Exploring the Correlates of Perceived Job Dangerousness Among Correctional Staff at a Maximum Security Prison

Abstract

In literature on correctional staff, one poorly understood antecedent of job stress and other negative outcomes is perceived danger from the job. Survey results from 272 staff at a state-run Midwestern maximum security prison were analyzed with Ordinary Least Squares (OLS) Regression to determine the relationships between personal/work environment variables and perceptions of job danger. Analyses revealed the effects of the personal variables were conditional on staff position (custody vs. non-custody). Irrespective of position, two of seven work environment variables studied (less input into decision making and more daily contact with prisoners) were related to greater perceived risk of harm from the job. Also, greater organizational formalization was related to greater perceived risk among custodial staff. Perceived danger from the job is a real issue, and the current results indicate workplace factors play a role.

Keywords

job dangerousness correctional staff work environment prison workers

Working in institutional corrections is a stressful and potentially dangerous experience (Lahm, 2009; Taxman & Gordon, 2009). Researchers have devoted much attention to correctional job stress. Higgins, Tewksbury, and Denney (2012) found more than 80 articles since 1985 dealing with job stress among correctional staff. Harmful for both correctional staff and agencies, job stress can lead to health problems for staff, such as hypertension, heart disease, and depression (Cheek & Miller, 1983; Finn, 1998). It can also promote elevated rates of burnout, divorce, post-traumatic stress disorder, and substance abuse (Black, 2001; Cornelius, 2001; Finn, 1998; Keinan & Maslach-Pines, 2007), along with increased absenteeism, higher turnover intent and turnover, lower job satisfaction, and lower organizational commitment (Dowden & Tellier, 2004; Lambert, Edwards, Camp, & Saylor, 2005; Mitchell, Mackenzie, Styve, & Gover, 2000; Tewksbury & Higgins, 2006). In light of its detrimental consequences, research on job stress among custody (i.e., correctional) officers is commonplace. Still, not all possible antecedents have received equal attention. Specifically, perceived danger of the job (also known as perceived job dangerousness), which refers to a feeling of being at risk of injury while on the job, has not gotten the consideration it warrants (Cullen, Link, Wolfe, & Frank, 1985; Gordon, Proulx, & Grant, 2013; Lai, Wang, & Kellar, 2012).

Feeling at risk of injury at work, whether that risk is actual or perceived, can be a major life strain. Research often considers the role of perceived danger or safety among correctional officers in predicting job stress. Taken in total, the research indicates officers who feel less safe report higher levels of job stress (Armstrong & Griffin, 2004; Blevins, Cullen, Frank, Sundt, & Holmes, 2006; Dowden & Tellier, 2004; Hartley, Davila, Marquart, & Mullings, 2013; Huckabee, 1992; Wells, Minor, Angel, Matz, & Amato, 2009). The perceived threat of violence wears on staff (Higgins et al., 2012). Besides increased job stress, perceived job danger or risk of victimization has been linked to lower levels of job satisfaction, perceived organizational, and organizational commitment (Cullen et al., 1985; Dowden & Tellier, 2004; Griffin, 2006; Lambert & Hogan, 2010; Taxman & Gordon, 2009). Although recommendations to reduce perceived danger from the job have been made, little research exists on the possible correlates of this harmful workplace factor (Liebling, 2008). Without research examining the correlates of danger or safety, reducing perceived risks from the job will be difficult and haphazard.

Working in corrections carries a risk of physical injury (Lai et al., 2012). The perceived risk of danger or risk of victimization varies among correctional staff, even at the same facility (Gordon, Moriarty, & Grant, 2003; Gordon et al., 2013; Lambert & Hogan, 2010). Although we have some indication of how danger, actual or perceived, influences other predictors such as stress, perceived organizational fairness, and commitment, the current literature has not adequately explored the correlates of perceived danger. There is, however, limited information regarding the correlates of the similar, but distinct, concept of perceived risk of victimization by inmates among correctional staff (see Gordon & Baker, 2015; Gordon et al., 2013; Lai et al., 2012; Stichman & Gordon, 2015; Taxman & Gordon, 2009); hence, the present study was undertaken to explore this gap in knowledge.

The current study advances the correctional literature in two distinct ways. First, it examines the correlates of perceived job danger, and, second, it considers differences between custodial and non-custodial staff. Specifically, this study explores the correlates of perceived danger of the job among correctional personnel employed at a maximum security facility. It considers workplace variables (i.e., input into decision making, formalization, integration, job feedback, instrumental communication, supervision consideration, and amount of daily inmate contact) and personal characteristics (supervisory status, gender, age, position, tenure, educational level, and race). In addition, this exploratory study examined the correlates of perceived danger among custodial staff and non-custodial staff. To date, non-custodial staff perceptions of danger or related concepts have not been examined among U.S. correctional personnel. Nevertheless, non-custodial employees comprise a sizable portion of all staff in most correctional facilities. Compared with custody staff, such staff generally have varying amounts and types of contact with inmates, and the work environment context may also differ (e.g., they may have more input into decisions).

Literature Review

Maintaining a safe and secure environment is a central component for all staff working in prisons. Yet prison environments are not inherently conducive to personal safety and security, and this can shape staff perceptions of danger. At a broader level, Haney (2006) characterized these environments as a combination of “high threat and low control” (p. 170). Prisoners and staff alike face omnipresent psychological and physical threats, which they often have very limited capacity to affect. Moreover, the adaptations that Haney says people use to deal with this combination (e.g., projecting tough veneers, suppressing emotionality, and being distrustful of others) frequently intensify matters. Interpersonal interactions based on views of others as a threat to safety and well-being, along with sensing minimal potential to exert control over situations, are not conducive to positive or even innocuous outcomes. As such, understanding correctional staff perceptions toward danger and safety is essential in determining the health and resiliency of a prison environment.

It is important to note that the current focus on perceived danger is conceptually different from fear and risk of victimization among correctional staff. Perceptions of danger are cognitive and more generalized/pervasive, while fear/risk is affective and likely more targeted and focused on particular sources (Gordon et al., 2003; Gordon et al., 2013). Due, however, to the lack of existing correctional research predicting perceived job danger, utilizing what is known regarding predictors of perceived victimization by inmates serves as a foundation. Within the prison environment, research identifies perceptions of danger, safety, or fear of victimization as an important predictor of factors such as perceived organizational fairness, job satisfaction, turnover, organizational commitment, and stress (Cullen et al., 1985; Dowden & Tellier, 2004; Griffin, 2001; Lambert, Cluse-Tolar, & Hogan, 2007). The impact of such relationships on the daily work environment, retention of correctional staff, and the climate and culture of the institutional environment is commonly investigated (Cullen et al., 1985; Dowden & Tellier, 2004; Griffin, 2001; Lambert, Cluse-Tolar, & Hogan, 2007; Lambert, Hogan, & Barton, 2002; Stichman & Gordon, 2015; Taxman & Gordon, 2009). Understanding the correlates of correctional staff perceptions on the dimension of workplace well-being or possible victimization by inmates is emerging in the literature (Gordon & Baker, 2015; Gordon et al., 2003; Gordon et al., 2013; Lai et al., 2012; Stichman & Gordon, 2015; Taxman & Gordon, 2009).

This body of research examines perceptions of correctional officers’ fear (emotional) and risk (cognitive) of primarily victimization by inmates at both the juvenile and adult levels, in the United States and overseas, and within single institutions or across multiple facilities. Seminal research by Gordon et al. (2003) explored perceptions of fear and risk of individual- and collective-level victimization by inmates among correctional officers working at two juvenile facilities. In general, the study identified officer gender as a consistent predictor of both individual and collective fear and risk of victimization. That is, female staff were concerned that one-on-one interactions or combined interactions of inmates could lead to psychological or physical harm. In addition, Gordon et al. (2003) identified that non-White officers and officers with less formal education had higher levels of fear of individual-level victimization by inmates; however, this study did not consider work environment factors, thus encouraging this study’s further exploration.

