Working on local time: Testing the job-demand-control-support model of stress with jail officers.

PURPOSE: The demands and consequences of the correctional workplace are well-documented, but researchers have yet to examine the correlates of work stress among a large multi-facility sample of jail officers. We framed our inquiry within the Job Demand-Control-Support model that has guided researchers in parallel studies of work stress among prison officers.
METHOD: Data on officers' background characteristics, job demands, work-related control or autonomy, support from coworkers or family members, safety risks, and work-related stress were examined across 1380 officers working in 19 jails.
RESULTS: Unreasonable workloads, perceptions of insufficient staff, role problems, less control or autonomy, a lack of support at work or home, and exposure to violence were associated with greater stress among jail officers. Jails with characteristics that threatened order and security-having more inmates per officer and greater levels of inmate violence-had higher levels of stress among officer workforces.
CONCLUSIONS: Findings largely support research on work stress among prison officers and may inform efforts to reduce stress and improve quality of life among jail and prison personnel. Delegating appropriate workloads, the maintenance of control, social supports, and the reduction of safety risks are all important for reducing work stress among jail officers.

Introduction

The maintenance of safety, security, and order in any jail or prison is the primary responsibility of front-line correctional officers (Jacobs, 1977; Lombardo, 1989), whose challenging and often dangerous work environment is associated with a variety of job-related consequences such as mental health problems, substance abuse issues, and suicide (Brower, 2013; Cullen, Link, Wolfe, & Frank, 1985; Konda, Reichard, & Tiesman, 2012; Schaufeli & Peeters, 2000; Spinaris, Denhof, & Kellaway, 2012; Steiner & Wooldredge, 2015). Workplace violence and nonfatal injuries that result in absenteeism are common among correctional officers (U.S. Department of Labor, 2016; Konda et al., 2012), as are health and occupational adversities such as stress, burnout, job dissatisfaction, and fatigue (Lambert & Paoline, 2005; Paoline, Lambert, & Hogan, 2015; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015; Triplett, Mullings, & Scarborough, 1996). Correctional officers also experience significant conflict between their work and family lives (Armstrong, Atkin-Plunk, & Wells, 2015; Lambert et al., 2002, Lambert et al., 2004; Lambert, Hogan, Camp, & Ventura, 2006), and have a heightened risk of developing Post-Traumatic Stress Disorder or PTSD (Jaegers et al., 2019, Spinaris et al., 2012).

Researchers have identified some of the factors that threaten officer well-being and safety (Cullen et al., 1985; Paoline et al., 2015; Steiner and Wooldredge, 2015, Steiner and Wooldredge, 2017), but most of this information is based on samples of correctional officers working in prisons rather than jails (Butler, Tasca, Zhang, & Carpenter, 2019). A non-trivial portion of correctional officers (i.e., about one-third) work in jails, where population turnover and a lack of resources and control mechanisms produce a chaotic environment fraught with safety and wellness risks (Lambert, Gordon, Paoline III, & Hogan, 2018). A recent study, for example, showed that 16.4% of jail officers reported being assaulted by inmates in the past year; one-half (55%) reported being threatened in the past three months and most of those officers were threatened two or more times (Ellison & Gainey, 2020). Due in part to the threat of victimization and exposure to violence and human suffering (e.g., inmate-staff assaults, attempted suicides, deaths, disease), as well as the demands, lack of control, and isolating nature of their work environment, jail officers' jobs are extremely stressful. Given the expanding role of these facilities in local, state, and federal correctional systems (Caudill et al., 2014; Irwin, 1985), understanding how these work environments affect officers is paramount.

To expand the small body of literature regarding safety and wellness among jail staff (Applegate & Paoline, 2007; Ellison & Gainey, 2020; Lambert et al., 2004, Lambert et al., 2018; Lambert & Paoline, 2005), we examine the factors that affect work stress among a sample of officers working in 19 jails. Guided by the Job Demand-Control-Support model utilized in parallel research among prison officers (i.e., see Johnson & Hall, 1988; Steiner & Wooldredge, 2015), we examine whether increased job demands, less job control or support, or safety risks increase levels of work stress among jail officers. The potential implications of our findings are then discussed within the context of the small but growing body of literature on jail officer safety and wellness.

A model of work stress for jail officers

In 2017, approximately 226,000 staff members in more than 3000 jails were supervising about 750,000 individuals accused of crimes, awaiting sentencing, serving a short sentence (i.e., typically 1–2 years but longer in recent years), or awaiting transfer to state and federal prisons (Ellison, Steiner, & Wright, 2018; Zeng, 2019). Accused criminals ranging from vagrants to murderers or probation and parole violators are admitted to local jails—a process that plays out over 10.5 million times per year. Many individuals arrive to jail under the influence of drugs or alcohol, off their prescribed psychotropic medication, and/or defiant towards authority figures (Irwin, 1985), and have enduring physical and mental health deficits (e.g., symptoms of drug/alcohol withdrawal) that complicate their supervision and management. Moreover, because individuals are only incarcerated for an average of 25 days and many jails are overcrowded, underfunded, and understaffed, services and programming options such as needs-based placements in detoxification, work-release, or trauma-informed units (e.g., Veterans modules) may be full or unavailable altogether (Ellison et al., 2018). Static risk assessments (e.g., based on criminal history or offense type) classify inmates into custody levels but do not address inmates' behavioral issues or proclivities. Even when regularly updated (i.e., jails vary greatly in data management practices), data are not readily available to officers and they cannot absorb this information across the hundreds if not thousands of inmates they supervise (Lambert et al., 2018). In short, jail officers supervise a diverse array of criminal and/or antisocial personalities under situational and institutional conditions that limit remunerative controls or compliance mechanisms (see Huebner, 2003). Thus, despite supervising (on average) less serious offenders than their prison counterparts, jail officers work under incredibly chaotic and unpredictable conditions that lead to work stress (Ellison, 2017; Ellison & Gainey, 2020; Lambert et al., 2018).

