Research Reports

Attitudes of Federal Correctional Officers Towards Offenders

Michel Larivière
David Robinson
Research Division
Correctional Service of Canada
February 1996

Table of Contents

Attitudes of Federal Correctional Officers Towards Offenders

Executive Summary

Introduction and review of the literature

Method

Measuring Correctional Orientations/Correctional Attitudes

Results

Conclusion

References

Executive Summary

Correctional officers (COs) are the largest category of front-line correctional employees and they maintain a high frequency of contact with inmates. It has been argued that the development of good relationships between COs and inmates can be an important avenue for influencing offenders in pro-social ways. Thus, correctional officers are viewed as key agents in the process of rehabilitation. The quality of relationships between COs and inmates is also likely to promote more stable institutional environments whereby offenders are easier to manage on a day-to-day basis. When CO attitudes are positive and supportive of offender rehabilitation, the CO will be better equipped to promote favorable correctional outcomes among offenders.

This study focused on three dimensions of CO attitudes toward offenders or "correctional orientations": empathy, punitiveness, and support for rehabilitation. Empathy referred to the COs willingness to understand the affective states of inmates (e.g., trust, compassion, advocacy for offenders). The punitiveness dimension is concerned with the degree to which COs endorse retribution and punishment of offenders as important correctional goals. Support for rehabilitation simply referred to belief in the efficacy of rehabilitation and agreement that programming is a valuable function within the correctional domain.

The study was based on data from the CSC Staff Survey conducted in 1994. A total of 1,970 COs participated in the national survey, which represented an overall response rate of 48% for this occupational group. The CO respondents were 83.1% male, had a mean age of 39.6 years, and a mean of 11.4 years of work experience in CSC .

The 1994 survey results indicated that 23.3% of COs exhibited empathic views of offenders, 76.2% held punitive views of corrections, and 53.6% supported rehabilitation. The data also showed that COs varied significantly from other occupational groups on their attitudes towards offenders. COs were less empathic, more punitive, and less supportive of rehabilitation than each of eight other occupational groups within the Service (Case Management Officers; Health Care/Psychology; Miscellaneous Administrative Officers; Administrative Support; Labour/Technical/CORCAN; Correctional Programs; Correctional Supervisors and Managers; Cost/Activity Centre Managers).

However, there was considerable variability in correctional orientations when the CO group was divided into separate sub-groups based on demographic and work setting variables.

Regional variations suggested that COs employed in the Pacific region held the most positive attitudes towards offenders, while CO s from Quebec were the least positive. Lower security level was associated with more positive attitudes toward offenders. COs from maximum and medium security levels were generally comparable on each of the dimensions studied. COs employed at the CO-I level were somewhat more negative in their correctional orientations than officers employed at the CO-II level.

Analyses by gender suggested that male and female officers are comparable on empathy and punitiveness. However, female officers were more supportive of rehabilitation than male officers. Older COs (e.g., 50 years and over) tended to express more positive attitudes toward their charges than younger COs (under 50 years). There was a similar trend when years of experience was examined. However, CO s at the earliest stages of their careers (i.e., less than 1 year of experience) were considerably more positive in their attitudes toward offenders than their more experienced counter-parts.

There were interesting correlations between attitudes towards offenders and a number of measures of occupational adjustment. For example, CO s who were more empathic, less punitive, and more supportive of rehabilitation were also more committed to CSC, happier in their jobs, and reported significantly less job stress. COs with more positive correctional orientations also expressed greater support for unit management, were more likely to endorse CSC objectives, felt more empowered on the job, experienced a greater sense of physical security on the job, and were more satisfied with their career development and most recent performance evaluations. In short, COs who expressed positive views about offenders were generally much happier in their careers than those who possessed negative views about offenders.

Attitudes of Federal Correctional Officers Towards Offenders

Introduction and review of the literature

There is general consensus that Correctional Officers (COs) play a major role in creating and maintaining the human environment within institutions (Hepburn & Knepper, 1993). Research about this occupational group often begins with the assertion that the CO is the most important person in an inmate's world (Guenther and Guenther, 1974). The nature of the CO-offender relationship is recognized as a potential intervention source for the development of more constructive and effective behaviour patterns. Because COs play such a crucial part in carrying out correctional objectives, academics and policy makers alike have suggested an expansion or evolution of the CO's role to include elements of treatment and rehabilitation. In addition to the custodial function of COs, some have proposed a "human services" function (Toch, 1978; Lombardo, 1985; Johnson 1987; Hepburn, 1993).

In the human services model, COs assist inmates with institutional problems and act as referral agents or advocates in a variety of situations (Johnson, 1987). In doing so, some have argued that front line correctional staff may obtain increased levels of job satisfaction and decreased levels of job stress. Recent findings have, in fact, demonstrated that job satisfaction is associated with the more intrinsic aspects of a job (Lombardo, 1982; Gruenberg, 1980; Kalleberg 1977; Mottaz, 1987). Intrinsic aspects of employment would include the freedom to plan one's work activities, the chance to use one's skills and talents, and the likelihood of personal growth on the job. Extrinsic aspects, on the other hand, include elements such as the amount of pay, job security, fringe benefits, and opportunities for advancement.