Additional research incorporating workplace variables as predictors of fear and risk of victimization by inmates surfaced (Gordon & Baker, 2015; Lai et al., 2012; Stichman & Gordon, 2015; Taxman & Gordon, 2009). While there is a lack of consideration of the many workplace variables, empirical research does provide a number of important organizational climate predictors. To illustrate, procedural justice was negatively related to fear of victimization by inmates (Taxman & Gordon, 2009). Adequate training and supervisory trust decreased perceptions of victimization (Lai et al., 2012). A sense of personal efficacy during interactions with inmates decreased fear of victimization (Stichman & Gordon, 2015). Job frustration predicted higher levels of emotional and cognitive fear of victimization (Gordon & Baker, 2015). In addition, for female officers working in male facilities, a sense of disorganization increased the level of cognitive fear of victimization by inmates (Gordon & Baker, 2015). It is also important to note that institutional composition was important in predicting safety as well. That is, correctional staff working in higher-level security prisons (Gordon & Baker, 2015; Gordon et al., 2013; Stichman & Gordon, 2015) and in male-only facilities (Gordon & Baker, 2015) reported higher perceptions of inmate fear and risk of victimization. The gender variable continued to be a consistent predictor of perceptions toward fear or safety when considering work environment variables; likewise, the race of the officers was an important predictor of perceptions of safety, with non-White officers reporting greater concern for their well-being. Finally, research indicates that the effects of personal factors are reduced when salient workplace factors are taken into account (Lai et al., 2012; Taxman & Gordon, 2009).

The current study seeks to expand the understanding of workplace safety by investigating the correlates of perceived job danger rather than perceptions toward victimization or fear. The concept of danger involves the risk of injury while on the job, whereas fear and risk of victimization consider emotional and cognitive responses, primarily, to inmate interactions. Research suggests the concepts should be viewed as distinct (see Gordon & Baker, 2015; Stichman & Gordon, 2015). As previously indicated, examination of the antecedent predictors of perceived job danger is lacking among correctional officers. The one investigation of the issue reported that gender, supervisory support, and co-worker support influenced perceptions of danger among corrections officers (Garcia, 2009). Specifically, the perception of danger or risk of injury among correctional officers was lower among men and those who perceived a solid level of backing from their supervisors and co-workers. This is not surprising when considering the larger body of literature examining perception of danger or safety in other professions, which identifies the trust and affective state of the leader as essential in developing safe environments (Conchie, Taylor, & Donald, 2012; Flin & Yule, 2004; Zohar, 2002). In addition, when positive relationships among supervisors and staff are achieved, the overall milieu of support is increased. Theoretically, this suggests improved support among staff and the ability to decrease the sense or actual risk of injury (Zohar, 2002; Zohar & Tenne-Gazit, 2008). The present study expands the scope of inquiry by including the variables of decision-making input, formalization, integration, job feedback, instrumental communication, supervision consideration, average daily contact with inmates, and position within the prison as possible correlates of perceived danger from the job.

Input into decision making considers the perceived degree of staff involvement in organizational matters (Slate & Vogel, 1997) referring to power sharing in an organization and allowing employees a voice in the process (Lambert, Minor, Wells, & Hogan, 2015). Granting employees a say in organizational matters enables a mechanism to express workplace concerns and offer recommendations, hopefully before an issue becomes a significant threat to staff safety. Theoretically, input into decision making expands the sense of an internal locus of control which can result in feeling less risk at work (Schmitz, Neumann, & Oppermann, 2000). However, a lack of control can lead to a sense of powerlessness, and heighten a perception that work is dangerous (Garland, Hogan, & Lambert, 2013). Taken together, a lack of perceived input into decision making can lead to feeling more vulnerable and less safe because of an externally imposed locus of control.

Formalization is the extent to which written rules and procedures are established within an organization (Lambert, Paoline, & Hogan, 2006; Price & Mueller, 1986). Simply put, it is the codification of organizational rules and procedures (Pandey & Scott, 2002). According to Taggart and Mays (1987), formalization is “the use of well-defined rules and regulations to govern the behavior of individuals so that actions within the organization become standardized” (p. 186). Formalization can provide staff guidance on how to handle duties, a sense of control to assist in avoiding conflict, as well as helping them be effective at their jobs (Lambert & Paoline, 2012; Pandey & Scott, 2002). Formalization is empirically distinct from “red tape” or excessive and even irrational bureaucracy (Pandey & Scott, 2002). Formalization provides structure and guidance, which should lead to reduced feelings of uncertainty and perceived risks in the workplace. Conversely, unclear or excessively cumbersome organizational procedures and expectations can not only result in frustration but also could lead to a “screw-up,” placing the staff member and possibly co-workers at risk of harm, increasing correctional personnel’s perception of danger.

Creating group cohesion among employees, work groups, departments, and/or divisions within an organization is referred to as integration (Mueller, Boyer, Price, & Iverson, 1994). Integration stresses that different units, departments, and groups work together rather than competing against one another (Lambert, Barton, Hogan, & Clarke, 2002; Mueller et al., 1994). Social integration or a sense of belongingness has been argued to be important for explaining fear of crime in the general population (Franklin & Franklin, 2009; Hale, 1996). Gordon et al. (2013) postulated that social integration in the workplace may explain different levels of fear of being at risk on the job, although when tested as a predictor of fear of victimization, the sense of belongingness was not statistically significant (Gordon & Baker, 2015). Integration can create a sense of group cohesion and belonging (Garland et al., 2013; Miller & Droge, 1986) and can create a more pleasant working experience, resulting in less strain (Lambert et al., 2015). As Garland et al. (2013) noted, integration refers to a philosophy of “we are all in this together” rather than “us versus them.” This sense of cooperation in the workplace could lead to a more pleasant working experience, resulting in fewer negative feelings, which could result in a perception that the job is less dangerous. Integration could result in a sense of support from co-workers, including co-workers providing guidance on how to handle work conflicts and reduce negative stimuli and threats from the job (Lambert & Paoline, 2012). However, a lack of integration could result in staff members not knowing where to go for help, and, in turn, making them feel less safe at work.

Job feedback refers to the level of information received on how well the job is being done and whether improvement is needed (Lambert, Hogan, & Barton, 2002). Staff need timely job feedback to determine whether they are performing their jobs properly or if changes are needed (Price & Mueller, 1986). Feedback not only can give staff direction, but it can also promote empowerment (Lambert, Hogan, & Barton, 2002), leading to a belief that staff can successfully handle the challenges of the job. A lack of job feedback can cause strain on staff, as they may not know whether they are performing their duties correctly and they may question their abilities on the job, which may make them feel more at risk of being harmed. A paucity of feedback can also lead to staff performing their work with prisoners less effectively (e.g., compromising security).

Instrumental communication refers to the formal transmission of information about the job by an organization to its members (Agho, Mueller, & Price, 1993). For correctional staff to do their jobs effectively, clear and meaningful communication is needed. As Lambert, Barton, et al. (2002) noted, instrumental communication means providing information to staff about procedures, issues, problems, and changes. Instrumental communication helps staff understand what is to be done and how it is to be done, as well as explaining issues that have arisen and how to address these issues. This can help prevent problems and work conflicts from arising, and, when problems or conflicts do occur, how to best deal with them before they escalate (Garland et al., 2013). Instrumental communication should allow staff members to feel like they are a productive and an integral part of the correctional agency (Lambert, Barton, et al., 2002). Communication allows staff to believe they possess skills and abilities to resolve problems that occur on the job, including conflict with inmates. Furthermore, communication provides staff a mechanism to raise their concerns and fears, allowing supervisors and administrators a chance to address them. A lack of meaningful organizational communication can frustrate staff and cause problems that place them at greater risk of injury on the job (Lambert et al., 2015). Indeed, job frustration was found to be a predictor of emotional and cognitive fear of victimization among correctional officers (Gordon & Baker, 2015).