Arising from exposure to adverse stimuli or experiences that cause extreme arousal, stress is defined as psychological discomfort or tension that disturbs an individual's equilibrium and prevents their ability to cope, manage, and/or respond to life experiences or circumstances (Cullen et al., 1985; Gerrig & Zimbardo, 2002; Steiner & Wooldredge, 2015). By extension, work stress can be considered a product of exposure to job-related adversities (e.g., exposure to violence) or occupational demands (e.g., constant vigilance) that exceed available coping resources (Schaufeli & Peeters, 2000). Among correctional officers, work stress may manifest itself as frustration, anger, anxiety, and/or extreme worry (Cullen et al., 1985; Steiner & Wooldredge, 2015), all of which may adversely affect officers both in and outside of work. Officer stress has been linked to a number of negative outcomes such as mental health problems, job dissatisfaction, drug and alcohol abuse, substandard job performance, and suicide risk (Finn, 2000; Jaegers et al., 2019; Lambert and Paoline, 2005, Lambert and Paoline, 2012; Paoline & Lambert, 2012; Spinaris et al., 2012). In addition, stressed officers may begin to neglect aspects of their job that are essential to their success, including seemingly trivial activities such as settling inmate requests in a timely fashion, or more essential facets of their job such as maintaining open lines of communication with coworkers or inmates (Ellison & Gainey, 2020; Steiner & Wooldredge, 2015). Separated from supports that are otherwise essential for preserving workplace safety and navigating or coping with difficulties inherent to their work environment (Cullen et al., 1985; Schaufeli & Peeters, 2000; Steiner and Wooldredge, 2015, Steiner and Wooldredge, 2017), stressed officers may instead rely on and perhaps overburden family members and friends to help them deal with workplace adversities (Armstrong et al., 2015).

Researchers have utilized the Job Demand-Control-Support model of occupational health to predict work stress among prison officers (Johnson & Hall, 1988), and posited that officers will exhibit greater levels of work stress when they experience more job demands, less control over their work environment, and/or less support from coworkers and supervisors (Steiner & Wooldredge, 2015). Similar job duties and related adversities likely exert similar pressures over jail and prison officers, but the jail environment may also produce unique stressors. Physical victimization and vicarious exposure to violence (e.g., assaults), for example, may be less prevalent because jail populations typically contain fewer violent offenders than prisons (Irwin, 1985), yet a substantial proportion of jail officers still believe their jobs are especially dangerous (Ellison & Gainey, 2020; Lambert et al., 2018). Moreover, safety risks and other adversities inherent to prison work (e.g., greater job demands or less job control and/or support at work) may lead to greater job-stress and adversely affect job satisfaction and performance (Steiner & Wooldredge, 2015), as well as create work-family conflict (Armstrong et al., 2015).

Job demands

Workload, time pressures, and job-related demands have emerged as relevant influences on work-related stress among correctional officers (Dollard & Winefield, 1998; Hausser, Mojzisch, Niesel, & Schulz-Hardt, 2010; Karasek, 1979; Steiner & Wooldredge, 2015; Van Der Doef & Maes, 1999). Officers are tasked with the responsibility of preserving not only their own safety and welfare, but that of inmates, their colleagues, and civilians who work in or visit the facility (e.g., lawyers, doctors, nurses, social workers, researchers). In addition, officers are simultaneously expected to juggle inmate requests, deescalate conflicts, manage inmate movement, assist civilians with their duties (e.g., medical staff with sick or pill call), and lead or support other officers and supervisors. Researchers have found that correctional officers who experience greater job demands and workload pressures may exhibit higher levels of work stress and prison officers who perceive insufficient staffing or time to complete their job duties have greater levels of stress (Dollard & Winefield, 1998; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015).

High incarceration rates, particularly in jails, have arguably only added to officers' workloads and time pressures because officers are asked to achieve their job-related tasks (i.e., maintain of supervision, order, and safety) across a greater number of individuals without corresponding increases in support or assistance (Beck, 2000; Finn, 2000; Lambert et al., 2018; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015; Stephan, 1997, Stephan, 2008). Hiring practices have simply not kept pace with the burgeoning correctional population, and the number of open law enforcement and corrections positions had reached all-time highs prior to the COVID-19 pandemic (Hyland, 2018). Despite some reductions in state prison populations (e.g., New York), jail populations have decreased more slowly or remained steady (Ellison et al., 2018; Kaeble & Cowhig, 2018). The glut of open positions in some jails coupled with the steady number of inmates has created a precarious situation where administrators and supervisors must maintain supervision and safety without adequate resources and coverage. Supervisors experience greater job demands, pressures, and stress due to their level of responsibility relative to line staff (i.e., Blau, Light, & Chamlin, 1986; Lambert, Hogan, & Griffin, 2007; Schaufeli & Peeters, 2000), and with fewer line staff, they assume even more responsibility to make do with who they have available.

Officers' ambiguous or conflicting job responsibilities or expectations (i.e., role problems) have also been known to increase work stress (Cullen et al., 1985; Schaufeli & Peeters, 2000). Correctional officers experience role problems because they balance the custodial need to maintain safety and security with the human service approach that emphasizes modicums of freedom and responsibility for inmates (Steiner & Wooldredge, 2015). Moreover, institutional policies, formal training procedures, and on-the-job socialization do little to alleviate officers' role conflict and complicate their professional role, particularly if they receive conflicting direction from two or more sources (Cullen et al., 1985; Lambert & Paoline, 2005; Schaufeli & Peeters, 2000). Without a clear path for balancing safety and security with rehabilitative goals, officers tend to vary substantially in their approaches to inmate management; some officers alleviate role conflict by using their discretion to manage inmates, while others rely on formal regulations and procedures (Liebling, Price, & Shefer, 2011). Regardless of how they go about their job, prison officers who experience ambiguous and/or conflicting job demands tend to experience more work stress (Cullen et al., 1985; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015).

The extent to which job demands affect levels of work stress among jail officers remains largely unknown, though recent trends in jail and prison operations warrant mention. For example, policy initiatives in states such as California (i.e., Public Safety Realignment Act) have reduced prison overcrowding by sentencing low-level felony offenders to local jails instead of prison. These initiatives have arguably reduced prison populations but increased levels of violence in jails (Caudill et al., 2014; Lofstrom & Martin, 2015). In addition, burgeoning jail populations have likely increased officers' job demands (i.e., workload and time pressures) by forcing them to interact with more criminally involved or sophisticated individuals, and by further limiting the availability of privileges and programming options that provide incentives for rule compliance (e.g., work assignments). On the other hand, the lack of inmate activity may provide officers with the ability to focus primarily on security and control (i.e., the primary goals in jail), and cause them fewer role problems because security responsibilities are less conflicted by the competing goal of rehabilitation more commonly emphasized in prison environments.

Control

Individuals with greater job control persevere through the difficulties of job demands by exercising their authority to make work-related decisions and utilizing their personal skillset to accomplish job-related tasks (Hausser et al., 2010; Karasek, 1979; Van Der Doef & Maes, 1999). Officially, correctional officers have less occupational autonomy because rigid institutional rules and policies stipulate how officers should carry out their job duties and manage inmates (Dollard & Winefield, 1998; Hepburn, 1985; Jacobs, 1977). Yet institutional policies do not guide officers in every situation and strict adherence to rules would create disorder and a lack of control by irking inmates and increasing defiance of rules and directives (Liebling, 2004; Liebling et al., 2011; Lombardo, 1989). Instead, compliance and control (i.e., when inmates adhere to jail rules and officers' commands) is achieved through discretionary decisions that decrease officer stress and balance security needs with inmates' desires for modicums of freedom (Dollard & Winefield, 1998; Slate & Vogel, 1997; Steiner & Wooldredge, 2015; Wright, Saylor, Gilman, & Camp, 1997).