It has been documented that COs who exhibited human service orientations enjoyed greater job satisfaction, increased intrinsic rewards, and less role ambiguity (Hepburn & Knepper, 1993). According to Hepburn (1985), COs themselves recognize that their overall effectiveness is based more on their interpersonal skills than on their use of rewards or punishment. If this is the case, then promoting a human services approach may hold the promise of more effective, humane, and rewarding correctional environments. "Given the appropriate training and support systems, each officer would be not only a rule enforcer, but also a lay counselor, a dispute mediator, an administrative ombudsperson, and a treatment aide" (Hepburn & Knepper, 1993; p. 315).

CSC's Mission shares much in common with the human services model. Both view positive human relationships as the basis for effective intervention with offenders. The Mission focuses on the dignity and potential of individuals and reflects the evolution of corrections in modern society; from punishment and custody, to assistance and reintegration.

The 12 week CO training program for new recruits begins with an introduction to CSC's Mission Statement, Core Values, Guiding Principles, and Strategic Objectives. Among other things, recruits examine effective interpersonal skills, cross cultural differences, family violence, and suicide prevention. Although COs are expected to fulfill the dual mandate of security and assistance / intervention at the end of their training, the emphasis on how to encourage offenders to become law-abiding citizens is clear from the outset.

This approach appears to be consistent with national opinion polls indicating that the public views rehabilitation as a legitimate and important goal of the correctional system (Cullen, 1989; Adams, 1990). Moreover, polls show that support for rehabilitation surpasses support for more punitive objectives such as deterrence and/or retribution (Cullen, Clark, and Wozniak, 1985; Cullen, Golden, and Cullen, 1983; Thomson and Ragona, 1987). Considering all of the above, it becomes important to address the issue of correctional orientations held by front-line staff. Moreover, there is a need to identify the predictors or correlates of positive and negative attitudes towards offenders.

In part the research literature suggests that a substantial proportion of COs hold unfavorable attitudes toward inmates. Eighty-six percent of Jacobs' (1978) sample of COs believed that "prisoners try to take advantage of you whenever they can". One quarter of the sample also believed that punishment was the main reason for putting offenders in prison. Freeman and Johnson (1982) studied CO attitudes towards inmate health: 20% agreed that inmates faked illness while 47% were uncertain. In addition, 25% stated that they were uncertain whether inmates who needed resuscitation would receive it from COs trained to provide emergency medical care. Nacci and Kane's (1984) sample of COs were inclined to believe that inmates themselves are to blame when they are victims of sexual assaults. They also overestimated the prevalence of homosexuality in institutions. Friel (1984) submitted that COs do not understand what inmates believe to be important.

Chang and Zastrow (1976) argued that the typical view of inmates held by COs is so poor that it may render any rehabilitative efforts futile. They also suggested that COs are implicitly a factor in perpetuating the criminal behaviour for which individuals are incarcerated. These authors based the latter argument on "looking-glass self" postulates which assume that people develop their self-concepts (who and what they are) in terms of how others relate to them. According to this theory, individuals labeled as untrustworthy or untreatable will likely fulfill the expectation. In a Canadian sample, Weekes, Pelletier and Beaudette (1995) found that 52.4% of officers thought that offenders could be rehabilitated and only 20.7% believed that sex offenders could be rehabilitated.

However, not all studies return pessimistic assessments of CO attitudes. The literature also indicates that many COs are supportive of rehabilitative and human service goals and do not embrace exclusively punitive or custodial views. Toch & Klofas' (1982) survey revealed that 75% of respondents agreed with the statement: "It's important for a CO to have compassion". Sixty-one percent agreed that "the way you get respect from inmates is to take an interest in them". Other research by Teske and Williamson (1979) demonstrated that most COs in their sample were highly positive in their attitudes toward selected treatment programs. Wheeler (1961) and again Kauffmann (1981) concluded that COs are more sympathetic than inmates and other COs think they are. In fact, there appears to be a disjunction between the typical CO's public presentation and personal beliefs (Cullen, Lutze, Link, Wolfe; 1989). Although they may embrace a rehabilitative/empathic orientation in private, they are inclined to present a punitive demeanor in public. Because the COs in Cullen's study overestimated the degree to which colleagues held punitive orientations (pluralistic ignorance), the researchers predicted that COs would demonstrate negative attitudes while in the company of co-workers, perhaps fearing sanctions should they disclose otherwise. These dynamics would in turn promote a punitive/custodial sub-culture despite individual convictions to the contrary.

As the studies cited above indicate, research on COs is not a neglected area in the criminal justice and social science literature (Philliber, 1987). Many researchers have attempted to identify the determinants of correctional attitudes and have examined a variety of factors: gender, prison settings, age, length of service.

Gender

Jurik and Halemba (1984) reported that 55% of female officers in their sample indicated that their primary reason for assuming their job was an interest in human service work or in inmate rehabilitation. In contrast, only 20% of male officer respondents gave this as their primary reason for employment. As a group, male officers were more likely than female officers to list other, more extrinsic reasons for employment (e.g. job security, fringe benefits, etc.). However, no significant relationship was discovered between gender and attitudes toward inmates. For instance, women did not appear more sensitive to the needs of inmates than men.

Crouch and Alpert (1982) found that both male and female guards entered correctional work with similar scores on measures of punitiveness and aggressiveness. However, males became significantly more punitive and aggressive, whereas females became less so after only six months of employment. This striking divergence in scores was likely due to the specific job demands required of male and female officers in this sample. Jurik (1985) on the other hand, found no attitudinal differences in her sample of experienced male and female COs.

