transferred during this six month period, data were examined only since the admission or transfer date.

Analyses

Since coding for the number of instances of each of the above variables resulted in positively skewed distributions, all variables were instead coded dichotomously as ‘present’ or ‘not present’. This necessitated the use of non-parametric analytic procedures; a number of hierarchical log-linear models were computed for each variable.¹⁶

Results

Sample

Institutional Information

The proportions of the total sample of 506 women housed at each institution can be seen in Table 6. EIFW housed the largest proportion of women.

Table 6. Percentage of Women Housed at Each Institution at Pre- and Post-test

Institution	Pre-test (n = 253)	Post-test (n = 253)	Total (N = 506)
Edmonton Institution for Women	25%	32%	29%
Grand Valley Institution	22%	27%	25%
Joliette Institution	23%	25%	24%
Nova Institution	11%	16%	14%
Regional Psychiatric Centre	4%	-	2%
Saskatchewan Penitentiary	8%	-	4%
Regional Reception Centre	3%	-	1%
Springhill Institution	4%	-	2%

Of the 253 pre-test participants, 83% were incarcerated in GP and 17% were incarcerated at maximum security. Of the 253 post-test participants, 87% were incarcerated in GP and 13% were incarcerated in the SU.

Demographic Information

Of the women, 63% were Caucasian, 24% were Aboriginal, 8% were Black, and 4% another visible minority. Though these distributions did not differ by time, they did by unit, χ2(3, N = 504) = 18.34, p < .001. Specifically, the SU included a larger proportion of Aboriginal women (43% as compared to 21%) than did the GP sample. The women’s ages ranged from 19.9 to 65.8 years, with a mean of 35.5 years (SD = 9.69). Average age did not differ either from pre- to post-test or by unit.

Offender Management System Data

Participation in Pro-Social Activities

The proportions of women whose Offender Management System file indicated participation in pro-social activities are presented in Table 7 both by time and by unit.

Table 7. Percentage of Offenders Participating in Pro-Social Activities

Indicator	By Time		By Unit
	Pre-test (n = 253)	Post-test (n = 253)	GP (n = 430)	SU (n = 76)
Visits
All visits b	59%	61%	63%	42%
Private family visits	7%	8%	8%	4%
Employment
All employment a	55%	65%	60%	62%
Education
All education a	34%	51%	42%	46%
Programs
Any program a	80%	92%	86%	87%
Substance abuse a	19%	24%	24%	7%
Violence prevention	4%	2%	4%	1%
Cognitive / living skills	27%	21%	25%	16%
Women-specific	11%	16%	14%	11%
Psychology c	16%	23%	16%	38%
Other a	4%	1%	2%	1%

For about half of the indicators of participation in pro-social activities, analyses demonstrated that time, unit, and the variable of interest were not related to one another. On the other hand, more women participated in employment and educational activities, programming generally, and substance abuse programming at post- than at pre-test. Additionally, the proportion of women participating in psychological programming increased from pre- to post-test overall, with fewer GP than SU women receiving visits at each testing session. Conversely, fewer women participated in ‘other’ programs at post-test. Finally, more GP women than SU women received visits.

Poor Institutional Adjustment

Table 8 presents, separately by time and by unit, the proportions of women whose Offender Management System files included indications of institutional maladjustment.

Table 8. Percentage of Offenders Displaying Indicators of Poor Institutional Adjustment

Indicator	By Time		By Unit
	Pre-test (n = 253)	Post-test (n = 253)	GP (n = 430)	SU (n = 76)
Institutional Incidents
All incidents b	39%	44%	34%	83%
Assault b	17%	20%	13%	47%
Self-harm b	4%	5%	3%	14%
Substance abuse	1%	2%	1%	3%
Disturbance b	6%	8%	3%	29%
Disciplinary b	30%	30%	24%	66%
Drug Test Results
All positive results a	4%	7%	6%	1%
Involuntary Segregation Admissions
All admissions b	26%	26%	17%	75%
Danger to inmate b	4%	8%	3%	21%
Jeopardize safety of institution / individual b	22%	23%	15%	62%

In all cases except for incidents relating to substance abuse, a relationship was found between either time or unit and the indicator of poor institutional adjustment. Across all women, there were more positive urinalysis results at post- than at pre-test. In terms of participation in institutional incidents and of involuntary segregation admissions, however, the differences were at the unit level, with greater proportions of SU women than GP women displaying these indicators of poor institutional adjustment. Taken together, these results suggest, not surprisingly, that the SU women display a more problematic profile in terms of institutional adjustment than do the GP women.

DISCUSSION

Study Limitations

Study 1 is limited in part by its generalizability. Not all institutions were represented in the same numbers, either from pre- to post-test, or by mode of inquiry. Due to the aggregative nature of data analyses, the study’s results are most representative of those institutions where many staff and inmates participated in the study. Moreover, those staff and inmates who agreed to participate in the study may differ in some way from those who did not; it is impossible to determine the extent to which the study’s results are reflective of self-selection bias.

A further limitation, for the quantitative component of the study specifically, is that of sample size. As there are relatively few women offenders in the federal correctional system, and therefore, few staff working with them, it is difficult to recruit sufficient participants for statistical purposes. The size of the study’s sample impacts statistical analyses in two inter-related ways. First, it decreases the likelihood that statistical analyses will be able to detect effects in the data. This concern is particularly salient for the staff sample, which, due to matching across testing periods, was reduced to only 32 persons. The fact that statistical trends were considered for this sample, however, helps to mitigate this limitation. Second, the small sample size limits the more detailed analyses that can be completed. Specifically, too few participants and staff members completed the standardized assessment measures to allow for separate analyses by institution, yet the fact that there were some differences by institution within the qualitative data suggests that such analyses would have produced informative results.