Jail officers who are more successful in eliciting inmate compliance might also find their jobs less stressful, but it is important to recognize that inmates' deficits and short average stays may complicate officers' management and control efforts. Irwin (1985) observed that many jail inmates are detached from conventional society and some teeter on the edge of social and physical existence; officers' routine exposure to this degree of human depravity and despair may affect their perceptions of inmates, make them more cynical, and limit the existence of shared values and morals that could aid officers in generating rule compliance. In addition, short periods of incarceration and the corresponding lack of programming deprives officers of remunerative controls that indirectly influence inmate behavior and generate incentives for compliance (e.g., Ellison & Gainey, 2020). Further, the transitory nature of most jail populations interrupts any degree of social cohesion that might develop between officers and inmates; with only 25 days served (on average), there is little opportunity for jail officers to get to know inmates, their backgrounds and proclivities, or their desires, all of which may aid officers in their control efforts.

Support

Officers may successfully navigate a demanding and relatively unmanageable work environment so long as they possess supportive relationships with their colleagues and/or supervisors (Cullen et al., 1985). Coworkers and supervisors provide officers with affective or instrumental assistance and resource options that help them cope with the difficulty of their job demands, and these resources may insulate officers from, or help them deal with, workplace adversities (Cohen & Wills, 1985; Cullen et al., 1985; La Rocco & Jones, 1978; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015). Researchers have linked coworker support, supervisor support, and a constructive work environment with greater well-being among prison officers (Cheek & Miller, 1983; Dollard & Winefield, 1998; Poole & Regoli, 1980; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015; Van Der Doef & Maes, 1999), as well as jail officers (Lambert et al., 2018; Lambert & Paoline, 2005).

Emotional support and assistance from sources outside of work may also be critical for alleviating work-related stress among correctional officers. Many officers do not share stressful aspects of their jobs or symptoms related to stress with their coworkers and supervisors for fear that they will be treated differently or experience some sort of reprisal (Armstrong et al., 2015). Rather than relying on their colleagues for support, officers may rely on family members to cope with the difficulties of their jobs and manage work-related stress (Cullen et al., 1985; Triplett et al., 1999). That is, emotional support from family members may provide officers with coping resources that are simply not available in the workplace, and a supportive family or significant other may be vital for reducing stress both in and outside of work. Researchers have yet to examine whether social supports at home alleviate work-related stress for jail officers.

Researchers have also noted that regular contact with inmates may result in more fatigue and burnout (Dollard & Winefield, 1998; Whitehead, 1989), rendering officers more stressed at work (Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015). On the other hand, less social distance between officers and inmates may be beneficial for supervision, keep officers busy, and allow them to pass more easily. Thus, both the quantity and quality of inmate contact might be important for predicting officers' stress levels, yet researchers have only focused on the former.

Workplace safety and exposure to violence

Direct and vicarious exposure to violence is an unfortunate reality for many correctional officers (Konda et al., 2012; Spinaris et al., 2012), and these experiences may lead to adverse health and occupational outcomes such as work stress, turnover, and absenteeism (e.g., see Cullen et al., 1985; Lambert & Paoline, 2005; Schaufeli & Peeters, 2000; Spinaris et al., 2012; Stohr, Lovrich Jr, Menke, & Zupan, 1994). Approximately one-in-five correctional officers report being assaulted during the previous year, and a majority say they are threatened by inmates on a regular basis and/or agree that their workplaces are especially dangerous (Ellison, 2017; Ellison & Gainey, 2020; Lambert et al., 2018; Steiner and Wooldredge, 2015, Steiner and Wooldredge, 2017). Additionally, officers experience secondary trauma and remain in a state of hypervigilance—a symptom of stress—when they are exposed to violent situations (Spinaris et al., 2012). Officers who directly observe violent incidents among inmates (e.g., suicides, fights), or witness their colleagues enduring victimization likely perceive a greater level of disorder and experience more stress (Cullen et al., 1985; Lombardo, 1989; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015). Taken together, officers likely experience symptoms of work stress resulting from exposure to violence, regardless of whether officers were directly victimized or merely witnessed such safety risks.

Jail-level influences on officer work stress

Environmental factors of some prisons and jails may influence the collective well-being of the correctional officers who work there, and these facility conditions may adversely affect officers' stress levels by increasing their job demands, limiting their job control, and/or placing them in more dangerous situations (Bierie, 2012; Blau et al., 1986; Steiner & Wooldredge, 2015; Wright et al., 1997). Larger and more crowded prisons, for example, increase officers' workloads and may overextend their capacities as managers and supervisors of the inmate population, especially in facilities where staff coverage has not kept pace with increases in the number of individuals incarcerated (Finn, 2000; Lambert et al., 2018; Steiner and Wooldredge, 2009, Steiner and Wooldredge, 2015). Large, crowded, and/or understaffed facilities also create greater uncertainty for officers and limit opportunities for the development of supportive relationships with supervisors, other officers, and inmates (Liebling, 2004). Thus, larger and more crowded jails create greater job demands and all but eliminate most aspects of control and social support that the workforce relies on to limit work-related stress (Steiner & Wooldredge, 2015). In addition, officers in overcrowded and understaffed facilities typically have mandatory overtime, which reduces time with external social support circles. Population size, crowding, and/or the ratio of officers to inmates have never been considered relative to officers' stress levels across jails.

Higher levels of inmate violence—particularly assaults on officers—might also contribute to greater levels of stress for correctional officers (Steiner & Wooldredge, 2015). A high level of inmate violence may increase levels of work-related stress because officers feel their safety is more at risk in these facilities, and levels of disorder, rule-breaking, and violence (i.e., safety or the lack thereof) are often used measures of officer performance (Cullen et al., 1985; Lambert & Paoline, 2005; Lombardo, 1989; Schaufeli & Peeters, 2000; Spector, Coulter, Stockwell, & Matz, 2007; Sykes, 1958). Levels of violence may be lower in jails compared to prisons owing to lower proportions of violent offenders among their respective inmate populations, but jails with a higher level of inmate violence should still cause officers more stress than safer jails due to the level of disorder and lack of control that may characterize these workplaces (Ellison et al., 2018; Steiner & Wooldredge, 2015). However, like much of the officer-level factors that affect work stress, the jail-level correlates of work stress across facilities remain largely unexplored.