Quality and caring supervisors are a critical workplace resource for correctional staff. Good supervisors provide guidance, structure, control, and support for staff (Cherniss, 1980). Supervisors need to be considerate and supportive of staff (Armstrong & Griffin, 2004; Brough & Williams, 2007). Such qualities can lead to positive feelings among subordinates (Garland et al., 2013). Staff look to their supervisors to help cope with the demands of the job and to deal effectively with work problems and conflicts (Cullen et al., 1985). In addition, supportive supervisors can be a resource for correctional staff to deal with potential workplace strains and conflicts (Cullen et al., 1985; Lambert, Hogan, & Tucker, 2009). A lack of quality supervision can hinder staff in doing their jobs and lead to feelings that the workplace is a negative place.

Gordon et al. (2003; Gordon et al., 2013) reported that fear of being victimized by inmates was of greater concern to correctional staff than being harmed by co-workers. This suggests that greater interaction with inmates could result in greater feelings of being at risk on the job. Regular interaction with inmates can be stressful and may also result in a staff member feeling more at risk of harm (Higgins et al., 2012; Whiteacre, 2006). As such, daily interaction with inmates was hypothesized to be positively associated with perceptions that the job is dangerous.

Examining difference among position type held within prisons is not new. Prior research indicates custodial staff and non-custodial staff are affected differently by the work milieu (Armstrong & Griffin, 2004; Lambert, 2004). The role each plays within the work setting is important to achieving a safe and secure environment, and the research is mixed on the overall impact of varied roles. To illustrate, research has uncovered differences between custodial and non-custodial staff regarding job satisfaction, organizational commitment, and job stress (Lambert, 2004; Robinson, Porporino, & Simourd, 1996). In general, this literature identifies higher stress and lower satisfaction and commitment among custodial staff; however, Armstrong and Griffin (2004) revealed relatively equal levels of stress or strain across position type. More recently, Lai et al. (2012) investigated the effects of position on perceptions of safety and uncovered that custody staff reported higher perceptions of work-related danger compared with non-custody staff. Thus, position was hypothesized to be positively associated with perceived danger, although this relationship has not been well-investigated among correctional staff in the United States.

Method

Participants

All available staff (N = 420) at a state prison were surveyed, including custody officers, unit management personnel (i.e., counselors, case managers, and unit managers), prison industry supervisors, food service supervisors, facility maintenance employees, medical employees, and hobby and recreational staff.¹ The staff were informed that the survey was voluntary and that their responses would remain anonymous. In 2004 at the time of the survey, the prison housed approximately 1,100 male, felony offenders who were serving long sentences (i.e., 10 years or more), and the prison was classified as a maximum security institution. A total of 272 usable surveys were returned, resulting in a response rate of 65%. Among those who responded, 24% indicated that they were a supervisor of other prison staff. About 76% indicated that they were men. The median age was 44, and age ranged from 21 to 61. The mean age was 42.55, with a standard deviation of 8.32. Fifty percent indicated they held a custody position (i.e., correctional officer). The median number of years working at the prison was nine, and tenure ranged from zero to 26. In terms of highest educational level, 10% were high school graduates, 47% had some college but no degree, 20% had an associate’s degree, 16% had a bachelor’s degree, and 5% had a graduate or professional degree. Approximately 81% of the participants indicated that they were White, 8% Black, 2% Hispanic, 3% Native American, and 5% another race. In an effort to compare the survey respondents with the overall worker population at the prison, the human resource office at the prison provided some demographic information. Of all staff working at the correctional institution, 77% were male, 86% were White, and 53% were custody officers. In addition, the human resource office estimated that the median age was 45, and the average tenure at the facility was approximately 10 years. Based on demographic characteristics (i.e., gender, race, position, age, and tenure), the participants appeared to be similar to those who were employed at the institution at the time of the survey. Non-responders may have differed by other variables (e.g., workplace perceptions) which could not be tested.

Variables

Dependent variable

Perceived danger from the job was the dependent variable, and it was measured using four items (e.g., “In my job, a person stands a good chance of getting hurt”) from Cullen et al. (1985). See the appendix for all the items for this variable. The items were answered using a five-point Likert-type scale of strongly disagree (coded 1) to strongly agree (coded 5). Based on a Principal Axis factor analysis, all the items loaded on one factor; the Cronbach’s alpha value was .82, indicating that the items had unidimensionality. The responses to the three items were summed together to form an additive index. Higher levels on this variable signify higher levels of concern of being victimized on the job.

Independent variables

For the work environment variables, items were summed to form additive indexes—see the appendix for all the items. The items for input into decision making, formalization, integration, feedback, and supervision consideration were answered using a five-point Likert-type scale of strongly disagree (coded 1) to strongly agree (coded 5). The instrumental communication items were measured using five response options ranging from not informed at all (coded 1) to very well-informed (coded 5). Input into decision making was measured using seven items (e.g., “I have input into matters that affect me at my job”) from Curry, Wakefield, Price, and Mueller (1986), which had a Cronbach’s alpha value of .84. Six items (e.g., “At this prison, written rules and procedures are highly emphasized”) adopted from Finlay, Martin, Roman, and Blum (1995) and Oldham and Hackman (1981) were used to measure formalization and had a Cronbach’s alpha of .61. Integration was measured by five items (e.g., “At this prison, there is a great deal of departmental interaction on most decisions”) adapted from Miller and Droge (1986), which had a Cronbach’s alpha of .73. Job performance feedback was measured by two items (e.g., “I am given adequate information on how well I am performing my job”) adapted from Wright and Saylor (1992), which had a Cronbach’s alpha of .64. Instrumental communication was measured using five items (e.g., “How well-informed are you regarding what is to be done?”) from Curry et al. (1986), and had a Cronbach’s alpha of .85. Three items (e.g., “My supervisor demonstrates sensitivity to such personal needs as shift and leave requests by fairly balancing them with the needs of the prison”) from Wright and Saylor (1992) were used to create an index measuring perception of supervisor consideration, and the Cronbach’s alpha was .77. A factor analysis using the Principal Axis technique was calculated for the items used to measure the independent variables. Each loaded on the predicted factor, indicating unidimensionality of the index. Higher levels on these variables indicate higher levels of perceptions of input into decision making, formalization, integration, job feedback, instrumental communication, and supervisor consideration, respectively. Based on the literature, input into decision making, formalization, integration, job feedback, instrumental communication, and supervision were hypothesized to have a negative relationship with perception of danger. In addition, a positive relationship was anticipated for daily interaction with inmates and perceived danger.

Average daily contact was measured using the following question: “On a typical day, about how much time do you spend in direct contact with inmates (e.g., supervising, talking with, counting, training, counseling, etc.).” The response options were less than 1 hr (17%, coded as 1), 1 to 2 hr (11%, coded as 2), 3 to 4 hr (16%, coded as 3), 5 to 6 hr (20%, coded as 4), and >7 hr (36%, coded as 5). Finally, the personal characteristics of supervisor status, gender, age, position, tenure, educational level, and race were included more as control than explanatory variables. Supervisory status was coded with non-supervisors as 0 (76%) and supervisors as 1 (24%). Gender was a dichotomous variable, where women (24%) were coded as 0 and men (76%) as 1. Age was measured in continuous years, and the median age was 44 years. Position represented if the participant worked in a custody (coded as 1, 50%) or a non-custody position (coded as 0, 50%). Tenure was measured in years at working at the prison, and the median tenure at the prison was 9 years. Educational level measured if the participant’s highest educational degree was a college degree (coded as 1, 41%) or not (i.e., high school diploma or some college but no degree, coded as 0, 59%). Race was a dichotomous variable representing if the participant was non-White (coded as 0, 19%) or White (coded as 1, 81%).