In a study of new CSC recruits, Plecas and Maxim (1987) revealed significant gender differences with expected levels of satisfaction on the job. On seven of these items (co-workers, supervisors, opportunity to make a contribution, physical setting, the general work atmosphere, status of the employer, and stress) women officers expected to be less satisfied than their male counterparts. Only on one item, the "working with inmates" dimension, did women recruits express higher expected levels of satisfaction than male recruits. In this same study, gender differences appeared along other dimensions. For example, women recruits believed that more inmates were mentally ill than did the men recruits. Also, women were significantly more likely to believe that a greater number of inmates could benefit from work programs; that fewer needed to be locked up; that fewer had "had a fair chance in life"; and that more were "basically decent people". Both attitudes toward general correctional policy as well as CSC were examined by these authors. With regards to correctional policy, female recruits tended to be less punitive than males. For instance, women were less likely to believe that harsh punishment deters or that physical punishment is necessary in dealing with inmates. Also, the women were more inclined to support inmate rights and to believe that inmates should have some say in running institutions.

In a more recent study of CO's employed by CSC, Robinson, Porporino and Simourd (1993) found that females were somewhat more supportive of rehabilitation than males. However, after controlling for a variety of demographic and career orientation variables, the relationship between gender and attitudes toward rehabilitation disappeared. Having a positive attitude towards correctional work in general, exhibiting a desire to work with people, and showing a desire for career development and growth were the most important predictors of positive attitudes towards rehabilitation in that study.

Setting

The correctional setting itself has been studied as a possible moderator of staff attitudes toward inmates. Jurik (1985) found more favorable attitudes toward inmates in minimum security prisons. Street (1965), found more positive attitudes and higher levels of primary relationships among workers in treatment settings. Farmer (1977) found cynicism to be more prevalent in treatment than in custodial settings whereas Brown et al. (1971) argued that the emphasis of the prison (treatment or custody) will interact with the duties of workers to affect attitudes toward inmates

Age and Experience

While data exploring the relationship between age and experience on correctional orientations has not always yielded clear results, it appears that attitudes towards inmates improve with age and time on the job. Crouch and Alpert (1982) report no significant relationship between age and punitiveness but Jurik (1985) reports a significant positive relationship between age and favorable attitudes toward inmates. Similarly, Toch and Klofas (1982) report a negative relationship between age and custody orientation.

These results are similar to other studies in the literature which have examined workers in various employment settings. It has been shown that older workers are generally more adjusted to their positions in life and tend to be more satisfied. It is possible that experience allows an individual to make a more balanced assessment of what one can possibly expect from work (Plecas and Maxim, 1987).

In the Plecas and Maxim (1987) study of CSC recruits, it was found that older recruits had slightly more positive attitudes than their younger colleagues. They were more likely to believe that inmates are "basically decent people" and would "help a friend in need". Also, the older the recruit, the less likely they were to believe that inmates lacked any capacity for rehabilitation and the less likely they were to believe that inmates "really needed to be locked up". That is not to say that these recruits as a whole were overly positive in their attitudes towards offenders. For example, recruits believed that 23% of offenders were mentally ill, that 37% lacked any capacity for rehabilitation, that only 31% truly appreciate the help offered by staff, and that only 35% are really interested in leading law-abiding lives. The authors argued that if these were valid indicators of recruits' perceptions, then it would seem reasonable to conclude that as a group recruits were not expecting to have a major impact on the lives of the inmate population.

In terms of experience, it has been reported that length of time as a CO is negatively related to favorable attitudes toward inmates (Jurik, 1985) and to belief in the rehabilitative potential of inmates (Shamir and Drory, 1981). However, these studies did not examine the entire career span of subjects in their sample. In studies examining the entire career of individuals, the relationship between tenure and attitudes does not seem to follow a typical linear pattern. One model suggests that newly recruited officers (with their idealism), and senior officers (having "mellowed" somewhat), will display the fewest negative attitudes toward inmates. Mid-career officers, on the other hand (for whom idealism is gone and relief of retirement is distant) would tend to hold the most negative perceptions of inmates (Philliber, 1987).

There is evidence in the literature suggesting that attitudes change rather quickly in a CO's career. At nine months after induction training, CSC COs had become more punishment oriented and less supportive of inmates' rights. There was increased agreement with items such as "physical punishment is necessary in dealing with criminals", "harsh punishment will deter people from committing crimes", and "federal prisoners do not deserve any civil rights". During a second follow-up study 9 months later, the researchers found that attitudes had stabilized at these more negative levels (Plecas and Maxim, 1987).

In the present study, we examine variations in attitudes towards offenders as a function of each of the variables discussed above. CO attitudes were examined using three dimensions: empathy toward offenders, support for rehabilitation, and level of punitiveness. We extended our analyses to other variables unique to our setting including CO level and region. In addition, we examined the association between attitudes towards offenders and other, more broader work variables including job satisfaction, stress, and commitment to the organization.

METHODOLOGY

The current study is based on data derived from the CSC All-Staff Survey conducted in March 1994. With an overall participation rate of 61%, the All-Staff Survey provided a rich database to explore both the demographic and attitudinal variables listed above. A total of 13 themes were covered in the survey's questionnaire ranging from unit and case management, to job stress, quality of the work environment, information technology, and attitudes towards inmates. The survey also included basic demographic information.