Current study

Correctional officers are exposed to many situations that may lead to health and safety risks such as stress and victimization, and researchers have sought to examine the antecedents and consequences of these work-related experiences. This study is focused on the potential factors that may influence work-related stress among jail officers, who have received much less empirical attention than their prison counterparts despite working a similar if not more stressful environment. Based on the Job Demand-Control-Support model, and data collected from a large sample of officers working in 19 jails and 10 agencies, we examined the following hypotheses:

Jail officers with greater job demands will experience more work stress.

Greater control over the work environment will reduce work stress among officers.

Officers who perceive greater support in and out of the workplace will experience less work stress.

Officers who are directly victimized or exposed to violence more often will experience more work stress.

Across jails, levels of work stress will be greater in facilities where order and safety are more difficult to maintain.

Method

Data regarding officers' backgrounds, perceptions of their work environment and support from coworkers and family and so forth were collected at 19 jails operated by 10 agencies during a two-year period from 2017 to 2019. One facility was located in a Midwestern state while the remainder of the participating jails were in a large Mid-Atlantic state, and facilities covered urban, suburban, and rural jurisdictions. Depending on jail capabilities, officers were surveyed in person via paper and pencil or via the internet using a web-based survey software, and every officer that was working and/or on the payroll during the data collection period was given an opportunity to fill out a survey. For paper and pencil delivery, surveys were delivered to participants during morning and afternoon roll calls and then collected after officers' shifts; this process was then repeated for two to three consecutive days at each jail. Initial response rates ranged from 71 to 98% across facilities. Alternatively, some jails provided officers with email addresses, and electronic survey requests were sent to officers with one-, three-, and seven-week reminders (Dillman, Smyth, & Christian, 2014). Response rates were lower in facilities where surveys were delivered electronically (49–83%), in part because many officers did not regularly check their email and/or could not or did not access email outside of work, and because lists of employee emails were not always updated prior to being shared with researchers (i.e., online sampling frames sometimes included those who quit, were terminated, or otherwise had not worked in months).

Participants

Data collection initially yielded 1464 responses, or a 71% response rate relative to the target population of 2071 jail officers after excluding 181 officers who jail administrators identified as unreachable (e.g., officers out sick, on FMLA leave, or vacation). An additional 84 responses were later determined unusable because officers had failed to answer questions related to the study outcomes, provided missing or incomplete responses to demographic questions, or otherwise failed to complete large swaths of the questionnaire relevant to this study. After excluding incomplete responses listwise, the sample was reduced to 1380 officers (usable response rate = 67%). Any additional missing data on the measures utilized in this study (<1% of cases) were imputed with mean or modal responses from officers in the same facility, and any missing data on scale items were replaced with the standardized mean of zero during scale construction. Outcomes and predictor variables were compared before and after cases were deleted and adjustments were made for missing data; no significant differences were revealed.

Measures

The officer-level measures used for the study were derived from survey items that captured officers' perceptions of how stressful their job was, as well as aspects relevant to the potential development of work stress such as officers' job demands, level of job control, degree of social support, and exposure to safety risks at work. Jail measures were obtained via official reports provided by facility administrators. The measures included in the final models are described in Table 1 .

Table 1

Description of study variables.

	Mean	(SD)	Range
Outcome
Work stress	0.00	(0.77)	−2.18–1.59
Individual-level variables
Job Demands
Perceived adequate staff to maintain safety and security	0.30	(0.46)	0–1
Perceived adequate time to meet job demands	0.57	(0.49)	0–1
Perceived unreasonable workload	0.41	(0.49)	0–1
Role problems	0.00	(0.78)	−1.76–1.83
Supervisor	0.28	(0.45)	0–1
Job control
Perceived order	0.00	(0.88)	−2.72–1.57
Autonomy	0.00	(0.81)	−2.12–1.88
Perceived cooperation from inmates	0.00	(0.83)	−2.04–1.90
Support
Coworker support	0.00	(0.73)	−2.30–1.69
Supervisor support	0.00	(0.84)	−2.54–1.30
Family support	0.00	(0.85)	−2.00–1.22
Natural log inmates talked with per shift	3.47	(1.19)	0–6.91
Safety
Assaulted	0.16	(0.37)	0–1
Natural log times threatened by inmates	0.92	(1.08)	0–4.34
Natural log times witnessed violence	1.66	(1.14)	0–4.26
Perceived dangerousness	0.00	(0.83)	−1.80–1.78
Age	36.79	(11.20)	18–67
Male	0.66	(0.47)	0–1
White	0.68	(0.47)	0–1
N1 = 1380
Level-2: Jails
Average daily population	590.84	(424.68)	50–1465
Crowding	1.15	(0.53)	0.48–2.67
Ratio of inmates to officers	4.82	(1.24)	2.50–6.83
Rate of violence committed per 100 inmates	110.94	(62.79)	19–219.49
N2 = 19

The outcome examined in this study—work stress— was a six-item scale derived from prior studies of work stress among correctional officers (e.g., Cullen et al., 1985; Lambert & Paoline, 2005; Steiner & Wooldredge, 2015). Officers were asked about their level of agreement with statements such as “A lot of times my job makes me frustrated or angry,” “I am usually under a lot of pressure at work,” and “When I am at work I often feel tense or uptight” (responses were strongly disagree, disagree, agree, and strongly agree) (α = 0.87). Items selected for the work stress scale, as well as the remainder of the scales used in this study can be viewed in the Appendix. Items were confirmed via principle axis factoring (Eigen values <1.0) and scales were created by summing standardized scores of scale items and dividing by the number of items in each scale.

The officer-level predictors relevant to work stress included officer demands (perceived adequacy of staff to maintain safety and security or of time to complete job tasks, unreasonable workload, role problems, supervisor), control (perceived order, job autonomy, and inmate cooperation), support (coworker support, supervisor support, family support, natural log of the number of inmates talked with per shift), and safety (victim of assault, natural log of the times officers were threatened, natural log of times witnessed inmate violence towards inmates and staff, and perceived dangerousness). In addition, the following analyses included controls for age, sex (male), and race (white). Facility-level measures relevant to predicting levels of work stress across jails included average daily population, crowding, ratio of inmates to officers, and rate of assaults committed per 100 inmates in the population.