Results

The descriptive statistics for the variables are presented in Table 1. There appeared to be significant variation in both the dependent and independent variables (i.e., none were constants). Statistical tests indicated that the variables were normally distributed, and checks for univariate and multivariate outliers were conducted.

Table 1.

Descriptive Statistics for Study Variables.

Variables	Description	Min	Max	Med	M	SD
Super status	0 = Non-supervisor and 1 = Supervisor	0	1	0	0.24	0.42
Gender	0 = Female and 1 = Male	0	0	1	0.76	0.43
Age	Measured in years	21	61	44	42.55	8.32
Position	0 = Non-Custody Staff and 1 = Custody Staff	0	1	0.50	0.50	0.50
Tenure	Measured in years at the prison	0	26	9	9.94	6.82
Educ level	0 = No college degree and 1 = College degree	0	1	0	0.41	0.49
Race	0 = non-White and 1 = White	0	1	1	0.81	0.39
Input into DM	7-item additive index, α = .84	7	35	21	20.57	5.39
Formalization	6-item additive index, α = .61	14	30	23	22.93	2.60
Integration	5-item additive index, α = .73	5	22	14	13.44	3.04
Feedback	2-item additive index, α = .64	2	10	7	6.74	1.61
Instr comm	5-item additive index, α = .85	5	25	18	17.65	3.69
Sup consider	3-item additive index, α = .77	3	15	10	9.51	2.79
Avg contact	1 = <1 hr (17%); 2 = 1-2 hr (11%); 3 = 3-4 hr (16%); 4 = 5-6 hr (20%); 5 = >7 hr (36%)	1	5	4	3.49	1.48
Danger of job	4-item additive index, α = .82	4	20	14	13.58	3.54

Note. α = Cronbach’s alpha value; Super Status = supervisory status; Educ Level = educational level; Input into DM = input into decision making; Instr Comm = instrumental communication; Sup Consider = supervisor consideration; Avg Contact = average daily contact with inmates; Danger of Job = perceived danger from the job. The mean for a dichotomous variable (i.e., coded 0 and 1) represents the percentage of participants who are coded 1 on the variable. The number of participants was 272.

The bivariate correlations among the variables are presented in Table 2. Male staff and staff holding the position of correctional officer generally reported feeling at greater risk of injury while at work. Staff possessing a college degree and White staff, on average, reported lower perceptions of risk from the job than staff without a college degree and non-White staff. Higher perceptions of input into decision making, workplace integration, job feedback, instrumental communication, and supervisor consideration were all negatively correlated with perceptions of danger from the job (i.e., increases in each were associated with decreases in feeling at risk on the job). Average daily contact with inmates was positively correlated with perceived risk. Correctional officers in general were less likely to report having earned some type of college degree and reported higher levels of average daily contact with inmates. Interestingly, custody staff also tended to report lower levels of perceptions of input into decision making, formalization, workplace integration, job feedback, and supervisor consideration. As already noted, the aforementioned variables that have a significant correlation with the custody variable also had significant associations with the dependent variable. Custody officers also tended to report higher levels of feeling at risk at work. These bivariate correlations suggest that the effects of these variables on perceived danger of the job may vary between custody and non-custody staff. As such, three different multivariate regression equations were run—one for all the staff, one for just custody staff, and one for just non-custody staff.

Table 2.

Correlation Matrix for Study Variables.

Variable	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.	14.
1. Super status	1.00
2. Gender (1 = male)	0.06	1.00
3. Age	0.17**	0.09	1.00
4. Position (1 = custody)	−0.33**	0.23**	−0.18**	1.00
5. Tenure	0.14*	0.24**	0.40**	0.06	1.00
6. Educ level (1 = college)	0.14*	−0.10	−0.01	−0.20**	−0.25**	1.00
7. Race (1 = White)	−0.02	−0.01	0.06	0.07	0.04	−0.06	1.00
8. Input into DM	0.31**	−0.14*	−0.10	−0.36**	−0.16*	0.18**	0.06	1.00
9. Formalization	0.08	−0.03	−0.03	−0.12*	0.00	0.04	0.18**	0.36**	1.00
10. Integration	0.09	−0.12	−0.03	−0.16**	−0.15*	0.17**	0.05	0.61**	0.38**	1.00
11. Feedback	0.09	−0.06	−0.02	−0.16**	0.01	0.00	0.19**	0.58**	0.38**	0.43**	1.00
12. Instr Comm	0.03	0.12	−0.08	−0.11	−0.13*	0.07	0.07	0.42**	0.47**	0.36**	0.42**	1.00
13. Sup consider	0.26**	−0.14*	0.02	−0.42**	0.02	0.08	0.12*	0.75**	0.40**	0.43**	0.63**	0.41**	1.00
14. Avg contact	−0.22**	0.21**	−0.19**	0.24**	−0.07	−0.17**	−0.11	−0.12*	−0.14*	−0.19**	−0.11	−0.10	−0.17**	1.00
15. Danger	−0.10	0.24**	−0.10	0.35**	0.04	−0.17**	−0.13*	−0.34**	−0.07	−0.25**	−0.24**	−0.13*	−0.28**	0.43**

Note. Super Status = supervisory status; Educ Level = educational level; Input into DM = input into decision making; Instr Comm = instrumental communication; Sup Consider = supervisor consideration; Avg Contact = average daily contact with inmates; Danger = perceived danger from the job. The number of participants was 272.

p ≤ .05. **p ≤ .01.

An Ordinary Least Squares (OLS) regression analysis was conducted for the entire group of staff (N = 272), and the results are reported in Table 3. The Variance Inflation Factor (VIF) and tolerance statistics indicated no problem with multicollinearity among the independent variables. High multicollinearity occurs when an independent variable shares a very large part of its variance with the other independent variables in the regression equation, and this hinders the estimation of the effects on the dependent variable by the independent variables in question. VIF scores above five or tolerance values below .20 indicate a problem with multicollinearity among the independent variables (Tabachnick & Fidell, 2013). None of the VIF or tolerance scores were close to the cut-off points. Specifically, the VIF values ranged from 1.07 to 2.77, and the tolerance values ranged from .38 to .93. In addition, the issues of outliers, influential cases, normality, linearity and homoskedasticity of residuals, and independence of errors in the regression analysis were tested and addressed (Tabachnick & Fidell, 2013).

Table 3.

Ordinary Least Squares Regression Results With Perceived Danger From the Job as the Dependent Variable for All Correctional Staff, a Subset of Custody Staff (i.e., Correctional Officers) Only, and a Subset of Non-Custody Staff Only.