Summary of Findings

This study aimed to assess the impacts of the openings of the secure units on all affected parties: women classified as maximum security and staff working with them, as well as women classified as minimum and medium security and staff who work with them. Both qualitative and quantitative techniques were used to evaluate the effects of the openings. As the analysed data reflect only impacts occurring within approximately six months of the opening, findings reflect only immediate impacts. Nonetheless, a number of areas for potential improvements emerged.

Do the secure units meet the needs of the women classified as maximum security?

In general, these data suggest that there have been marginal improvements in the extent to which the needs of the secure unit women are being met, but that there is room for amelioration. According to self-reported data, the women’s involvement in most programs, visits, spiritual activities, and recreational activities did not increase once transferred from the co-located units to the regional women’s institutions. A number of staff and inmates also indicated that the programs offered tended not to be correctional programs necessary for completion of correctional plan objectives. (The exception was Dialectical Behaviour Therapy, which was appreciated by women and staff alike.) The results of Study 2, however, showed that a larger proportion of women were involved in educational activities, as well as substance abuse programming and programming generally after the openings, revealing marginal improvement in these areas.

With regards to involvement in work-related activities, the data were less clear. While women reported spending less time in work and vocational training after the units’ opening, OMS data demonstrated that the proportion of women involved in work increased after the units’ opening. There are two possible explanations for this discrepancy. It may be that those women who provided self-report data were not representative of the relevant population of women. Alternatively, these results may be attributable to the different definition of involvement in work-related activities used for each analysis. While the self-report data measured hours per week involved in work, the OMS data represented whether the women had participated in work-related activities at all during the six month period under study. It is possible, then, that these results indicate that more individual women were involved in work after the units’ openings, but that each woman is working fewer hours. It is impossible, however, to know which of these alternate explanations is correct, and therefore these results are inconclusive.

Regardless, it is noteworthy that many secure unit respondents suggested that the work tasks available were too limited to fill their time and did not allow for the acquisition of transferable skills (e.g., cleaning for 30 minutes each day). Accordingly, a few women expressed that they were bored, and in some cases, unmotivated. Since the completion of this study, this concern has also been reported elsewhere (Glube, Audette, Henriksen & Stobbe, 2007).

The secure unit women also reported frustration regarding communication and relationships. The women felt that there was room for improvement in this area both prior to the opening of the units and afterwards, when they indicated that information and directions received were sometimes inconsistent, particularly when general population staff were called upon to work in the secure units. Additionally, quantitative data indicated that, relative to the general population women, the secure unit women perceived staff to have poorer relationships with offenders. Finally, the women reported more interpersonal problems after being transferred to the units than they did prior to their transfers.

This did not, however, lead to an increase in institutional misbehaviour. Involvement in incidents and involuntary segregation admissions remained unchanged from prior to after the units’ opening, though the proportion of positive urinalysis analyses increased. This suggests that in the main, despite increased frustrations, the women’s behaviour remained stable. One possible reason for this is the availability of Dialectical Behaviour Therapy, in which the majority of the women participated; Dialectical Behaviour Therapy has previously been found to assist high need women offenders in managing their emotions and behaviours (Sly & Taylor, 2003).

Do the secure units affect the secure unit staff?

It is clear that the opening of the units had direct and profound impacts on the staff working there. Most of the secure unit staff who provided qualitative data indicated that their levels of stress, frustration, and burnout have increased, while their job satisfaction and morale have decreased. A number of the respondents indicated that the stress is so serious that it affects their personal lives. Sources of stress included the perceived dangers associated with working with women classified as maximum security, a perceived lack of support from management, pressure to work excessive overtime, and concerns relating to safety.

Quantitative results revealed that secure unit staff were more concerned about their safety than were their counterparts working in the general population. Moreover, after the openings, staff were experiencing more burnout and had less confidence in the ability of the Service’s policies and procedures to protect them from harm from inmates. They were also experiencing more burnout related to depersonalization than were general population staff. This may have been related to the finding that the frequency of incidents (including verbal and physical assaults, self-injurious behaviour, a hanging, and a hostage-taking) was higher than anticipated, and the level of assistance in the form of Critical Incident Stress Management, in some cases, lower than expected. Some staff also felt that they would like greater access to static security resources to use in managing inmates and preventing incidents.

One positive impact of the openings on the secure unit staff, on the other hand, was to increase the cohesiveness of the secure unit teams. Many secure unit staff indicated that the challenges specific to working in the secure unit environment, along with the support among staff, have led to increased team unity.

Does the addition of the secure units affect the functioning of the rest of the institution?

Results suggested that the opening of the secure units has affected the functioning of the remainder of the institution considerably. Specifically, staff working in the general population experienced increases in stress and frustration and decreases in job satisfaction and morale, particularly when called upon to work in the secure units. In addition to the stressors associated with the units themselves, general population staff had the added challenge of not having completed the same training as the secure unit staff. Many general population staff indicated that their perceived safety when working in the secure units would be increased with further training.

Quantitative data revealed that general population staff’s burnout due to emotional exhaustion also increased since prior to the units’ opening. A number of participants who provided qualitative feedback indicated that they resented being asked to work in the secure units, either on a regular basis or in emergencies. Data gathered from inmate participants also suggested that the women perceived the general population staff to be a more unified group at the second testing session; this may be linked to staff working together to manage higher levels of stress and frustration.