Although the operationalization and distribution of many variables is clear by viewing Table 1, some measures may require a more detailed explanation. Staffing, time, and workload pressures were each captured with measures that assessed whether officers strongly agreed or agreed with the statements “There is enough staff to maintain the safety and security of this facility,” “There is enough time to perform all of the tasks required of me on my shift,” and “The amount of work on my shift is unreasonable,” respectively (see Steiner & Wooldredge, 2015). Based on studies of the correlates of work stress (Cullen et al., 1985; Lambert, Altheimer, & Hogan, 2010; Lambert & Paoline, 2005; Poole & Regoli, 1980; Rizzo, House, & Lirtzman, 1970), role problems (α =0.85) was a five item scale that assessed officers' level of agreement (i.e., strongly agree through strongly disagree) with statements such as “I work on unnecessary projects or tasks on the job.” Based on scales adopted from the work of Lambert et al. (2007) and Steiner and Wooldredge (2015), autonomy was measured with a scale comprised of three survey items that asked officers to indicate the degree to which they agreed with statements such as “I have a great deal of freedom as to how I do my job.” (α = 0.74). Similarly, coworker, supervisor, and family support (α = 0.71, α = 0.86, & α = 0.87, respectively) were four item scales that asked officers about their level of agreement with statements such as “My immediate supervisor is helpful in getting me what I need to do my job effectively,” “I generally receive help from my coworkers when I ask for it,” and “My family helps me relax when I am worried about work.” Perceived order was captured with a two-item scale that measured how much they agreed with the statements “I rarely have problems on my shift,” and “Inmates generally follow the rules during my shift” (α = 0.71). Derived from studies of cooperation and normative compliance (Nix, 2017; Sunshine & Tyler, 2003; Tyler, 1990, Tyler, 2010), perceived cooperation from inmates was a four-item scale that assessed officers' level of agreement with statements like “Inmates help me identify problems in my work area” (α = 0.77).

Based on studies of safety among correctional officers working in prison (e.g., Steiner & Wooldredge, 2017), individuals were asked whether they were physically assaulted in the past year, how many times inmates had threatened to hurt them in the past 3 months, and how many times they witnessed inmate violence (i.e., including fights between inmates and self-injurious behavior) and saw other staff members be assaulted in the past year. Perceived danger (see Ellison & Gainey, 2020; Lambert et al., 2018; Gordon, Moriarty, & Grant, 2003) was a scale comprised of five items that asked officers how likely it was that inmates would threaten their safety, threaten their family's safety, or injure them while they broke up fights, walked around the housing unit, or escorted inmates to and/or from housing units (responses were very unlikely, unlikely, likely, very likely) (α = 0.88). The natural logs of inmates talked with per shift, as well as the number of times officers were threatened or exposed to violent incidents involving inmates or other staff members was taken in order to reduce the skew in officers' responses.1 Non-white and female officers were treated as reference groups for race and sex, respectively. All independent variables were assessed for (multi)collinearity, which was not a significant problem here. In these models, the lowest tolerance level was 0.52. and the highest VIF = 1.91, suggesting little or no evidence of a multicollinearity problem (i.e., Allison, 1999 recommended exercising caution with tolerance levels less than 0.40 or VIFs greater than 2.50).

Pertaining to the jail-level correlates of work stress, the ratio of inmates to officers was calculated by dividing the average daily population of each jail by the number of employed officers, while crowding was created by dividing average daily population by design capacity (Steiner & Wooldredge, 2009). Rate of observed inmate violence, an aggregate measure of officers' responses to the facility level, was the total number of times respondents witnessed different types of violence (i.e., an inmate physically assault a staff member, assault an inmate, or engage in self-harm or attempt suicide in the past year), divided by the average daily population of the facility and multiplied by 100.

Analysis

The hierarchical structure of the data (i.e., officers nested within jails) suggests that a multilevel modeling technique is appropriate in order to: 1) separate individual- and facility-level hypotheses and analyses, 2) adjust for the correlated error across officers working in the same jail, and 3) remove between-jail variation in officer characteristics that might correspond with differences in levels of work stress across jails. Work stress—a ratio outcome—was examined with a traditional hierarchical linear model. Analyses began by first testing whether unconditional models (with no predictors) detected significant variation in work stress at level-1 (i.e., officer) and level-2 (i.e., jail), justifying the use of a multilevel model (p ≤ .05). All level-1 measures were group-mean centered to remove any similarities among the respective officer populations (i.e., in each jail) that might contribute to variation in levels of work stress across facilities (Raudenbush & Bryk, 2002). By group mean centering all predictors, the results of the officer-level analyses are independent of facility-level influences. Second, random effects models with all officer-level measures were estimated. Relationships that did not vary across jails (p ≤ .05) were fixed. Finally, jail-level predictors were added to the final models at level-2. Facility-level variables were added one at a time in order to ensure model stability and identify any findings that could be potentially misleading due to the limited number of jails in the study sample.

Findings

The analysis of job demands, control, support, and safety variables on levels of work stress among jail officers uncovered a number of interesting findings (Table 2 ). Before identifying the results of the multilevel analysis, however, it is worth noting a few patterns across the items in the work stress scale. Nearly 60% (n = 791) of officers agreed or strongly agreed with the statement “A lot of times my job makes me very frustrated or angry,” for example, and nearly as many (n = 770) said they were often tense or uptight at work. Sixty-five percent of jail officers (n = 899) felt they were usually under a lot of pressure at work, and 761 officers (about 55%) agreed that there were a lot of aspects of their job that make them upset. Yet some officers were clearly not as stressed; 660 agreed that they were usually calm and at ease when working and nearly a quarter (23%) felt that most of the time they did not have much to worry about at work. Thus, there is clearly variation in job stress among jail officers; some officers felt their jobs routinely made them upset, angry, uptight, or frustrated, yet others were able to navigate these stressors and feel relatively at ease when working.

Table 2

Effects of officers' job demands, control, support, and safety on work stress.

	b	(SE)
Intercept	−0.01	(0.05)
Job Demands
Perceived adequate staff to maintain safety and security	−0.09*	(0.04)
Perceived adequate time to meet job demands	−0.07	(0.06)
Perceived unreasonable workload	0.34**	(0.06)
Role problems	0.20**	(0.03)
Supervisor	0.03	(0.04)
Job control
Perceived order	−0.09**	(0.02)
Autonomy	−0.13**	(0.02)
Perceived cooperation from inmates	−0.003	(0.02)
Support
Coworker support	−0.08**	(0.02)
Supervisor support	−0.06**	(0.02)
Family support	−0.06**	(0.02)
Natural log inmates talked with per shift	−0.02	(0.01)
Safety
Assaulted	0.01	(0.04)
Natural log times threatened by inmates	0.04*	(0.02)
Natural log times witnessed violence	0.04*	(0.02)
Perceived dangerousness	0.18**	(0.02)
Age	−0.001	(0.001)
Male	−0.13**	(0.03)
White	0.02	(0.04)
Proportion variation within jails	0.89
Proportion variation within jails explained	0.49

Notes: **p < .01. *p < .05. Unstandardized coefficients reported with standard errors in parentheses. Italicized coefficients indicate relationship varies across jails (p < .05).