Variable	All staff		Custody staff		Non-custody staff
Variable	B	C	B	β	B	β
Personal characteristics
Supervisory status (1 = Supervisor)	0.97	.12*	−0.03	−.01	1.52	.20*
Gender (1 = male)	0.68	.08	−0.57	−.07	1.19	.10
Age	−0.01	−.02	0.03	.08	−0.48	−.10
Position (1 = custody)	1.52	.22**	n/a	n/a	n/a	n/a
Tenure	−0.02	−.03	−0.04	−.08	−0.04	−.08
Educational level (1 = college degree)	−0.30	−.04	−0.33	−.05	−0.64	−.08
Race (1 = White)	−1.01	−.11*	−0.52	−.06	−1.18	−.12
Work environment variables
Input in decision making	−0.21	−.32**	−0.25	−.42**	−0.18	−.25*
Formalization	0.21	.15*	0.19	.16*	0.17	.13
Integration	−0.03	−.02	−0.16	−.15	0.12	.11
Feedback	−0.19	−.09	0.03	.02	−0.39	.12
Instrumental communication	0.02	.02	0.09	.11	−0.07	−.06
Supervision consideration	0.11	.09	0.04	.03	0.28	.16
Average daily inmate contact	0.80	.34**	0.44	.21*	1.10	.46**
F value (df)	9.57 (14, 242)**		3.58 (13, 119)**		5.22 (13, 110)**
R²/adjusted R²	.36/.32		.28/.20		.38/.31

Note. See Table 1 for a description of the variables and how they are coded. B = unstandardized coefficient; β = the standardized coefficient; n/a = not applicable; df = degrees of freedom. The total number of participants was 272, and after listwise deletion of missing responses, the number of participants for the regression analysis was 256. The total number of custody officers was 137, and after listwise deletion of missing responses, the number of custody officers for the regression analysis was 132. The total non-custody staff was 135, and after listwise deletion of missing responses, the number of non-custody staff for the regression analysis was 123.

p ≤ .05. **p ≤ .01.

The R² value was .36, which means that about 36% of the variance in the dependent variable was explained by the independent variables. Among the personal characteristics, supervisory status, position, and race had statistically significant associations with the dependent variable. Supervisors and custody officers reported higher levels of feeling at risk at work, while White staff reported lower levels. Among the work environment variables, perceptions of input into decision making, formalization, and average daily contact with inmates had statistically significant associations with the dependent variable. Perceptions of greater input into decision making was associated with lower feelings of being at risk of harm at work. Conversely, increases in perceptions of formalization and greater daily contact with inmates were linked with higher perceived risk. The finding that formalization was linked to increased perceived danger was contrary to what was predicted, where it was theorized to result in lower levels of fear for safety. Last, based on the standardized regression coefficient (β), average daily contact with inmates had the largest effect on perceived risk, followed closely by input into decision making. Position had the third largest effect, followed by formalization, and then supervisory status. Race had the smallest sized effect.

A second OLS regression analysis was conducted using only participants who held the position of correctional officer (n = 137), and the results are also reported in Table 3. Based on the tolerance and VIF scores, multicollinearity was not a problem. Specifically, the VIF values ranged from 1.10 to 3.23, and the tolerance values ranged from .31 to .91. In addition, the issues of outliers, influential cases, normality, linearity and homoskedasticity of residuals, and independence of errors in the regression analysis were tested and addressed (Tabachnick & Fidell, 2013). The independent variables accounted for approximately 28% of the observed variance in the dependent variable. For custody officers, none of the personal characteristics had a significant effect. Input into decision making had a negative association, which means that the more input a person felt he or she had at work, the less the person felt at risk. Increases in perceptions of formalization were associated with greater concerns over safety, which again was contrary to what was hypothesized. The more inmate contact reported, the greater the level of feeling being at risk. Finally, input into decision making had the greatest sized effect, average daily contact had the second largest sized effect, and formalization had the smallest sized effect.

A third OLS regression analysis was conducted using only participants who held non-custody positions (i.e., were not custody officers, n = 135), and the results are also reported in Table 3. Based on the tolerance and VIF scores, multicollinearity was not a problem. Specifically, the VIF values ranged from 1.12 to 3.08, and the tolerance values ranged from .33 to .90. In addition, the issues of outliers, influential cases, normality, linearity, and homoskedasticity of residuals, and independence of errors in the regression analysis were tested and addressed (Tabachnick & Fidell, 2013). The independent variables accounted for approximately 38% of the observed variance in the dependent variable, which was 10 percentage points higher than that in the regression analysis for custody officers. Among non-custody staff, the only personal characteristics to have a significant association was supervisory status; non-custody supervisors reported higher levels of fear than their non-supervisory counterparts. Input into decision making had a negative association, as it did for custody staff; the regression coefficient for this variable was smaller for non-custody staff as compared with custody staff. In addition, increases in daily contact with inmates was positively associated with the dependent variable, as was also the case with custody officers; however, the size of the regression coefficient for non-custody staff was more than twice the size observed for their counterparts in custody. Formalization approached but did not reach statistical significance for non-custody staff. Finally, among non-custody staff, average daily contact had the largest sized effect, followed by input into decision making, and then supervisory status.

Discussion and Conclusion

Research indicates that one source of correctional staff stress and other negative outcomes is the perception of exposure to danger on the job (Armstrong & Griffin, 2004), but research has not given sufficient attention to possible antecedents of such perceptions or to possible differences between custody and non-custody staff. The current investigation identified several personal variables (i.e., supervisory status, position, and race) and organizational characteristics (i.e., input into decision making, formalization, and contact with inmates) as related to overall perceptions of danger on the job. The input and contact variables were common correlates of perceived danger for custody and non-custody staff.

Comparatively, the present study yielded findings both consistent and inconsistent with the limited previous research. Similar to prior research on personal variables (Gordon et al., 2003; Gordon et al., 2013; Taxman & Gordon, 2009), the data for all staff in this study revealed that age and tenure were not significantly related to perceived danger on the job, whereas position was (Lai et al., 2012). Contrary to previous studies (Gordon et al., 2003; Gordon et al., 2013; Taxman & Gordon, 2009), neither gender nor educational level were significantly related to such perceptions. It could be that staff at the maximum security prison studied were at greater risk of harm, or it could be that the current study included different workplace variables than those used in past studies. The latter suggests that once workplace factors are taken into account, personal characteristics do not play as great a role in predicting perceived risk of danger on the job. The present study revealed higher perceptions of danger among non-Whites, which is consistent with prior research examining perception of fear of victimization (see Gordon & Baker, 2015; Gordon et al., 2013; Stichman & Gordon, 2015; Taxman & Gordon, 2009). This held only for the entire sample of correctional staff; the race variable was not significant for the subset of custody officers or among non-custody staff, but it did come close to statistical significance. It appears that race could play a greater role among non-custody staff than their custody counterparts, but, due to a lack of statistical significance, it cannot be conclusively determined. Interestingly, supervisory status was a significant predictor for all staff and among non-custody staff, but not for custody staff.

The findings underscore the relevance of separately analyzing custody and non-custody staff. As indicated, position was a significant predictor in the analysis of all the staff, with custody officers reporting greater concern of risk from the job. The custody staff are responsible for maintaining institutional order and security and also have to restrain disruptive and potentially violent inmates (Armstrong & Griffin, 2004). It could be that custody staff are in posts that make them feel more vulnerable, such as working in large housing units. This is not to say that non-custody staff are free from risk, especially those who have regular contact with inmates in open areas such as the dining hall and housing unit, and these staff often do not have the same safety equipment such as body alarms, as do custody officers. Non-custody staff in this study occupied a wide array of positions, and it was not possible to look at level of perceived risk by each specific position. It could that some non-custody positions have greater concern for safety than do other positions. Future research needs to determine why position was a significant predictor in the current study. In the end, the current finding does not mean that custody staff are at greater actual danger, but they perceive themselves to be. It is important to note that some risk of danger is pervasive in a maximum security prison, and threat of harm should be taken very seriously by all staff.

The finding that non-White staff expressed greater concern about personal safety than Whites could partly reflect the fact that 86% of the staff at the institution studied were White, and such, White staff may feel less vulnerable because they are part of the majority. Minority groups, in terms of their overall representation, may experience tokenism (Kanter, 1977). This may result in non-White staff feeling less part of the organization, and, as such, at greater risk of harm. This is an untested explanation. Interestingly, the largest group of inmates at the prison were non-White, mainly African American. Unfortunately, no data were available on racial dynamics among staff at the institution. If such data were available, one could investigate the degree to which non-White staff were less likely than White counterparts to feel they could rely on co-workers for assistance and support in the event of threatening occurrences. Furthermore, race was not a significant predictor in the analyses of custody staff and non-custody staff; however, the regression coefficient for race among non-custody staff was larger for non-custody staff as compared with custody officers. The race variable may not have attained statistical significance for non-custody staff because of fewer cases reducing the power of analysis. It was not possible to break down the non-White staff by position and race/ethnicity to determine whether different groups expressed greater concern for their safety than other groups. The relationship between race and perceptions of being at risk requires further study.