Procedure

The All-Staff Survey questionnaire consisted of approximately 200 items but few staff members were required to complete all sections. Questionnaires for the survey were sent to all CSC facilities including Regional and National Headquarters. Worksite contacts were identified and became responsible for coordinating the administration of the survey, distributing questionnaires, monitoring participation, and taking action to encourage greater response rates where necessary. Completed questionnaires were sealed in an envelope, returned to site coordinators, then forwarded to an independent consulting firm for analysis.

Current Sample

Sixty one percent of all eligible respondents completed the questionnaire Response rates for each occupational group are presented in Table 1. As illustrated in the table, COs responded at a lower rate (48%) than the overall average of 61%. The best response rates occurred in the Miscellaneous/ Administrative and Cost/Activity Centre Managers (CACM) category at 83% followed by Case Management Officers at 70%.

Table 1. Response rates by occupational group

Job Type	Respondents (n)	Total Population	Response Rate
Correctional Officers (CO)	1,970	4,080	48%
Case Management Officers (CMO)	623	888	70%
Health Care/Psychology (HCP)	331	587	56%
Miscellaneous/Administrative Officers (MAO) and Cost/Activity Centre Managers (CACM)	1,015	1,219	83%
Administrative Support (AS)	1,017	1,685	60%
Labour, Technical, CORCAN (LTC)	707	1,226	58%
Correctional Programs (CP)	198	481	41%
Correctional Supervisors and Managers (CSM)	234	634	37%
Not Reported	515	NA	N/A
Total	6,610	10,800	61%

Demographic variables

Table 2 presents demographic information for all respondents and for COs separately. While age and language are similar in both groups, we noted that males accounted for 83% of the CO group as opposed to 62% of the total sample.

Table 2. Demographics - CSC All-Staff Survey respondents

Variable	All Respondents	Correctional Officers
Sample Size	N= 6,610	N= 1,970
Gender: Male	62.5%	83.1%
Gender: Female	37.5%	16.9%
Language: English	63.9%	65.1
Language: French	36.1%	34.9%
Mean age	40.2 years S.D.=10.2	39.6 years S.D.=8.7
Mean experience with CSC	11.8 years S.D.=10.1	11.4 years S.D.=7.2
Proportion of respondents by Region: Pacific	15.4%	18.2%
Proportion of respondents by Region: Prairies	19.8%	24.8%
Proportion of respondents by Region: Ontario	18.6%	18.1%
Proportion of respondents by Region: Quebec	29.1%	29.3%
Proportion of respondents by Region: Atlantic	11.9%	9.5%
Visible minority/ethnic group	7.4%	9.6%
Aboriginal	2.5%	4.0%
Disability	3.4%	4.0%

Measuring Correctional Orientations/Correctional Attitudes

Composite Scale Development

Fourteen questionnaire items were used to form three scales to assess different dimensions of attitudes towards offenders: empathy, punitiveness, and support for rehabilitation. The empathy scale examined the respondent's willingness to understand the affective state of inmates. It also tapped issues of trust, compassion, and advocacy for offenders. The punitive composite described the degree to which a respondent favoured punishment, deterrence, and retribution for criminal behaviour. The third composite, support for rehabilitation, quantified an employee's support for treatment/intervention and for correctional programming in general.

Participants recorded their level of agreement/disagreement with each item statement based on a 7-point rating scale ranging from Strongly Disagree to Strongly Agree. For each composite scale, the respective items were summed and divided by the total number of items (e.g. 4 items in the case of punitiveness) in order to provide an average score. Respondents with average scores within the "agree to strongly agree" range were classified as endorsing that attitude. Individuals were therefore classified as either "empathic/non-empathic", "punitive/non-punitive", and as either "supportive/non-supportive" of rehabilitation. Table 3 lists individual questionnaire items making up each composite scale and shows the percentage of responses for each of the items used in the scales.

Table 4. Response distributions for composite scales

EMPATHY (=0.80)	Strongly Disagree	Disagree	Partially Disagree	Undecided	Partially Agree	Agree	Strongly Agree
Q1. Staff should work hard to earn trust from offenders.	11.5 %	16.8 %	11.6 %	5.1%	26.1%	24.4%	4.4%
Q2. It's important for staff to have compassion for offenders.	23.2%	25.6%	14.8%	4.6%	21%	8.3%	1.6%
Q3. The way to get respect from offenders is to take an interest in them.	12.4%	18.9%	13.8%	4.6%	28.8%	18.4%	3.2%
Q4. Sometimes staff should advocate for an offender.	14.0%	15.9%	11.0%	11.7%	31.3%	14.3%	1.8%
Q11. You can't ever completely trust an offender. (reversed)	0.6%	4.5%	7.0%	2.2%	18.1%	30.6%	36.9%
Q12. A good principle is to not get "close" to offenders. (reversed)	1.6%	5.9%	8.7%	2.1%	19.7%	31.4%	30.7%
Q13. If staff are lenient with offenders, they will take advantage of them. (reversed)	1.1%	4.3%	9.4%	1.7%	29.4%	27.8%	26.5%

Table 4 (continued)