The general population women were also seriously impacted by the openings. They experienced increases in stress, tension, and in number of interpersonal problems since the openings as a result of associated impacts on their privileges. More specifically, both staff and women reported that the general population women’s movement, access to visits and recreational activities, and access to staff were more restricted. Of particular concern was the practice of assigning general population women to secure unit primary workers’ caseloads; the women in this situation expressed considerable frustration, as they indicated that they very rarely saw their primary workers. Moreover, the women indicated that the atmosphere of the institutions had become more strict and security-focused subsequent to the unit’s opening.

Quantitative data also indicated that the proportion of general population women whose files included indications of pro-social adjustment remained largely unchanged from before to after the units’ opening. However, the women reported engaging in fewer hours of work-related activity each week after the opening of the secure units. In contrast, OMS data indicate that the proportion of women involved in educational and employment activities, as well as in programming, increased after the units’ opening. As such, the results regarding work-related activity are inconclusive. Finally, according to these data, the only difference in institutional behaviour among the general population women since the units’ openings was a slight increase in positive urinalysis results.

Are there adequate resources and support for establishing the secure units?

Staff expressed their perception that further training would increase their own and inmates’ safety. While secure unit staff would like refresher training in the areas of use of force and arrest and control, general population staff would like to receive training equivalent to that of the secure unit staff. Until such time as general population staff receive supplementary training, many staff (both general population and secure unit) indicated that they feel that supplementing the secure unit’s roster with general population staff is an unsafe practice. A number of staff members also expressed unease that many secure unit staff were relatively inexperienced.

Another area of concern is that of communication and support. Qualitative data included a number of mentions of front-line staff feeling that their concerns were not being heard by management, and that their input was being ignored. Data suggested that this is an area of difficulty both in managing the secure units and in managing the women’s facilities as a whole.

Does the opening of the secure unit create any positive or negative unintended effects?

One of the most important unintended effects was a sense of divisiveness that developed between the general population and secure unit staff. For the general population staff, this may have been attributable to resentment of the expanded professional responsibilities and higher levels of inmate frustration associated with the opening. Similar reasons may have been applicable for secure unit staff - that is, resentment resulting from their perception that those general population staff asked to work in the secure unit sometimes were unhelpful and provided conflicting information to the women. It is also possible, however, that this divisiveness was related ‘in-group / out-group’ thinking brought upon by the strong cohesiveness among the secure unit staff.

The presence of a number of other unintended effects was also explored. Though it was originally hoped that the proximity of the general population would lead to an increase in the secure unit women’s motivation to decrease their security levels, this did not seem to be the case. Additionally, there were suggestions that the secure units be moved to a separate, self-contained location, and that the building of a Special Handling Unit for women offenders be considered. It should be noted that this suggestion runs counter to a number of the Service’s goals for women offenders, including allowing women to be housed in their own regions, providing opportunity for integration among women of various security levels, and focusing on dynamic as opposed to static security.

On a more positive note, it was initially feared by some respondents that, after the openings, women from the general population may find themselves reclassified as maximum-security with greater frequency than prior to the units opening; fortunately, in the main, this did not occur. Also positively, it was found that many of the women of the general population cared about the well-being of the secure unit women, and wanted them to be more involved in the institutional community.

Recommendations

The findings of this study lead the researchers to propose three recommendations. Specifically, it is recommended that steps be taken to increase the number of staff at the regional women’s facilities, that communication be ameliorated, and that the impacts of the units’ opening be reexamined once the units have been open for a longer period of time. The latter is particularly important as study results reflect only the immediate (i.e., 6 month) impacts of the opening of the secure units.

Increased Staffing

Although the fiscal limitations of this recommendation are recognized, increasing the number of staff in the facilities would have a number of impacts. Indeed, the issue of staff insufficiency was also mentioned in the Expert Committee Review of CSC ’s Ten-Year Status Report on Women’s Corrections, 1996-2006 (Glube et al., 2007), published after this study was completed. Notably, since respondents from a number of facilities indicated that there was a great deal of overtime being worked, it may actually be more fiscally responsible to expand staff rosters and diminish the frequency with which staff is paid at overtime rates.

From a more humanistic point of view, increased staffing levels would likely benefit staff in both general population and the secure unit. A major advantage of having more staff would be that the frequency with which general population staff would be asked to work in the secure unit would be diminished substantially. This would allay concerns relating to general population staff not having secure unit training, as well as the issue of secure unit staff having to work with colleagues who are not happy to work in the unit. This would likely lead to improvements in levels of tension and in job satisfaction for both general population and secure unit staff.

Increasing staffing levels would also be useful to general population and secure unit women. Because dedicated secure unit staff would all have completed the necessary training, secure unit women would be more likely to receive more consistent information, as well as more consistent behavioural reinforcement. General population women would benefit from not having their primary workers stationed in the secure unit, thereby avoiding the delays and difficulties associated with this. Moreover, both offender groups would also benefit from having more staff available to support and supervise them, including, for general population women, on escorted temporary absences, and for secure unit women, on activities off the pods or with the general population women, once they have reached level four. This may lead to decreases in frustration and tension for the women. As frustration and tension are related to interpersonal aggression, such changes could in turn lead to a safer environment for staff and inmates alike.

A related suggestion would be to increase the recruitment of volunteers within the regional women’s institutions. Volunteers may be found to fill a number of roles, and may therefore take on a portion of the tasks facing staff, allowing staff a better opportunity to meet their primary responsibilities. It is suggested that a greater number of volunteers be recruited to act as escorts on escorted temporary absences for those minimum-security women for whom a citizen escort is appropriate. Additionally, it would be useful to have volunteers involve the women at all levels of security in activities and non-core programs. This would both alleviate some of the boredom experienced by women and free CSC program staff to focus on those programs most critical to the women’s correctional plans.