The data also revealed that supervisory staff perceived greater risk of danger than non-supervisors among non-custody but not custody staff. Safety threats could be more routinized for supervisors working in custody, as these supervisors are likely to deal with a myriad of threats and incidents over time. Supervisors working outside custody may experience threats less frequently and less directly. Their work is more focused on rehabilitative, service, legalistic, or bureaucratic tasks, but as supervisors, they also likely feel some responsibility for the safety of other staff and prisoners. Compared with supervisors working in custody, their exposure to danger may be less routine and more vicarious. Paradoxically, then, being more distanced from direct threat of danger could render non-custodial supervisors more sensitive to and concerned about potential dangers in the environment. Further research needs to clarify the interaction of supervisory status and custodial versus non-custodial position as regards perceptions of danger. In addition, as indicated in Table 2, supervisory status and position were negatively correlated (r = −.33, p ≤ .01), indicating that supervisors in this study were less likely to work in custody. As such, it could be that there were not a sufficient number of supervisors in custody to reach a statistically significant finding between supervisory status and perceived danger of the job.

The relationship of work environment variables to perceived job danger is not well-understood due to a scarcity of research. Favorable perceptions of procedural justice (Taxman & Gordon, 2009), positive views toward training and supervisors (Lai et al., 2012), positive self-empowerment to control inmate behaviors (Stichman & Gordon, 2015), and lower job frustration (Gordon & Baker, 2015) have been linked with staff perceiving a lower job victimization. Although the present study did not investigate these specific factors, it did consider work-related items. Specifically, two work environment variables (i.e., input into decision making and average daily contact with inmates) were significantly related to perceptions of danger among both custody and non-custody staff. More daily contact with prisoners likely translates into greater opportunity for risks to safety. Interestingly, amount of inmate contact appeared to have a slightly stronger effect for non-custody staff than it did for custody officers. Custody officers reported on average greater contact with inmates (see Table 2, r = .24, p ≤ .01). It could be that greater contact with inmates provides officers with a greater sense of confidence and increased skill in diffusing potentially dangerous situations. What future research needs to determine is how perceptions of danger vary with the kind or quality of contact with prisoners, because staff approach can influence prisoner response. For example, staff whose approach to interacting with inmates is derogatory and disrespectful could have very different perceptions of danger than staff who display professionalism and basic respect. Besides operating on the individual level, this phenomenon operates on the aggregate level in an institution (Liebling, 2008) and, therefore, has major implications for the enculturation of newer staff to a given prison environment.

Input into decision making was the strongest correlate of perceived risk among custody officers and had the second greatest sized effect among non-custody staff. It was also a significant predictor in the regression analysis of all staff. Having input empowers staff to express their viewpoints and recommendations and probably expands their internal locus of control at work. In an environment where uncertainty often runs high and threat of danger is ever-present (Haney, 2006), staff who are given meaningful opportunities for input probably feel less threatened. In contrast, high external control and a lack of input can render staff feeling less able to affect their own welfare and more vulnerable to danger (Garland et al., 2013; Schmitz et al., 2000).

For all staff and custody officers, formalization had a positive association; however, while in a positive direction, the relationship failed to reach statistical significance for non-custody staff. The lack of statistical significance could be the result of having only 135 cases, which can affect the power of analysis. The positive relationship between formalization and perceived danger of the job was contrary to that hypothesized. Inspection of the formalization items reveals that the items tap availability and ubiquity of policies and procedures rather than the extent to which staff see these as being of good quality (e.g., effective at furthering institutional goals, facilitating rather than hamstringing staff, etc.). Given this, staff are aware and recognize policies exist, but staff report a higher sense of risk of injury. What is unknown is if this sense of danger is due to inconsistency in following such formal policies, or if it is due to regular following of rules seen as placing staff at risk. In addition, there was a significant positive bivariate relationship between formalization and input into decisions (r = .36, p ≤ .01). This pattern may seem anomalous, given the inverse relationship of input into decisions with perceived risk discussed above. But it is worth remembering that having input into decisions (especially rather mundane ones) is not the same as affecting formal policies. Institutional policy formation involves processes of negotiation and compromise that are often politically charged. The outcome can be well-intended policies and procedures that are frustrating and unwieldy for staff to implement, less effectual than desired, or even counterproductive. Hence, it appears that the measure used did not adequately distinguish the kind of cogent formalization that might enhance perceptions of safety from irrational formalization permeated with “red tape” (Pandey & Scott, 2002) and inefficiency. Instead of providing staff structure and guidance, the latter may undermine their autonomy, making them sense greater risk of injury on the job.

Assuming the results are replicated, there are important policy implications. There is a need for additional scholarly work in this area to understand perceptions of being at risk of harm and the variables which influence these perceptions. Without this information, it will be difficult to reduce perceptions of harm. Similarly, correctional administrators need to be aware that level of perceived danger from the job is an issue, and the current results indicate workplace factors play a role. Improving perceptions of input into decision making may reduce fear of safety for both custody and non-custody staff. Correctional administrators should hold focus groups with staff, particularly custody officers, to find out how to improve input by staff, including how to make the workplace safer. Devaluing staff through denial of meaningful participation in decisions may yield some short-term conveniences for administrators but can be counterproductive over the long term if it leads staff to feel restricted and threatened on the job. Administrators need to reach out to staff, particularly custody officers, to find out why their perceptions of formalization are linked to being more concerned with safety. As noted before, excessive or unclear formalization can result in “red tape” which actually hinders rather than help people (Pandey & Scott, 2002). Administrators need to strive toward effective formalization which provides people structure and guidance for carrying out responsibilities without undermining their autonomy and efficacy; simply, the point is to strive for cogent over irrational formalization. Furthermore, a relatively high level of daily contact between staff and prisoners is unavoidable in all but the most segregated correctional housing units. High daily contact need not invariably be associated with staff perceptions of danger. The quality of staff–prisoner interaction matters greatly, and perceived risk can probably be reduced through interaction with prisoners that demonstrates professionalism and respect, combined with responsiveness to legitimate prisoner living needs in such areas as safety, nutrition, health care, and sanitation. Both scholars and correctional administrators need to explore ways to help insure that higher levels of inmate contact will not promote increased concerns of well-being at work.

Like many studies, the current one has shortcomings. Staff at a single maximum prison were surveyed, and at this prison the majority of staff were White (81%) and male (76%). The results may be situational or contextual and vary by different types of correctional facilities. Staff at other correctional facilities need to be studied. As mentioned above, there were no data on staff racial dynamics or characteristics of contact with prisoners, and, as just discussed, there were issues with the formalization measure. In addition, staff working at other correctional facilities need to be studied to determine if the findings can be replicated. As this study was cross-sectional, it is not possible to show a causal relationship between the workplace variables and perceived risk of the job. Longitudinal data are needed to determine the causal direction of the associations between work environment variables and concern for safety. Future research is needed to determine what other factors account for the variance of perceptions of danger presented by the job. The R² values for the three regression equations were .36, .28, and .38, which means that 62% to 72% of the variance is not accounted for by the independent variables used in the current study. As such, future studies should include other variables to explain the variation in perceived danger of the job. Actual past victimization by the respondent could help explain perceptions of job risk. Research on fear of crime among civilians has often identified past victimization as a significant predictor (Lai et al., 2012; Oh & Kim, 2009), but not always (Yin, 1980). Perceptions of risk play a role in being fearful; level of disorder of a neighborhood and known victimization of others can also play a role in fear of crime among the public (Yin, 1980). Future research should explore how other workplace factors may be linked to fear of safety. For example, future studies may wish to determine whether organizational (i.e., administrative), supervisor, and/or co-worker social support is associated with perceived danger of the job. Research indicates that different forms of social support can help staff deal with workplace stressors and strains (Dowden & Tellier, 2004). Another limitation is that the data for the current study were collected a decade ago. The safety technology in prisons, such as cameras and more sophisticated body alarms, may lead to lower concerns over safety. In addition, there could be generational effects on perceptions of danger at work and workplace variables. While not directly studying the correlates of feeling at risk, Cheeseman and Downey (2012) reported that there were generational or cohort effects on both job stress and job satisfaction in their study of custody officers. Thus, studies using newer data are warranted.