PUNITIVENESS (=0.64)	Strongly Disagree	Disagree	Partially Disagree	Undecided	Partially Agree	Agree	Strongly Agree
Q7. There would be much less crime if prisons were more uncomfortable	5.8%	11.3%	5.6%	6.1%	18.4%	19.1%	33.7%
Q8. Improving prisons for inmates makes them worse for staff	4.3%	15.9%	12.6%	6.4%	18.6%	18.9%	23.5%
Q9. A military regime is the best way of running a prison.	6.0%	14.2%	12.7%	7.5%	27.4%	13.5%	18.9%
Q16. We should stop viewing offenders as victims of society.	2.8%	2.8%	2.7%	1.5%	10.7%	28.8%	50.7%

Table 4 (continued)

SUPPORT FOR REHABILITATION (=0.72)	Strongly Disagree	Disagree	Partially Disagree	Undecided	Partially Agree	Agree	Strongly Agree
Q10. Rehabilitation programs are a waste of time and money. (reversed)	5.9%	17.8%	17.2%	5.7%	25.8%	11.6%	16.1%
Q15. Rehabilitating an offender is just as important as making an offender pay for his or her crime.	4.9%	7.6%	6.8%	6.1%	26.0%	36.0%	12.8%
Q17. I would support expanding the rehabilitation programs which are presently being offered in our institutions.	15.7%	20.6%	8.6%	10.8%	19.2%	18.2%	7.0%

RESULTS

Attitudes of COs in comparison to other operational groups

Staff were generally positive in their attitudes toward offenders, and typically expressed views which were consistent with CSC's Mission (Figure 1). For example, 62% of Case Management Officers were considered empathic toward offenders (e.g., desire to make social contact and earn their trust) while less than 34% of those in the Correctional Programs group were considered punitive (e.g., emphasis on deterrence and/or retribution). Also among operational staff, support for rehabilitation reached above 90% among CMOs and over 70% among Correctional Supervisors. On the other hand, only 23.3% of COs were considered empathic according to our measures (p< .001). Moreover, COs were the most punitive (76.2%; p< .001) and the least supportive of rehabilitation (53.6%; p< .001) than any other occupational group. As Figure 1 shows, attitudes of COs differed significantly from all other occupational groups within CSC. The LTC category (Labour, Technical, and CORCAN) was the group most similar to COs on the attitudinal variables measured.

Figure 1. Correctional Orientation by Occupational Group

CO

Correctional Officers

LTC

Labour, Technical, CORCAN

CMO

Case Management Officers

CP

Correctional Programs

HCP

Health Care/Psychology

CSM

Correctional Supervisors/Managers

Regional differences

As can be seen in Figure 2, COs are generally punitive, non-empathic, and only moderately supportive of rehabilitation regardless of the region in which they work. A notable finding, however, concerned COs in the Québec region who held the most negative attitudes toward inmates; an observation which was consistent across all three dependent variables. Québec region COs were significantly less empathic (19%), more punitive (86%), and less likely to support rehabilitation (49.1%) when compared to colleagues in other regions of the country (p <.001). COs in the Pacific region were the most empathic (31%) and the least punitive (64.6%), while support for rehabilitation was greatest in the Atlantic (65%) and Ontario regions (61.2%).

Figure 2. Regional Differences in Correctional Orientation

Gender differences

In terms of empathy, female COs (22.8%) were not significantly different from male COs (23.8%). The same held true when we considered level of punitiveness: 72.2% of females were considered punitive compared with 76.7% of males. However, as can be seen in figure 3, females were significantly more likely to support rehabilitating offenders (64.7%) than males (51.9%; p< .001).

CO-Level

CO-2s were significantly more empathic (27.7%) than CO-1s (17.7%; p< .001) slightly less punitive (74.2% vs. 78.8%; p< .05), and more supportive of rehabilitation (60.5% vs. 44.6%; p< .001). Figure 4 displays the relationship between CO level and correctional orientation.

Figure 3

Figure 4

Age and Experience

Correctional attitudes were found to vary when different age groups were compared. As shown in figure 5, there was a clear trend toward more positive attitudes with increasing age. For example, only 8% of the 20-29 age group had average scores within the empathic range. In contrast, 42.1% of the 50+ age group (p< .001) were empathic according to their responses on the questionnaire. Similarly, 81.3% of the 20-29 year olds were classified as punitive as opposed to only 64% of the most senior group (p< .001). Belief in the rehabilitative potential of inmates increased with age, albeit to a lesser extent: 49.4% of the 30-39 year group supported rehabilitating offenders compared to 59.1% for the 50+ group (p< .05).

It should be kept in mind however, that the data presented in figure 5 may be demonstrating cohort differences (e.g., differences in training and organizational policy at the time of career entry) in addition to the effect of age per se. A longitudinal study would be required to sort out these effects and allow a more accurate discussion of age related factors. Nonetheless, we can statethat empathy, and support for rehabilitation are more prevalent among CSC's most senior correctional staff.

Figure 5

We observed striking attitudinal differences amongst those recently recruited as CO's, however. Figure 6 highlights these differences by comparing those with less than 1 year of experience and those with 1 and 2 years of experience. Results from the most senior group of workers are also included as a basis for comparison.