Ameliorated Communication

The second recommendation is to take steps to ameliorate the communication between staff and management, as well as between the women and staff. Notably, this recommendation was also made previously in an unrelated context. Researchers studying incidents involving federally sentenced women hostage-takers also suggested that improving communication, both between staff and management and between staff and inmates, may lead to fewer such hostage-takings (Taylor & Flight, 2003).

The current findings suggest that there is still room for improvement in these lines of communication. With regards to staff and management, it may be beneficial to develop standardized procedures for expressing concerns and ensuring these concerns are responded to; this may assist with staff’s feelings of not being heard. This would also reduce the number of situations wherein management are truly unaware of staff’s concerns, where staff are unaware that management has considered a concern, or where the reasons for management decisions are unclear or unknown.

The communication between staff and women should also be addressed. In this area, implementation of the increased staffing recommendation may be beneficial, as having dedicated staff for each unit would likely eliminate situations where incorrect information is communicated due to uncertain knowledge of policies and procedures. This would improve the level of consistency in information and direction for secure unit women.

Improved communication could also be helpful with the general population women. Many of these women indicated that they had been incorrectly informed of the potential impacts of the openings of the secure units. Where possible, it would be positive to involve representatives of the general population in discussions of how to manage these impacts. During the site visits conducted for this study, it was noted that the general population women had many suggestions regarding the issue of shared spaces; having an opportunity to have these suggestions considered may result in the women being more satisfied with the final decisions.

A final area where increased communication may have positive impacts is between the regional facilities. Among the women, some facilities are seen as superior to others in certain respects (e.g., in terms of employment and programming activities, staffing practices, meals). It may be useful for the Service to provide an opportunity for representatives of each facility to meet and discuss their respective procedures and systems in order to share best practices. In addition to potentially improving facilities’ practices in certain areas, this would add consistency to the women’s experiences of the regional facilities, thereby facilitating their transfers from one to another.

Reexamination of Impacts

Finally, it is recommended that the impacts of the opening of the secure units be reexamined once these units have been open for a longer time. The present results represent impacts occurring within the first six months of the units’ operation. It is not uncommon for difficulties to occur in relation to new initiatives, and for some time to be necessary to iron out initial problems. Indeed, other correctional researchers have also identified initial difficulties associated with institutional adaptations. For instance, in reviewing American prisons adapted to accommodate a new group of offenders with different needs (in this case, offenders from another jurisdiction), Green (2003) noted "start-up problems" (p. 59), including inadequate dissemination of information and staff finding it difficult to manage offenders with whom they were inexperienced. It seems likely that the same phenomenon played a role for the secure units, and that the resolution of these implementation difficulties would positively impact prospective findings.

In the current context, a number of important changes have already been put into practice since the opening of the units and the completion of this research. These include certain structural changes and clarification of relevant policies. As such, it would be beneficial to reexamine the impacts associated with the units again in the future. Such an examination would have the added advantage of allowing for the inclusion of the secure unit at Fraser Valley Institution, which was omitted from this study because it opened after data collection ended.

Conclusion

The present report summarizes the results of two studies on the immediate impacts of the opening of the secure units at the regional women’s facilities. A number of areas were identified as having potential for improvement, specifically with regards to the impacts on the women classified as maximum security, the staff who work with them, and the functioning of the institutions. Among other themes, all impacted parties expressed that stress and tension rates have increased, that communication has been problematic, and that there have been more negatives than positives associated with the opening of the units.

A number of specific findings are particularly noteworthy in light of recommendations stemming from previous research with women classified as maximum security. For example, though Warner (1998) emphasized the importance of staff consistency in dealing with women classified as maximum security, these data suggest that this has been problematic. Similarly, while both McDonagh (1999) and the Secure Unit Operational Plan (CSC, 2003) underlined the importance of individualized and intensive programming for the women of the secure units, the present findings suggest that, with the exception of Dialectical Behaviour Therapy, women and staff perceive limited programming options. The results of Study 2, however, indicate that program involvement has either increased (substance abuse programs, programs generally) or remained unchanged since the opening of the unit. A possible explanation for this discrepancy would be that those programs which are offered are not those perceived by the women and staff as being appropriate to their needs.

Though these findings are not uniformly positive in nature, they do provide valuable information. Using this knowledge, wardens, team leaders, and regional and national headquarters staff will better know in which areas to focus their efforts. As mentioned, it is also expected that a number of the difficulties associated with the units’ opening will be minimized over time as problems are encountered and steps taken towards resolution. Moreover, recommendations have been provided to decrease or help manage the impacts of the units on women and staff at the regional facilities. These recommendations, along with awareness of the challenges specific to the opening of the units, could also be of utility to the Service in managing any substantial transitions in future.

REFERENCES

Blanchette, K. (1997). Risk and need among federally sentenced female offenders: A comparison of minimum-, medium-, and maximum-security inmates. Report R-58. Ottawa, ON: Correctional Service Canada.

Correctional Service Canada (2003). Secure unit operational plan. Ottawa, ON: Correctional Service Canada.

Correctional Service Canada (2005). The changing federal offender population: Profiles and forecasts, 2005. Ottawa, ON: Correctional Service Canada.

Cullen, F.T., Link, B.G., Wolfe, N.T., & Frank, J. (1985). The social dimensions of correctional officer stress. Justice Quarterly, 2, 505-533.

Delveaux, K. & Blanchette, K. (2000). Results of the evaluation of the Peer Support Program at Nova Institution for Women. Report R-87. Ottawa, ON: Correctional Service Canada.