Officer-level effects on work stress

Officers experienced greater levels of work stress when their jobs required more of them, or in other words, they experienced greater job demands (Table 2). For example, officers who felt there were insufficient staff to maintain safety and security experienced greater levels of work stress than those who felt there was adequate staff. More directly tied to officers' job demands, those who thought they had an unreasonable workload experienced a higher level of stress than individuals who felt their workload was more reasonable. In addition, officers who experienced conflicting job responsibilities or expectations—greater role problems—had a higher level of stress. The perception of enough time to meet job demands and a supervisor role had no effect on officers' work stress levels, however.

Jail officers also experienced greater levels of work stress when their work environments were more disorderly or they had less control at work. Officers who perceived their work environment relatively free of inmate rule violations and issues had lower levels of work stress than individuals who felt work was more disorderly. In addition, jail officers who had more control or autonomy at work experienced lower levels of job-related stress. The perceived level of cooperation officers received from inmates did not affect their level of work stress.

Regarding the relationship between officer support and work stress, higher levels of support were associated with lower levels of stress. Officers who felt they had more coworker, supervisor, and family support had lower work stress levels than individuals with less coworker, supervisor, or family support, respectively. The number of inmates talked with per shift had no effect on officers' job stress levels, however.

Officers' experiences with, and perceptions of safety related incidents were also related to their work stress levels. Officers who were threatened more often by inmates experienced a higher level of stress than their colleagues who were threatened less. In addition, officers who witnessed more inmate-inmate and inmate-staff assaults had greater levels of job stress, as did those who perceived their work environment to be dangerous as opposed to safe. Job stress levels were no different among jail officers who were victimized by inmates compared to those who were not assaulted.

Finally, regarding the controls for officers' demographic characteristics, we found that male officers tended to be less stressed than female officers. White and non-white officers had similar levels of stress, however, as did officers of different ages. Together, the officer-level predictors explained nearly half (49%) of the variation in work-stress within jails.

Jail-level effects on work stress levels across facilities

Pertaining to the potential sources of work stress across facilities (Table 3 ), jails that had more inmates per officer had higher levels of work stress among their respective officer workforces. In addition, officers were more stressed in jails that had higher rates of inmate violence. The average daily population and crowding level of jails were not significantly related to the level of officer work stress between jails. The jail-level predictors explained 35% of the variation in officer work stress between facilities.

Table 3

Environmental effects on work stress levels across jails.

	b	(se)
Intercept	−0.01	(0.05)
Average daily population	−0.0001	(0.0001)
Crowding	−0.07	(0.09)
Ratio of inmates to officers	0.09*	(0.04)
Rate of violence committed per 100 inmates	0.001t	(0.001)
Proportion variation within jails	0.11
Proportion variation within jails explained	0.35

Notes: **p < .01. *p < .05 tp < .10. Unstandardized coefficients reported with standard errors in parentheses. Models include all of the level-1 effects shown in Table 2.

Discussion

The purpose of this study was to enhance what is known about the correlates of stress among correctional officers working in jails or facilities designed for short-term confinement. While much is known about the correlates and consequences of work stress among prison officers, much less attention has been paid to the origins and outcomes of work stress for officers working in local jails and other facilities where inmates often stay for relatively brief periods of time. This inquiry is especially relevant because one-third of correctional officers work in jails (Applegate & Paoline, 2007; Ellison, 2017; Lambert & Paoline, 2008), and the chaotic, transitory, and ever-changing nature of jail populations may create unique stressors and safety risks for correctional officers (Caudill et al., 2014; Irwin, 1985; Lambert et al., 2018). Given the distinctions between jail and prison (cf., Irwin; Jacobs, 1977; Marquart and Crouch, 1984, Marquart and Crouch, 1985), we posited that some circumstances surrounding work stress among jail officers may be unique, but also noted the parallel job duties and related experiences that may result in similar stressors among jail and prison staff. In general, our findings largely comported with researchers' findings pertaining to work stress among samples of prison officers (Steiner & Wooldredge, 2015).

Based on our guiding theoretical foundation and what was known about the correlates of work stress among correctional officers, we found that officers' stress levels were tied to their demand-, control-, support-, and safety-related experiences and perceptions at work, and that officers' stress levels varied across jails depending on facility characteristics or conditions that threatened order and security. Across a diverse sample of jail security staff, we found that perceptions of a more demanding work environment—insufficient staff to maintain safety and security, an unreasonable workload, or greater role problems and conflict at work—were associated with higher levels of work stress (Hypothesis 1). Positive experiences maintaining order and greater occupational liberty, on the other hand, were related to lower levels of work stress (Hypothesis 2), as was greater support both in and outside of work (Hypothesis 3). Exposure to violence and more dangerous work conditions were related to greater work stress (Hypothesis 4), and across facilities, levels of work stress were collectively higher among officers working in jails where control and safety were more difficult to achieve (Hypothesis 5).

These findings hold implications for the examination of work stress in correctional environments, imply strong support for the Job-Demand-Control-Support model of stress among jail officers, and closely aligned with findings from parallel studies of correctional officers working in prisons (Steiner & Wooldredge, 2015). Excessive job demands and negative perceptions of policies resulted in more stress for jail officers. Examples of how excessive job demands occur include staffing shortages that result in more operational tasks per officer and unreasonable workloads that force officers to neglect some aspects of their job in the interest of maintaining the vital concerns of safety and security. Poor guidance from administrators, such as when officers felt they had more ambiguous or conflicting responsibilities (i.e., experienced greater role problems), produced more stress for officers. This also means that officers with positive perceptions of their administration's expectations regarding policies for officers and clarity regarding job responsibilities reported less work stress and were more comfortable at work, a critical attribute of a successful correctional officer. Officers who perceived a lack of time to meet job demands and officers who were supervisors had similar levels of stress compared to their counterparts, suggesting that the importance of job demands superseded time restraints or level of authority in terms of job stress.