The current study used the measure of perceived danger of the job developed by Cullen et al. (1985). Other studies (Gordon & Baker, 2015; Gordon et al., 2013; Lai et al., 2012; Stichman & Gordon, 2015; Taxman & Gordon, 2009) have used fear of victimization measures (i.e., the emotional concern or threat of psychological or physical risk) and perceived risk of victimization (i.e., the cognitive likelihood of psychological or physical harm). Future research needs to ascertain the degree to which use of these various measures is associated with similar results. Furthermore, there is a need to break down the dependent variable according to perceived danger of the job from inmates and from co-workers (Gordon et al., 2003; Lai et al., 2012; Taxman & Gordon, 2009). In their study of juvenile correctional workers, Gordon et al. (2013) studied both fear of victimization by inmates and fellow staff. They found that staff were more fearful of victimization by juvenile inmates than by their co-workers, suggesting that the different types of perceptions of fear differ, which means the correlates of each may vary as well. In addition, Gordon et al. (2003; Gordon et al., 2013) measured fear of victimization and risk of victimization at the individual level (e.g., being assaulted by a single inmate) and collectively (e.g., being injured in a riot). There is clearly a need for far more research on the correlates of perceived danger of the job.

In closing, staff are the heart and soul of correctional facilities. People are inclined to react adversely when feeling their safety is threatened, and various negative outcomes are then likely to result (Commission on Safety and Abuse in America’s Prisons, 2006). It therefore behooves everyone associated with corrections, including taxpayers, for staff and prisoners alike to feel as safe as possible in the prison environment. Continued research on the correlates of staff perceptions of danger from the job can play an important part in furthering this end. There is definite need for more research on variables that may lead correctional staff to perceive their jobs as dangerous. This is critical information if scholars and administrators are to develop strategies to deal effectively with perceived danger of the job, which empirical research indicates promotes job stress and related negative outcomes. At the very least, it is hoped that the current study will spark further research in this salient area.

Footnotes

Appendix

Acknowledgements

The authors thank Janet Lambert for editing and proofreading the article. The authors also thank the Editor and anonymous reviewers for their comments and suggestions, which improved the article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Eric G. Lambert is a faculty member in the Department of Legal Studies at the University of Mississippi. He received his PhD from the School of Criminal Justice at the State University of New York at Albany. His research interests include organizational issues, job and organizational effects on the attitudes, intentions, and behaviors of criminal justice employees, and the international perceptions, attitudes, and views on criminal justice issues.

Kevin I. Minor is professor of Justice Studies at Eastern Kentucky University. In addition to correctional staff issues, his current research focuses on correctional law and the administration of capital and other extremist punishments in the United States.

Jill Gordon is an associate professor in Criminal Justice program at Virginia Commonwealth University. Her research interests center on correctional employees’ attitudes toward work, clients and the organization and assessment of correctional policy and programs. She is currently working on a multi-site project examining perceptions of culture and climate among correctional officers and inmates. Her work has appeared in numerous outlets including Justice Quarterly, Criminal Justice Policy Review, International Journal of Offender Therapy and Comparative Criminology, and Journal of Crime and Justice.

James B. Wells is professor of correctional and juvenile justice studies and director of the Center for Criminal Justice Education and Research at Eastern Kentucky University. His research interests and numerous publications focus on evaluating training and operational initiatives in adult corrections and juvenile justice as well as on developing gender-based descriptions of violence and safety in women’s correctional facilities.

Nancy L. Hogan is a professor in the School of Criminal Justice at Ferris State University in Michigan, USA. She received her PhD in Justice Studies from Arizona State University in Arizona, USA. Her research interests include job satisfaction and organizational issues of correctional staff, use of force, and cognitive behavioral treatment for inmates.

References

Agho

Mueller

Price

(1993). Determinants of employee job satisfaction: An empirical test of a causal model. Human Relations, 46, 1007-1027.

Armstrong

Griffin

(2004). Does the job matter? Comparing correlates of stress among treatment and correctional staff in prisons. Journal of Criminal Justice, 32, 577-592.

Black

. (2001, October). Correctional employee stress and strain. Corrections Today, 63(6), 82-85.

Blevins

Cullen

Frank

Sundt

Holmes

(2006). Stress and satisfaction among juvenile correctional workers: A test of competing models. Journal of Offender Rehabilitation, 44, 55-79.

Brough

Williams

(2007). Managing occupational stress in a high-risk industry. Criminal Justice and Behavior, 34, 555-567.

Cheek

Miller

(1983). The experience of stress for correctional officers: A double-bind theory of correctional stress. Journal of Criminal Justice, 11, 105-120.

Cheeseman

Downey

(2012). Talking “bout my generation”: The effect of “generation” on correctional employee perceptions of work stress and job satisfaction. The Prison Journal, 92, 24-44.

Cherniss

(1980). Staff burnout: Job stress in the human services. Beverly Hills, CA: Sage.

Commission on Safety and Abuse in America’s Prisons. (2006). Confronting confinement. New York, NY: Vera Institute of Justice.

10.

Conchie

Taylor

Donald

(2012). Promoting safety voice with safety-specific transformational leadership: The mediating role of two dimensions of trust. Journal of Occupational Health Psychology, 17, 105-115.

11.

Cornelius

(2001). The correctional officer: A practical guide. Durham, NC: Carolina Academic Press.

12.

Cullen

Link

Wolfe

Frank

(1985). The social dimensions of correctional officer stress. Justice Quarterly, 2, 505-533.

13.

Curry

Wakefield

Price

Mueller

(1986). On the causal ordering of job satisfaction and organizational commitment. Academy of Management Journal, 29, 847-858.

14.

Dowden

Tellier

(2004). Predicting work-related stress in correctional officers: A meta-analysis. Journal of Criminal Justice, 32, 31-47.

15.

Finlay

Martin

Roman

Blum

(1995). Organizational structure and job satisfaction: Do bureaucratic organizations produce more satisfied employees? Administration & Society, 27, 427-450.

16.

Finn

(1998). Correctional officer stress: A cause for concern and additional help. Federal Probation, 62(2), 65-75.

17.

Flin

Yule

(2004). Leadership for safety: Industrial experience. Quality and Safety in Health Care, 13, 45-51.

18.

Franklin

(2009). Predicting fear of crime: Considering differences across gender. Feminist Criminology, 4, 83-106.

19.

Garcia

(2009). Individual and institutional demographic and organizational climate correlates of perceived danger among federal correctional officers (Doctoral dissertation). Temple University, Philadelphia, PA.

20.

Garland

Hogan

Lambert

(2013). Antecedents of role stress among correctional staff: A replication and expansion. Criminal Justice Policy Review, 25, 527-550.

21.

Gordon

Baker

(2015). An analysis of correctional officers’ fear of victimization by inmates: Examining the influence of fear facilitators and fear inhibitors. Criminal Justice Policy Review. Advance online publication. Retrieved from http://cjp.sagepub.com/content/early/recent

22.