Horowitz, L. M., Rosenberg, S. E., Baer, B. A., Ureno, G., & Villasenor, V. S. (1988). Inventory of Interpersonal Problems: Psychometric properties and clinical applications. Journal of Consulting and Clinical Psychology, 56, 885-892.

Glube, C., Audette, M., Henriksen, S., & Stobbe, B. (2007). Moving forward with women’s corrections: The expert committee review on the Correctional Service of Canada’s ten-year status report on women’s corrections, 1996-2006. Ottawa, ON: Correctional Service Canada.

Greene, J.A. (2003). Lack of correctional services: The adverse effect on human rights. In A. Coyle, R. Neufield, & A. Campbell (Eds.), Capitalist punishment: Prison privitizaton and human rights (pp. 56-66). Atlanta, GA: Clarity Press.

Irving, J., Taylor, K., & Blanchette, K. (2001). Results of an evaluation of the Intensive Treatment Program at Burnaby Correctional Centre for Women. Report R-128. Ottawa, ON: Correctional Service Canada.

Maslach, B., Jackson, S.E., & Leiter, M.P. (1996). Maslach Burnout Inventory manual (3rd ed.). Palo Alto, CA: Consulting Psychologists Press.

McDonagh, D. (1999). Federally sentenced women maximum security interview project: "Not letting the time do you". Ottawa, ON: Correctional Service Canada.

Morin, S. (1999). Federally sentenced women maximum security interview project: Whatever happened to the promises of Creating Choices? Ottawa, ON: Correctional Service Canada.

Paulhus, D.L. (1998). Paulhus Deception Scales: The Balanced Inventory of Desirable Responding. Toronto, ON: Multi-Health Systems Inc.

Robinson, D., Lefaive, P., & Muirhead, M. (1997). Results of the 1996 CSC staff survey: A synopsis. Brief B-17. Ottawa, ON: Correctional Service Canada.

Rosenbaum, M. (1980). A schedule for assessing self-control behaviors: Preliminary findings. Behavioral Therapy, 11, 109-121.

Sly, A. & Taylor, K. (2003). Preliminary evaluation of Dialectical Behaviour Therapy within a women’s structured living environment. Report R-145. Ottawa, ON: Correctional Service Canada.

Sly, A., Taylor, K, & Blanchette, K. (2003). An evaluation framework for Dialectical Behaviour Therapy with women offenders. Ottawa, ON: Correctional Service Canada.

Syed, F. & Blanchette, K. (2000). Results of the evaluation of the Peer Support Program at Joliette Institution for Women. Report R-88. Ottawa, ON: Correctional Service Canada.

Task Force on Federally Sentenced Women. (1990). Creating Choices: Report of the Task Force on Federally Sentenced Women. Ottawa, ON: Correctional Service Canada.

Taylor, K. & Flight, J. (2003). Hostage-taking incidents involving women inmates: A profile and exploratory investigation. Report R-143. Ottawa, ON: Correctional Service Canada.

Triplett, R., Mullings, J.L., & Scarborough, K. (1999). Examining the effects of work-home conflict on work-related stress among correctional officers: A path model. Journal of Criminal Justice, 27, 371-385.

U.S. Department of Justice (1990). Prison Social Climate Survey: Inmate Version. Washington, DC: Federal Bureau of Prisons.

U.S. Department of Justice (1993). Prison Social Climate Survey: Staff Version. Washington, DC: Federal Bureau of Prisons.

Warner, A. (1998). Implementing Choices at regional facilities: Program proposals for women offenders with special needs. Ottawa, ON: Correctional Service Canada.

Warr, P., Cook, J., & Wall, T. (1979). Scales for the measurement of some work attitudes and aspects of psychological well-being. Journal of Occupational Psychology, 52, 129-148.

Wolfus, B. & Stasiak, E. (1996). The Correctional Environment Status Inventory: A tool for measuring the quality of correctional environments. Brampton, ON: Correctional Institute, Ministry of the Solicitor General and Correctional Services.

 

Appendix A: Staff and Inmate Interview Protocols

Guide for interview with GP Psychologist and/or GP staff members

Support

Do you feel that the secure unit is considered important by staff at this facility?

How about:
Mental health professionals
Primary Workers
Management / Warden / Deputy Warden
RHQ / NHQ

Have you noticed any resistance to the opening of the secure unit?
By staff?
By women?
By RHQ or NHQ?

Do you feel that GP staff support the secure unit staff?

Do you feel that women support the secure unit staff?

Do you feel that RHQ/NHQ support the secure unit staff?

Implementation

Would you say that the pace at which the secure unit opened was sufficient?

Do you have any comments regarding the opening itself?

Is it possible that GP staff would be needed to work in the secure unit (such as in the case of an emergency or to fill in for someone else?)
If Yes: Do you feel that you are appropriately trained to do so?

Relationships

How would you assess the reactions/relationships between the max women and the secure unit staff?

Have you noticed any changes in the general atmosphere on the facility since the opening of the secure unit?

Have you noticed any changes in the relationships amongst staff members since the opening of the secure unit?

Have you noticed any changes in the relationships between staff and women since the opening of the secure unit?

Have you noticed any changes in the relationships amongst inmates since the opening of the secure unit?

Impacts

What was the transition like when the max women were transferred into the secure units?
(a) For GP staff?
(b) For GP women?

Would you say that the primary workers have adjusted to this transition?

Would you say that the behavioural counsellors have adjusted to this transition?

Would you say that the women have adjusted to this transition?

What do you view as being the greatest source of stress related to its opening?

Have you noticed any changes in the GP staff since the opening of the secure unit? (stress, job satisfaction, burnout, perception of safety?)