Our findings regarding officers' perceptions of their ability to maintain control over their work environment as well as officers' perceptions of rule compliance and disorder, collectively, allude to the importance of inmate culture within facilities. Extrapolating these findings suggests that when inmates follow institutional rules and officers have fewer problems on their shift, they tend to encounter less stress. Officers probably receive very little voluntary assistance from inmates and thus cooperation has a negligible effect on stress, but when inmates create fewer issues on their shift, their work environment is less chaotic and tense. In addition, although officers are fairly limited in how they go about their jobs due to rigid institutional rules and policies (Dollard & Winefield, 1998; Hepburn, 1985; Jacobs, 1977), those who felt they retained a fair amount of discretion were less stressed. Given that institutional rules and policies provide little guidance for handling every situation (Liebling, 2004), officers utilize their discretion and personal skillset to accomplish job-related tasks (Hausser et al., 2010; Karasek, 1979; Van Der Doef & Maes, 1999), and modicums of freedom likely allow greater flexibility to adapt to all work-related situations (Steiner & Wooldredge, 2015). Having the discretion to address unique situations allows officers to maintain order in a way they see fit, which could reduce stress.

Exposure to violence and more dangerous work conditions were also associated with more work stress among jail officers. Specifically, officers who were threatened more often and who witnessed greater numbers of violent incidents exhibited more work stress, as did those who felt their personal safety was more at risk. Repeated direct victimization (i.e., being threatened) and exposure to violence (i.e., witnessing inmate-inmate violence, inmate-staff violence, or self-harm/suicide) may cause officers to remain in a state of hypervigilance and exhibit symptoms of work stress that may eventually manifest into symptoms of post-traumatic stress disorder or PTSD (Spinaris et al., 2012). Curiously, however, officers who were assaulted in the past year had similar levels of stress compared to officers who were not victimized; it is possible that the constant threat of violence is actually more psychologically stressful than victimization, which rarely occurs and therefore may not have the variation necessary to influence work stress (Ellison & Gainey, 2020). Officers who felt it was likely that they would be injured while carrying out job duties, meanwhile, experienced more work stress. A higher frequency of threats, exposure to violent incidents, and the likelihood of being injured by inmates may also signify greater disorder and a lack of control in the workplace, and because officers are often evaluated by the extent to which these incidents occur, they may experience greater work stress when violence is more commonplace (Steiner and Wooldredge, 2015, Steiner and Wooldredge, 2017). It is important to note here, however, that many officers who completed surveys did not report being assaulted; only 16% reported a recent assault in the past year. This suggests that while the jail environment may be stressful and the threat of violence ever present, inmate-on-officer assaults is not commonplace across all experiences. Much like the vast distinctions between Irwin's (1985) jail inmate culture and Skarbek's (2014) jail inmate culture, the heterogeneity of jail cultures across institutions likely influences the level of violence and the perceived threat of violence.

Officers who maintained supportive relationships with coworkers, supervisors, and family members were likely insulated from developing stress because they possess the coping mechanisms for dealing with workplace stressors. Our findings align with previous research on the role of social support as a protective factor for officers supervising the incarcerated (Cheek & Miller, 1983; Dollard & Winefield, 1998; Lambert et al., 2018; Lambert & Paoline, 2005; Poole & Regoli, 1980; Schaufeli & Peeters, 2000; Steiner & Wooldredge, 2015; Van Der Doef & Maes, 1999). Coworkers and supervisors are substantial resources for officers as they wrestle with the difficulty of their job demands (Cohen & Wills, 1985; Cullen et al., 1985; La Rocco & Jones, 1978), and these types of supports may insulate officers from work stress by providing affective and instrumental assistance or resources that help officers cope with the more difficult aspects of their job (Cohen & Wills, 1985; Cullen et al., 1985; La Rocco & Jones, 1978). The number of inmates officers talked with had no effect on officers' stress levels, which may imply that the number of inmates officers talk to is less important than perhaps the types of inmates officers interact with on a daily basis (e.g., violent inmates). Supportive elements outside of work were also critical for insulating officers from work stress; officers who felt their family members understood how difficult their job was or felt more comfortable taking about work with family members experienced lower levels of stress. Researchers should continue exploring the ways that social support mechanisms enhance officers' abilities to cope more successfully with the more difficult aspects of their job (Lambert et al., 2018). To our knowledge, our study was the first to examine the benefits of both work- and family-related support for reducing work stress among jail officers, and although more research should certainly be conducted, our findings suggest that family support is another critical resource for helping officers cope with workplace adversities.

At the facility level, jails that maintain adequate coverage of officers relative to inmates might delegate a more manageable number of tasks per officer and afford security staff the ability to maintain social contacts with inmates, other officers, and/or their supervisors that are beneficial for preserving personal safety and wellness (Finn, 2000; Lambert et al., 2018; Ellison & Gainey, 2020; Steiner & Wooldredge, 2015). While the size of jail populations and the extent of crowding were less influential for the level of stress across jails, we found that facilities that operated with more inmates per officer and those with higher rates inmate violence had greater levels of stress across their respective officer corps. In facilities where staff coverage has not kept pace with increases in the number of individuals incarcerated, officers' workloads may be increased, which overextends their capacities as managers and supervisors of the inmate population (Lambert et al., 2018; Steiner & Wooldredge, 2009). Understaffed facilities also create greater uncertainty for officers and limit opportunities for the development of supportive relationships with supervisors, other officers, and inmates (Liebling, 2004). Understaffed facilities also typically have mandatory overtime, which further limits officers' time with external social support circles (e.g. family or friends). Researchers need to continue exploring the officer- and facility-level correlates of safety risks for jail and prison officers, as these studies hold unique implications for preserving their physical and mental health (Ellison & Gainey, 2020, Steiner & Wooldredge, 2017). In particular, facility conditions that shift levels of officer wellness and safety across jails and prisons must be examined, as this scholarship may hold unique implications for preserving officer safety and institutional order (Bieire, 2012).

Although officers' demographic characteristics were included only as control variables, and the interpretation of those findings is largely beyond the scope of this study, we found that officers' sex, but not their age or race, affected their level of work stress. The finding that female officers experienced more work stress should not be entirely surprising, however, in part because women have to work in a historically male-dominated profession and may deal with unique demands that increase work-family conflict and work stress (Armstrong et al., 2015; Castle & Martin, 2006; Triplett et al., 1999). Regarding the effects of race and age, our findings partly comport with that of Steiner and Wooldredge (2015). Like prison officers, jail officers encountered similar levels of stress regardless of their age, but our results also showed that race was not a significant predictor of work stress. Thus, our finding provides additional evidence for the idea that white and non-white jail officers encounter a similar degree of stress (e.g., Castle & Martin, 2006), and could reflect the greater diversity of jail versus prison workforces that may overcome impediments to communication and interaction that increase officer stress in prison (Jacobs, 1977; Pastore & Maguire, 2003; Steiner & Wooldredge, 2015; Stephan, 1997, Stephan, 2008).