Figure 6

Thirty-one percent of COs with less than 1 year on the job were identified as empathic compared to 17.3% of those with 1 to 2 years of work with CSC (p< .05). There were similar attitudinal differences with regards to punitiveness: 53.1% of the newest recruits were punitive as opposed to 68% of those in the 1-2 year cohort (p< .005). Likewise, support for rehabilitation was significantly stronger at 78.1% among new recruits and 59.5% in the 1-2 year group (p< .05). We compared these results with CSC's most experienced workers (25 years and greater): 74.1% of these COs showed support for rehabilitation, 57% were empathic, and 52.3% were considered punitive. This replicates a pattern seen elsewhere in the literature whereby the most senior and the most junior workers have the most in common in terms of attitude. As was the case with age however, we cannot conclude from this type of study that experience is the cause of attitudinal change. The nature of the groups themselves may explain at least part of these results.

Correctional Orientation by Security Level

Correctional orientations also varied according to security level. As can be seen in Figure 7, COs operating at Minimum Security institutions had the most positive correctional attitudes. Minimum Security COs were the most empathic (39.6% p< .001) and the least punitive (70.3% p< .05). In addition, Minimum security COs were appreciably more supportive of efforts to rehabilitate offenders (71.6%) when compared with those in Maximum (51.2%) and Medium (51.8%) security levels (p< .001). In fact, support for rehabilitation amongst these COs compared favourably with those in most other occupational categories.

Figure 7

Commitment to CSC, Job Stress, and Job Satisfaction

We also looked at how attitudes towards inmates were related to other perceptions about correctional work and the organization in general. In Figure 8 we examine the impact of being empathic, non-punitive and supportive of rehabilitation on the alternative indices of occupational and organizational adjustment. As can be seen in the first chart, empathic COs were significantly more committed to their jobs (60.1% vs. 38.8%), more satisfied with the work (63% vs. 52.5%), and also experienced less job stress (48.1% vs. 59.5%). We observed similar results when punitiveness and support for rehabilitation were considered (p< .001). The link between attitudes towards offenders and commitment to CSC appears to be particularly strong.

Significant positive correlations were identified between the correctional orientation variables and still other aspects of work that were measured in the All-Staff Survey. Empathic, non-punitive, and pro-rehabilitation COs were more likely to support unit management (p< .0001), to endorse CSC objectives (p< .0001), and to feel empowered on the job (p< .0001). These individuals were also more satisfied with their most recent performance appraisal (p< .0001) and with career management (p< .0001). The data also revealed that workers with more positive attitudes also felt greater sense of personal security on the job (e.g., physical safety; p< .0001).

We also considered whether attitudes towards offenders were related to CO willingness to get involved in offender programming. Figure 9 shows a strong association between attitudes and the desire to participate in treatment/intervention. Eighty-six percent of empathic individuals showed an interest in participating in treatment compared to only 14% of those considered non-empathic. Not surprisingly, 80% of COs who were supportive of rehabilitation also wished to be more involved in treatment. Interestingly, over 40% of COs who had been categorized as punitive demonstrated a desire to participate in offender programming. Hence, holding punitive attitudes toward inmates does not necessarily preclude support of CSC's treatment efforts. On

Figure 8. Commitment to CSC, Job Stress, and Job Satisfaction by Correctional Orientation

Figure 9

the other hand, believing in rehabilitation and exhibiting empathy toward offenders appear to be crucial to developing a desire to become involved in programming.

CONCLUSION

Guenther and Guenther (1974; p.42) write that "the most durable and intensive relationships experienced by prisoners are with line COs". COs not only represent the largest group of Correctional staff but they are also among those who interact most with offenders. Attitudes will undoubtedly affect the quality of these interactions and although it has not been demonstrated empirically, we surmise that these interactions will in turn influence offender behavioural outcomes.

In many cases, the results of this study corroborated a number of findings in the literature. In general, COs showed little empathy for offenders and tended to support punitive objectives such as deterrence and/or retribution. Moreover, while about 50% of the COs surveyed were supportive of rehabilitation, they were proportionately less supportive of treatment than any other operational group. The present study does not explain why this is the case but points to those variables presently associated with more positive correctional orientations.

Although there was little association between gender and attitude, we found that women supported rehabilitation to a significantly greater extent than men. This finding has been reported elsewhere and continues to suggest that women select correctional work for different reasons than men; women seemingly preferring the intrinsic aspects of correctional work (e.g. intervention/interaction). This being the case, women may enjoy greater job satisfaction in treatment-oriented settings.

An interesting finding concerned the significant attitudinal differences between regions. As was shown, Pacific region COs were generally more empathic and less punitive than their colleagues in other regions. Notably, Québec region COs were the most punitive, the least empathic, and the least likely to support rehabilitative efforts than COs in any other region. Further research might explore possible reasons for these regional disparities.

Facility classification was also associated with correctional attitudes; employees in minimum security institutions were significantly more positive in their outlook than those in medium and maximum institutions. Because reduced security may promote more favorable attitudes, varying staff exposure to different security levels may have desirable results. Greater interaction with offenders who are preparing for release (or simply closer to release) may also be helpful.

A significant finding concerned the association between age and employment tenure and the way COs perceived offenders. COs appeared much more positive with increasing age and time on the job (a mellowing effect). An exception to this pattern, occurred in the early years of an employee's career. As was shown, new COs started work with fairly positive attitudes: approximately 80% of new COs favoured rehabilitation. This fell to 59% for CO's who had one year on the job. That type of finding has been reported empirically and anecdotally elsewhere (Wicks, 1980; Philliber, 1987;). Wicks (1980) writes:

Watching their initial entrance into the prison can be quite an experience. The hope on their faces, the positive anxiety of their motivated gait - at first, it's all there. Then, slowly and almost methodically, the smiles wane, the expectations atrophy, and the desires to perform in a positive fashion succumb to escapist fantasy and verbally acknowledged skepticism ...after six months to a year the period of hope and enthusiasm should almost be all over.