Have you noticed any changes in the GP women since the opening of the unit? (Such as how they deal with personal issues, incarceration?)

Have you faced any problems as a result of the opening?
If yes: How were these addressed?

What advantages have resulted from the opening?

What disadvantages have resulted from the opening?

Have you noticed any effects (either positive or negative) resulting from the opening of the secure unit that were not anticipated?

Are there any effects that you foresee the secure unit causing?

Prompt:
With regards to relationships between staff (GP and Secure)?
With regards to relationships between staff and women?
With regards to relationships amongst women (GP and GP and GP and Secure)?

General Comments

Is there anything that I have left out that you would like to mention, any comments, suggestions etc?

Guide for interview with women in the secure unit

How long have you been at this secure unit? ___________________________

Previous Location:______________________________________________
Length of time at that location:____________

Other Max Locations? __________________________________________
Length of time at each:________________________________________

Any time spent in multi? (ever medium security) Where?__________________
Length of time:___________________

Expectations

Now this might be difficult to remember but I'd like you to try your best. Prior to being transferred, how did you feel about the opening of the secure units?
Prompt: (what were your expectations, fears, hopes?)

Now that they have opened, how do you feel?

Why do you think the secure units were opened?

What programs/services are offered here?

What programs/services are you involved in?

Are there any negatives related to these programs/services?

Are there any positives regarding these programs/services?

Are the programs/services offered here very different from those at your previous facilities?

Implementation

How did you find the transfer to the secure units? Prompt: (proper pace, informed of when/where, stressful/anxiety-provoking, smooth, exciting?)

If required: Did you find it stressful?
What are the advantages? Prompt: (cleaner, less women, less fighting, more space, more interaction)
What are the disadvantages?
What is different from your previous location?

Is there anything you would have changed about the process of this transfer that would have made the transition easier for yourself and the other women?

Relationships

Are you still living with the same women you were with prior to this transfer?
If Yes: Have these relationships changed since being transferred to the secure unit? Prompt: (do you see them more/less, spend more/less time in yard, more/less time alone?)

Do you and the other women get along with the staff?

Is the relationship between the women and staff different here than it was at your previous facility?

Has your relationship with staff changed since you first arrived? Prompt: (more/less friendly? Do you feel they are there if you need them?)

Do you feel that there is more (or less, or the same) contact between staff and the women now than there was at your previous location? Prompt: (more chatting, more program time, more one-on-one time?)

Impacts

How have the secure units impacted you and other women you live with?

Have you noticed any changes in yourself and the other women since being transferred to the secure unit? Prompt: (How you deal with personal issues, your general outlook on life?)

Would you say that you/everyone has adjusted to the change?

Have any problems occurred as a result of the opening?
If yes: How were they addressed?

Are there any advantages of being in the new secure unit at a facility with minimum and medium security women? Prompt: (incentives to lower your own security level?)

Are there any disadvantages of being in the new secure unit at a facility with minimum and medium security women?

Since you have come here have there been any surprises, any changes that have occurred that you did not expect? Prompt: (anything to do with staff; programming; health care; food quality; physical structure - yard space, cell size?)

General comments

Do you have anything that you would like to add that we haven't already talked about? (Comments, opinions, feelings, etc)

Guide for interview with secure unit staff, Team Leader and Assistant Team Leader

Support

Do you feel that the secure unit is considered important by staff at this facility?

How about:
Mental health professionals
Primary Workers
Warden/Deputy Warden
RHQ/NHQ

Have you noticed any resistance to the opening of the secure unit?
By staff?
By the GP women?
By secure unit women?
By RHQ?
By NHQ?

Do you feel that you have adequate support by staff here at the institution for running the secure unit?

Do you feel that you have adequate support from RHQ and NHQ for running the secure unit?

Training

How were staff selected to work in the secure unit, was it volunteer/mandatory, and what was the selection criteria?

Where were most of staff working prior to working in the secure unit?

Do you think that the training and selection of secure unit staff was sufficient?

What from the training do you think was helpful? Why?

Would you recommend any changes to the training provided to the Secure unit staff? What?

In the case of an emergency or crisis, is it possible the GP staff would have to intervene/assist in the secure unit?
If Yes: Do you feel that the GP staff are adequately trained to do so?

Implementation

Would you say that the pace at which the secure unit was opened was appropriate?
Do you have any comments regarding the opening itself?

It is my understanding that the secure unit women will be sharing certain facilities such as Visits & Correspondence, gym, and some program rooms with the rest of the GP women. I also understand that when this occurs staff must try to conduct movement such that there is no (or minimal) interaction between the GP women and Secure unit women. How feasible has this issue of shared spaces/separation issues been?

Have there been any problems with movement on or off the unit?

It is my understanding that when maximum-security women are involved in major incidents that cause serious harm or seriously jeopardizes the safety of others, a management protocol informs the management of such women. Have you had to implement the management protocol since the women have been transferred? How effective do you find this?

Relationships

How would you describe the reactions/relationships between the max women and the secure unit staff?

Have you noticed any changes in these relationships in the 6 months since the unit has opened (from the first week the women were in the new unit until now)?

Does the secure unit staff work together as an interdisciplinary team, making decisions together? Is this effective?

Have you noticed any changes in the general atmosphere on the facility due to the secure unit opening?

Have you noticed any changes in the relationships amongst staff members due to the secure unit opening?

Have you noticed any changes in the relationships between the staff and women due to the secure unit opening?

Have you noticed any changes in the relationships amongst the GP women due to the secure unit opening?

Impacts

What was the transition like when the women were transferred into the secure units? For staff, for women?

Would you say that everyone has adjusted to this transition?