Gordon

Moriarty

Grant

(2003). Juvenile correctional officers’ perceived fear and risk of victimization: Examining individual and collective levels of victimization in two juvenile correctional centers in Virginia. Criminal Justice and Behavior, 30, 62-84.

23.

Gordon

Proulx

Grant

(2013). Trepidation among the “keepers”: Gendered perceptions of fear and risk of victimization among corrections officers. American Journal of Criminal Justice, 38, 245-265.

24.

Griffin

(2001). Job satisfaction among detention officers: Assessing the relative contribution of organizational climate variables. Journal of Criminal Justice, 29, 219-232.

25.

Griffin

(2006). Gender and stress: A comparative assessment of sources of stress among correctional officers. Journal of Contemporary Criminal Justice, 22, 4-25.

26.

Hale

(1996). Fear of crime: A review of the literature. International Review of Victimology, 4, 79-150.

27.

Haney

(2006). Reforming punishment: Psychological limits to the pains of imprisonment. Washington, DC: American Psychological Association.

28.

Hartley

Davila

Marquart

Mullings

(2013). Fear is a disease: The impact of fear and exposure to infectious disease on correctional officer job stress and job satisfaction. American Journal of Criminal Justice, 38, 323-340.

29.

Higgins

Tewksbury

Denney

(2012). Validating a measure of work stress for correctional staff: A structural equation modeling approach. Criminal Justice Policy Review, 24, 338-352.

30.

Huckabee

(1992). Stress in corrections: An overview of the issues. Journal of Criminal Justice, 20, 479-486.

31.

Kanter

(1977). Men and women of the corporation. New York, NY: Basic Books.

32.

Keinan

Maslach-Pines

(2007). Stress and burnout among prison personnel: Sources, outcomes, and intervention strategies. Criminal Justice and Behavior, 34, 380-398.

33.

Lahm

K. F.

(2009). Inmate assaults on prison staff: A multilevel examination of an overlooked form of prison violence. The Prison Journal, 89, 131-150.

34.

Lai

Y.-L.

Wang

H.-M.

Kellar

(2012). Workplace violence in correctional institutions in Taiwan: A study of correctional officers’ perceptions. International Journal of Comparative and Applied Criminal Justice, 36, 1-23.

35.

Lambert

(2004). The impact of job characteristics on correctional staff members. The Prison Journal, 84, 208-227.

36.

Lambert

Barton

Hogan

Clarke

(2002). The impact of instrumental communication and integration on correctional staff. The Justice Professional, 15, 181-193.

37.

Lambert

Cluse-Tolar

Hogan

(2007). This job is killing me: The impact of job characteristics on correctional staff job stress. Applied Psychology in Criminal Justice, 3, 117-142.

38.

Lambert

Edwards

Camp

Saylor

(2005). Here today, gone tomorrow, back again the next day: Absenteeism and its antecedents among federal correctional staff. Journal of Criminal Justice, 33, 165-175.

39.

Lambert

Hogan

(2010). Wanting change: The relationship of perceptions of organizational innovation with correctional staff job stress, job satisfaction, and organizational commitment. Criminal Justice Policy Review, 21, 160-184.

40.

Lambert

Hogan

Barton

(2002). Building commitment among correctional staff: The issue of feedback, promotional opportunities, and organizational fairness. Corrections Compendium, 27(3), 1-5, 24-28.

41.

Lambert

Hogan

Tucker

(2009). Problems at work: Exploring correlates of role stress among correctional staff. The Prison Journal, 89, 460-481.

42.

Lambert

Minor

Wells

Hogan

. (2015). Leave your job at work: The possible antecedents of work-family conflict among correctional staff. The Prison Journal, 95, 114-134.

43.

Lambert

Paoline

(2012). Exploring the possible antecedents of job involvement: An exploratory study among jail staff. Criminal Justice and Behavior, 39, 264-286.

44.

Lambert

Paoline

Hogan

(2006). The impact of centralization and formalization on correctional staff job satisfaction and organizational commitment. Criminal Justice Studies: A Critical Journal of Crime, Law and Society, 19, 23-44.

45.

Liebling

(2008). Why prison staff culture matters. In Byrne

J. M.

Humer

Taxman

F. S.

(Eds.), The culture of prison violence (pp. 105-122). Boston, MA: Allyn & Bacon.

46.

Miller

Droge

(1986). Psychological and traditional determinants of structure. Administrative Science Quarterly, 17, 333-357.

47.

Mitchell

Mackenzie

Styve

Gover

(2000). The impact of individual, organizational, and environmental attributes on voluntary turnover among juvenile correctional staff members. Justice Quarterly, 17, 333-357.

48.

Mueller

Boyer

Price

Iverson

(1994). Employee attachment and noncoercive conditions of work. Work and Occupations, 21, 179-212.

49.

J.-H.

Kim

(2009). Aging, neighborhood attachment, and fear of crime: Testing reciprocal effects. Journal of Community Psychology, 37, 21-40.

50.

Oldham

Hackman

(1981). Relationships between organizational structure and employee reactions: Comparing alternative frameworks. Administrative Science Quarterly, 26, 66-83.

51.

Pandey

S. K.

Scott

P. G.

(2002). Red tape: A review and assessment of concepts and measures. Journal of Public Administration Research and Theory, 12, 553-580.

52.

Price

Mueller

(1986). Absenteeism and turnover among hospital employees. Greenwich, CT: JAI Press.

53.

Robinson

Porporino

Simourd

(1996). Do different occupational groups vary on attitudes and work adjustment in corrections? Federal Probation, 60(3), 45-53.

54.

Schmitz

Neumann

Oppermann

(2000). Stress, burnout, and locus of control in German nurses. International Journal of Nursing Studies, 37, 95-99.

55.

Slate

Vogel

(1997). Participative management and correctional personnel: A study of the perceived atmosphere for participation in correctional decision making and its impact on employee stress and thoughts about quitting. Journal of Criminal Justice, 25, 397-408.

56.

Stichman

Gordon

(2015). A preliminary investigation of the effect of correctional officers’ bases of power on their fear and risk of victimization. Journal of Crime & Justice, 38, 543-558. doi:10.1080/0735648X.2014.929975

57.

Tabachnick

Fidell

(2013). Using multivariate statistics (6th ed.). Boston, MA: Pearson.

58.

Taggart

Mays

(1987). Organizational centralization in court administration: An empirical assessment. American Journal of Criminal Justice, 11, 180-198.

59.

Taxman

Gordon

(2009). Do fairness and equity matter? An examination of organizational justice among correctional officers in adult prisons. Criminal Justice and Behavior, 36, 695-711.

60.

Tewksbury

Higgins

(2006). Prison staff and work stress: The role of organizational and emotional influences. American Journal of Criminal Justice, 30, 247-266.

61.

Wells

Minor

Angel

Matz

Amato

(2009). Predictors of job stress among staff in juvenile correctional facilities. Criminal Justice and Behavior, 36, 245-258.

62.

Whiteacre

. (2006, June). Measuring job satisfaction and stress at a community corrections center: An evidenced-based study. Corrections Today, 68(3), 70-73.

63.

Wright

Saylor

(1992). Comparison of perceptions of the environment between minority and nonminority employees of the federal prison system. Journal of Criminal Justice, 20, 63-71.

64.

Yin

(1980). Fear of crime among the elderly: Some issues and suggestions. Social Problems, 27, 492-504.

65.

Zohar

(2002). The effects of leadership dimensions, safety climate, and assigned priorities on minor injuries in work groups. Journal of Organizational Behavior, 23, 75-92.

66.

Zohar

Tenne-Gazit

(2008). Transformational leadership and group interaction as climate antecedents: A social network analysis. Journal of Applied Psychology, 93, 744-757.