The All-Staff Survey confirmed this phenomenon but also demonstrated that elements of empathy and belief in human development reappear towards the end of a COs career. It may be beneficial to identify opportunities which enable senior staff to exercise leadership. Building correctional teams made of very senior and very inexperienced workers might be very rewarding for the former group while postponing disillusionment and skepticism in the latter group.

While CO-IIs are somewhat more positive than CO-Is, they remain considerably more negative than their colleagues in most other occupational categories. Given the case management duties associated with the CO-II position, one would have expected that the attitudinal profile of this group would more closely resemble the profile for the Case Management Officer group than the CO-I group. It may be that CO-IIs need more familiarity with the principles of effective rehabilitation and exposure to training that would facilitate a more empathic and less punitive stance toward offenders. Exposure to case management duties alone may simply not be enough to alter CO-II perspectives on offenders.

Some may argue that promoting more positive attitudes towards offenders is an enterprise most suited for induction training of COs. However, the current data provide compelling evidence to argue for ongoing training during a CO's career. The fact that COs enter the workplace with attitudes which are more consistent with the Mission than their more senior colleagues suggests that induction may be effective in promoting the appropriate attitudinal set for work in modern corrections.

However, the effects of training may be very short-lived without ongoing reinforcement and exposure to new correctional methods and procedures which focus on the rehabilitative side of the CO's role. The results of the study suggest that the first year of experience on the job may be crucial in determining how CO's perceive offenders. Some level of intervention (e.g. awareness training) may be appropriate during the first year of the CO's work experience to counteract influences that have a negative impact on attitudes. Awareness training may also be appropriate for COs who are in the middle of their correctional careers. It appears that this group, which represents the largest proportion of correctional staff, are least "in-step" with the attitudes the Service wishes to promote.

Unfortunately, we currently lack evidence about the relationship between staff attitudes and offender behaviour. In theory, it can be argued that the human context of corrections provides officers with ample occasions to positively influence the behaviour and outlook of offenders. Of course such potential influence could only be realized by COs who possessed the attitudes and values necessary to play a "helping" or "human service" role with their charges. There is recent evidence from a study of CORCAN shop instructors suggesting that staff attitudes may have important impacts on offenders (Gillis, Getkate, Robinson & Porporino, 1995). The CORCAN study showed that offenders who were exposed to shop instructors with dynamic leadership styles (e.g. more personal involvement with inmates, greater encouragement of inmate development), were more pro-social in their attitudes toward work. In addition, the study suggested that staff characteristics may also have an impact on concrete behavioural outcomes. Offenders exposed to more positive leadership styles were more punctual in their attendance at work. The results of the study suggest that there may be important payoffs from staff development investments that focus on changing how staff relate to offenders. By increasing CO empathy for offenders, for example, we may be able to facilitate interactions between staff and inmates that have beneficial correctional impacts.

Of additional significance was the relationship between attitudes and such dimensions as Job Stress, Job Satisfaction, and Commitment to CSC. Empathic, non-punitive, and pro-empathic workers were, without question, happier workers. Not only were they less stressed and more satisfied with their work, but they were typically more supportive of CSC's mission and objectives.

The link between agreement with organizational values and general satisfaction with work may be a positive indicator of the level of absorption of the CSC Mission by CSC staff. It may be that staff are becoming increasingly aware of the demanding challenges set by the Mission. Those staff who are not in agreement with the challenges may be less content in a correctional environment propelled by the current Mission principles. There may be less "fit" between their views and the ideals of the organization. In turn, the lack of "fit" may be causing them a degree of dissatisfaction in their correctional careers. If attitudes towards offenders and job satisfaction are related, there may be important implications for the recruitment process. In order to increase the fit between organizational and employee values, there may be a need to place a greater emphasis on attitudinal variables in the CO selection process.

REFERENCES

Adams, M. (1990). Canadian Attitudes Toward Crime and Justice. Forum, 2 (1),10-13.

Brown, B., Dupont, R., Kozel, N., & Spevacek, J. (1971). Staff Conceptions of Inmate Characteristics. Criminology, 9, 316-29.

Chang, D.H., Zastrow, CH. (1976). Inmates and Security Guards Perception of Themselves and Each Other. International Journal of Criminology and Penology, 1, 89-98.

Crouch, B.M., Alpert, P. (1982). Sex and Occupational Socialization Among Prison Guards: A Longitudinal Study. Criminal Justice and Behavior, 9, 159-76.

Cullen, F.T., Cullen, J.B., & Wozniak, J.F. (1988). Is rehabilitation dead? The myth of the punitive public. Journal of Criminal Justice, 16, 303-317.

Cullen, F., Lutze, F., Link, B., and Wolfe, N. (1989). The Correctional Orientation of Prison Guards: Do Officers Support Rehabilitation? Federal Probation, 53, 33-42.

Cullen, F., Clark, G., Wozniak, (1985). Explaining the Get Tough Movement: Can the Public be Blamed? Federal Probation, 49, 16-24.