Have you noticed any changes in the secure unit staff since the opening of the secure unit? Prompt: (stress, job satisfaction, burnout, perception of safety?)

Have you noticed any changes in the secure unit women since the opening of the unit? Prompt: (How they deal with personal issues, incarceration)

Have any problems occurred as a result of the opening?
If yes: How were these addressed?

What do you view as being the greatest source of stress related to its opening?

What advantages have resulted from the opening?

What disadvantages have resulted from the opening?

Have you noticed any effects (either positive or negative) of the secure unit that were not anticipated?

Are there any problems or advantages related to the opening of the secure unit that you might foresee occurring in the future? Prompt: (With regards to relationships between staff (GP and Secure)? Relationships between staff and women? Relationships amongst women (GP and Secure)?)

General Comments

Is there anything that I have left out that you would like to mention, any comments, suggestions etc?

Guide for interview with Warden and/or Deputy Warden

Support

Do you feel that the secure unit is considered important by staff at this facility?

How about:
Mental health professionals
Primary Workers
RHQ
NHQ

Have you noticed any resistance to the opening of the secure unit?
By staff?
By GP or Secure unit women?
By RHQ or NHQ?

Do you think that you have the support both regionally and nationally for running the secure unit?

Do you think that you have the support from staff for running the secure unit?

Do you think that you have the support from all (GP & secure) the women for running the secure unit?

Training

How were staff selected to work in the secure unit, was it volunteer/mandatory, and what was the selection criteria?

Do you think that the selection of secure unit staff was sufficient?

Where were most of staff working prior to working in the secure unit?

Do you think that the training of secure unit staff was sufficient?
What was helpful? Why?

Would you recommend any changes to the training of staff in the secure unit?

In the case of an emergency or crisis, is it possible the GP staff would have to intervene/assist in the secure unit?
If Yes: Do you feel that the GP staff are adequately trained to do so?

Implementation

Would you say that the pace at which the secure unit was opened was appropriate?
Do you have any comments regarding the opening itself?

It is my understanding that the secure unit women will be sharing certain facilities such as Visits & Correspondence, gym, and some program rooms with the rest of the GP women. I also understand that when this occurs staff must try to conduct movement such that there is no (or minimal) interaction between the GP women and Secure unit women. How feasible has this issue of shared spaces/separation issues been?

It is my understanding that when maximum-security women are involved in major incidents that cause serious harm or seriously jeopardizes the safety of others, a management protocol informs the management of such women. Have you had to implement the management protocol since the women have been transferred? How effective do you find this?

Relationships

Have you noticed any changes in the general atmosphere on the facility due to the opening of the secure unit?

Have you noticed any changes in relationships amongst staff members?

Have you noticed any changes in relationships between staff and inmates?

Have you noticed any changes in relationships amongst the women (GP & Secure)?

Impacts

What was the transition like when the women were transferred into the secure units?
For GP staff?
For GP women?

What would you say has been the most challenging transition resulting from the opening?
For GP staff?
For GP women?

Have you noticed any changes in the general population staff since the opening of the secure unit? (stress, job satisfaction, burnout, perception of safety?)

Have you noticed any changes in the general population women since the opening of the unit? (tension, perception of safety)

Have any problems occurred as a result of the opening?
If yes: How were these addressed?

What do you view as being the greatest source of stress (or disadvantage) related to its opening?

What do you view as being the greatest advantages resulting from the opening?

Have you noticed any positive effects of the secure unit that were not anticipated?

Have you noticed any negative effects of the secure unit that were not anticipated?

Are there any problems related to the opening of the secure units that you foresee occurring in the future? Prompt: (With regards to relationships between staff (GP and Secure), between staff and women, or amongst women (GP and Secure)?

General Comments

Is there anything that I have left out that you would like to mention, any comments, suggestions etc?

Appendix B: Focus Group Protocol

Focus group with general population inmates

Were the GP women well informed of the opening of the secure unit?
Were you told what this involved?
Did you know when this was happening?
When did you find out about this?
How did you find out?

Before the secure unit opened, how did most GP women feel about it opening at this facility? (What were your feelings, expectations, fears, hopes, wants, etc.)
Did most think it was a good thing?
Did most think it was a bad thing?
Did most not care either way?
Why was this?

Is the opening of the secure unit something that is talked about among the GP women?
What, if anything, has been said by the GP women about the secure unit?

Now that it has opened, do most GP women view the opening of the secure unit at this facility as a good thing, a bad thing?
Do most not care either way?
Have opinions about the secure unit changed since it has opened here?

Have the secure units impacted women in the general population?
Loss of privileges? (gym, movement, visitation?)
Has their opening created anxiety in any GP women?
Is there any interaction between GP women and secure unit women? (Is this good, bad, wanted, unwanted?)

What problems have occurred as a result of the opening?

What good things have resulted from the opening?

What do you think has been the greatest source of stress (biggest negative) related to its opening?

What do you think has been the greatest advantage (biggest positive) related to its opening?

Has the atmosphere here at the institution changed since the secure unit has opened?
Have things become stricter? (e.g., Is there more acting out?)
Have things become more lenient?
Have the relationships between staff and women changed? (e.g., Are staff more cautious, stressed, etc?)

Would you say that you/everyone has adjusted to the change?
Are things back to normal (the way they were prior to the opening) now since it has opened?

Have any changes occurred since the secure unit opened that you did not expect or were not prepared for?

Now I'd just like to provide everyone with the opportunity to talk to us about anything that you would like to share that we haven't discussed. This is your opportunity to provide the GP women with a voice that will be heard so feel free to talk about any feelings, concerns, suggestions, or comments that you might have, or any that you have heard expressed other women.