Cullen, F., Golden, K., Cullen, J. (1983). Is Child Saving Dead? Attitudes Toward Juvenile Rehabilitation in Illinois. Journal of Criminal Justice, 11(1), 1-13.

Farmer, R.E., (1977). Cynicism: A Factor in Corrections Work. Journal of Criminal Justice (5), 237-46.

Freeman, R.W., Johnson, L. (1982). Health Related Knowledge, Attitudes and Practices of Correctional Officers. Journal of Prison and Jail Health 2, (2) 125-38.

Friel, J. (1984). Staff Perceptions of Prisoner Life Tasks and Prisoner Perceptions of Prisoner Life Tasks in a Canadian Penitentiary. Canadian Journal of Criminology, 26, (3), 355-57.

Gillis, C., Getkate, M., Robinson, D., & Porporino, F. J. (1995). Correctional work supervisor leadership and credibility: Their influence on offender work motivation. Forum on Corrections Research, 7, 15-17.

Guenther, A., & Guenther, M. (1974). Screws vs. Thugs. Society, 12, 42-50.

Haney, C. Banks, C., and Zimbardo, P. (1973). Interpersonal Dynamics in a Simulated Prison. International Journal of Criminology and Penology 1, 69-97.

Gruenberg, B. (1980). The Happy Worker: An Analysis of Educational and Occupational Differences in Determinants of Job Satisfaction. American Journal of Sociology, 86, 247-71.

Hepburn, J.R., Knepper, P.E. (1993). Correctional officers as human services workers: the effect on job satisfaction. Justice Quarterly, 10, (2).

Hepburn, J.R., (1985). The Exercise of Power in Coercive Organizations: A Study of Prison Guards. Criminology 23, 146-64.

Jacobs, J.B. (1978). What Prison Guards Think: A Profile of the Illinois Force. Crime and Delinquency 25, 185-96.

Johnson, R. (1987). Hard Time. Monterey: Brooks/Cole.

Jurik, N.C., Halemba, J. (1984). Gender, Working Conditions, and the Job Satisfaction of Women in a Non-Traditional Occupation: FemaleCorrectional Officers in Men's Prisons. Sociological Quarterly 25, 551-66.

Jurik, N.C. (1985). Individual and organizational determinants of correctional officer attitudes toward inmates. Criminology, 23, 523-539.

Kauffmann, K. (1981). Prison Officers Attitudes and Perceptions of Attitudes: A Case of Pluralistic Ignorance. Journal of Research in Crime and Delinquency, 18, July 1981, 272-94.

Kalleberg, A. (1977). Work Values and Job Rewards: A Theory of Job Satisfaction. American Sociological Review, 42, 124-43.

Lombardo, L. (1982). Stress, Change and Collective Violence in Prison. In R. Johnson and H. Toch (eds.), The Pains of Imprisonment, pp. 77-93. Beverly Hills: Sage.

Lombardo, L. (1985). Mental Health Work in Prisons and Jails: Inmate Adjustment and Indigenous Correctional Personnel. Criminal Justice and Behaviour 12, 17-28.

Mottaz, C. (1985). The Relative Importance of Intrinsic and Extrinsic Rewards as Determinants of Work Satisfaction. Sociological Quarterly, 26, 365-86.

Nacci, P. Kane, T. (1984). Sex and Sexual Aggression in Federal Prisons. Federal Probation 48, 46-53.

Philliber, Susan (1987). Thy Brother's Keeper: A review of the Literature on Correctional Officers. Justice Quarterly, 4 (1)

Plecas, D.B., Maxim, P.S. (1987). CSC Correctional Officer Development Study: Recruit Survey. Report prepared for National Headquarters, Correctional Service of Canada.

Robinson, D., Porporino, F.J., Simourd, L. (1993). The influence of career orientation on support for rehabilitation among correctional staff. The Prison Journal 73, (2), 162-177.

Shamir, B., &, Drory, A. (1981). Some Correlates of Prison Guards' Beliefs. Criminal Justice and Behavior, 8, (2), 233-49.

Street, D. (1965). The Inmate Group in Custodial and Treatment Settings. American Sociological Review, 30, 40-55.

Teske, R., & Williamson, H. (1979). Correctional Officers Attitudes Toward Selected Treatment Programs. Criminal Justice and Behavior, 6, (1), 59-66.

Thomson, D., & Ragona, A. (1987). Popular Moderation Versus Governmental Authoritarianism: An Interactionist View of Public Sentiments Toward Criminal Sanctions. Crime and Delinquency, 33, 337-357.

Toch, H. (1978). Is a Correctional Officer by any other Name a Screw?. Criminal Justice Review, 3, 19-35.

Toch, H., and Klofas, J. (1982). Alienation and Desire for Job Enrichment Among Correction Officers. Federal Probation, 46, 35-44.

Walsh, R.T. (1976). Researcher-Participant Collaboration in Corrections Research: A Case Study. Canadian Journal of Criminology and Corrections, 18, 387-400.

Weekes, J., Pelletier, G & Beaudette, D. (1995). Correctional officers: How do they perceive sex offenders? International Journal of Offender Therapy and Comparative Criminology, 39, 55-61.

Wheeler, S. (1961). Role Conflict in Corrections Communities. In D.Cressey (ed.), The Prison: Studies in Institutional Organization and Change. New York: Holt, Rinehart, and Winston.

Wicks, R.J. (1980). Guard! Society's Professional Prisons. Houston: Gulf Publications.