Appendix C: Staff and Inmate Surveys

For concision, the lines provided in the original surveys for respondents to record their comments were excluded from this appendix.

Pre-Test Open Ended Survey - Staff Version

We want to know what you think. The following questions attempt to address any feelings or issues you may have in regards to the opening of the secure unit at your facility. Any suggestions or comments would be greatly appreciated (feel free to use back of page if necessary).

Do you view the opening of the secure unit as a positive or negative change? Explain.

Do you think that the opening of the secure unit will have an effect on your work environment? If so, how?

Do you think that the opening of the secure unit will have an effect on inmates in the general population?

Do you think that the opening of the secure units at the women's institutions will provide an effective way to manage maximum-security women? Why or why not?

What problems and/or advantages do you anticipate as a result of the opening of the secure unit?

Do you have any other comments or suggestions regarding the opening of the secure unit, and how you feel that yourself, fellow co-workers, or general population inmates will be effected?

Pre-Test Open-Ended Survey - Inmate Version

We want to know what you think. The following questions attempt to address any feelings or issues you may have in regards to the opening of the secure unit. Any suggestions or comments would be greatly appreciated (feel free to use back of page if necessary).

Do you view the opening of the secure unit as a positive or negative change? Explain.

How do you think that the opening of the secure units will impact you and other women you live with?

What problems and/or advantages do you anticipate as a result of the opening of the secure unit?

Do you have any other comments or suggestions regarding the opening of the secure unit, and how you feel that yourself, or others will be affected?

Post-Test Open Ended Survey - Staff Version

We want to know what you think. The following questions attempt to address any feelings or issues you may have in regards to the opening of the secure unit at your facility. Any suggestions or comments would be greatly appreciated (feel free to use back of page if necessary).

Now that it has opened, do you view the opening of the secure unit as a positive or negative change? Explain.

Do you think that the opening of the secure unit has had an effect on your work environment? If so, how?

Do you think that the opening of the secure unit has had an effect on inmates in the general population?

Do you think that the opening of the secure units at the women's institutions has provided an effective way to manage maximum-security women? Why or why not?

What problems and/or advantages have resulted from the opening of the secure unit?

Do you have any other comments or suggestions regarding the opening of the secure unit, and how you feel that yourself, fellow co-workers, or general population inmates have been effected?

Post-Test Open Ended Survey - Inmate Version

We want to know what you think. The following questions attempt to address any feelings or issues you may have in regards to the opening of the secure unit. Any suggestions or comments would be greatly appreciated (feel free to use back of page if necessary).

Now that it has opened, do you view the opening of the secure unit as a positive or negative change? Explain.

How do you think that the opening of the secure units has impacted you and other women you live with?

What problems and/or advantages have occurred as a result of the opening of the secure unit?

Do you have any other comments or suggestions regarding the opening of the secure unit, and how you feel that yourself, or others have been affected?

Appendix D: Relationships between Social Desirability and Assessment Measures

Table D1. Correlations between Social Desirability and Staff Assessment Measures

	Pre-test (n = 6 to 32)		Post-test (n = 6 to 32)
Measure	SDE	IM	SDE	IM
Job Satisfaction Survey
Overall Job Satisfaction	-.03	.21	.33*	.22
Intrinsic Job Satisfaction	.14	.21	.28	.32*
Extrinsic Job Satisfaction	.06	.22	.24	.26
Job Itself Intrinsic Satisfaction	.11	.13	.24	.24
Employee Relations Satisfaction	.13	.32*	.19	.35*
Total Score	.10	.24	.31	.34*
Maslach Burnout Inventory
Emotional Exhaustion	-.25	-.12	-.40**	-.47**
Depersonalization	-.13	-.25	-.46**	-.63****
Personal Accomplishment	.17	.07	.38*	.35*
Occupational Safety and Health Survey
Total Score	.02	.23	.23	.28
Stress and Coping Among Correctional Officers
Role Ambiguity	-.21	-.30*	-.44**	-.52***
Role Conflict	-.23	-.32*	-.21	-.51***
Career Development	-.26	-.31*	-.46**	-.41**
Safety	.16	-.09	-.18	-.12
Supervisory Support and Stress Scales
Supervisory Support	.10	.21	-.06	.35*
Peer Support	.08	.08	.25	.20
Work Stress	-.23	-.25	-.31*	-.23

Table D2. Correlations between Social Desirability and Offender Assessment Measures

	Pre-test (n = 27 to 50)		Post-test (n = 40 to 52)
Measure	SDE	IM	SDE	IM
Correctional Environment Status Inventory
Staff Cohesion a	.21	.23	.15	-.21
Staff Involvement	.05	.24	-.08	.04
Staff Treatment Focus	-.03	.16	-.08	.01
Clarity and Organization	.08	.23	-.05	.02
Offender Relationships	.25	-.05	-.11	-.20
Offender Treatment Orientation	.06	.10	-.02	-.04
Inventory of Interpersonal Problems
Domineering / Controlling	-.19	-.28	-.08	-.22
Vindictive / Self-Centred	-.09	-.15	.01	-.08
Cold / Distant	.00	-.10	.03	-.15
Socially Disinhibited	-.18	.13	-.04	-.07
Non-Assertive	.02	-.27	-.23	-.15
Overly Accommodating	-.37**	.04	-.30*	-.11
Self-Sacrificing	-.39**	.03	-.30*	-.13
Intrusive / Needy	-.37**	-.28	-.24	-.07
Total Score	-.31*	-.08	-.18	-.15
Self-Control Schedule
Total Score	.28	.26	.17	.39*

---