Research on correctional officer work stress is not only helpful for jail administrators and their officer corps, it provides some much-needed information for those who make decisions regarding jail resources and funding. For example, to the extent possible, these results suggest that administrators and supervisors must delegate a reasonable number of tasks per officer and maintain adequate levels of coverage across their facilities. Yet the large number of open law enforcement positions and inadequate compensation for officers—including both salary and benefits packages—has left many jail administrators with few options. Absent funding increases from local and state governments, administrators are often doing the best with what they are given. Though funding increases for correctional budgets are not always popular with taxpayers, additional resources could be diverted towards jail officers' compensation packages (i.e., salary, paid time off, healthcare, maternity/paternity leave, retirement), making positions more attractive to qualified job candidates. Furthermore, enhancing officer training to include a focus on self-care and trauma-informed care may help officers recognize and cope with stress as well as recognize it among inmates and their colleagues. These changes may increase staffing levels relative to inmates, increase officer safety, reduce stress, and improve their quality of life, all of which could have indirect and positive effects for their family and friends as well as inmates.

In addition, as legislatures consider policy changes such as California's realignment initiative, they must consider the additional burden being placed on local jails and increase funding for coverage relative to changes in the size and composition of inmate populations. If hiring more officers is not possible, jail administrators should consider funding team building exercises and ways for employees to air grievances, discuss work or personal issues, as well as workshops or training seminars aimed at reducing stress and improving safety (Ellison & Gainey, 2020; Steiner and Wooldredge, 2015, Steiner and Wooldredge, 2017). In a dangerous and stressful occupation such as corrections, institutional administrators would be well-advised to protect the largest portion of any correctional budget (i.e., personnel), and improve the future outlook for the hundreds of thousands of individuals who continually risk their well-being in the interest of public safety.

Finally, our findings support using the Job Demand-Control-Support model in gauging the relevant factors of jail security staff stress and welfare. Although the jail environment is characterized as transitory and administratively controlled (compared to the prison environment), the fact that our findings parallel studies of work stress among prison officer samples suggests that supervising the incarcerated population has similar stress points. This suggests the Job Demand-Control-Support model is well-suited for assessing stress among correctional officers and can be used as a guide for administrators interested in promoting a healthy corps of officers. After all, evidence suggests that work stress among correctional officers manifests as negative attitudes and emotionally-driven responses such as frustration and anger (Cullen et al., 1985; Steiner & Wooldredge, 2015), as well as unhealthy coping strategies (e.g., drug and/or alcohol abuse) and a host of negative interpersonal outcomes that could otherwise be addressed (Finn, 2000; Jaegers et al., 2019; Lambert and Paoline, 2005, Lambert and Paoline, 2012; Paoline & Lambert, 2012; Spinaris, Denhof, & Kellaway, 2012). For administrators focused on improving the jail environment, the Job Demand-Control-Support model could enhance their understanding of the sources of stress for custodial security officers and inform response and/or treatment efforts.

Limitations

Although the findings reported here are the results of a rigorous approach, there are limitations to the research design that should be considered alongside the findings and addressed through future research. At the officer-level of analysis, the cross-sectional survey design is potentially problematic for determining time ordering. Future studies should be designed longitudinally to truly capture the processes by which officers develop work stress. In addition, self-report data have been criticized because subjects tend to embellish or under-exaggerate their responses and participants also tend to have issues with recall. Yet most jails and prisons do not track officers' stress levels (i.e., there is no official source of data from which to study officer stress), and we suspect that issues with recall were limited to the independent variables which measured officers' exposure to violence. Nonetheless, to combat this possibility, survey items asked about exposure to violence within the last 3 months (i.e., threats) or one year (i.e., victimization or witnessed violence). It is also important to acknowledge the role other stressors may have had which were not measured in this study. For example, work-family conflict may play a role in the development of work stress among jail officers (see Armstrong et al., 2015), and we encourage researchers to continue exploring the potential correlates of work stress, including stressors that exist both in and outside of work (e.g., pay and living situation).

At the facility-level of analysis, our study is limited given the small number of facilities in the sample, particularly from one of the states from which data were gathered, and future research should aim to include additional facilities and states to confirm our findings. It is important to note that we examined zero-order estimates of jail-level relationships prior to creating our final level-2 model, and we reviewed results excluding one of the facilities and states examined here, but our findings were substantively similar. Researchers should also seek to examine additional measures of inmate populations and the types of individuals that comprise them (e.g., proportion violent offenders, [hetero] homogeneity of inmates, security level, rate of misconduct and/or self-harm), as well as facility design or supervision style (e.g., [in]direct supervision), to examine whether these factors influence levels of officer stress across facilities. Jurisdictional differences between jails make these statistics difficult to accurately assess, however, and many administrators are understandably protective of such information, making them difficult to obtain. Finally, during the process of data collection, it became obvious that the availability and use of Employee Assistance Programs (EAPs) varied widely from one jail to another, as did benefits packages. These considerations are important because EAP programs and benefits—particularly healthcare—can have important implications for aiding officers with workplace adversities, reducing stress, and improving quality of life. Thus, jails that have strong EAP programs and benefits packages might have lower levels of stress across their respective officer populations, though researchers have yet to examine this possibility.

Conclusion

The policy implications here, even considering the study limitations, highlight the relevance of understanding work-related stress for correctional officers. Work stress among prison officers has been the focus of scholarly attention for several decades. It is known that correctional officers working in prison are subject to, among other things, work stress resulting from exposure to violence and conflict between professional and personal lives (Armstrong et al., 2015; Konda et al., 2012; Lambert et al., 2002, Lambert et al., 2004; Lambert et al., 2006). Using the Job Demand-Control-Support model to predict work stress, our work is a part of a new line of inquiry designed to expand what is known about correctional officers working in the jail environment. The transitory nature of American jails means that officers working in them have contact with many more individuals than prison correctional officers. That is to say that jail officers, while more limited in the length of time they supervise individuals, have a broader reach into the community. And, while there were approximately 600,000 new admissions in state and federal prisons in 2017, there were 10.6 million admissions to county jails for that same year (Bronson & Carson, 2019; Zeng, 2019). The reach and influence of correctional officers working in local jails is broad and thus, supporting their well-being may well diffuse into the community. If nothing else, best practices for making jails less stressful and safer for officers may pay dividends, not only for the officers themselves, but also the inmates they supervise, their families, and their friends or acquaintances. Indeed, an officer that is less stressed may be better able to prioritize important aspects of their job such as inmate safety, as well as allow them to deal with or tend to problems outside of work, which may improve their overall quality of life.

The importance of properly equipped correctional officers working in jails is timely given the current debate about incarceration and policies designed to reduce the number of individuals in prison as well as the amount of time people spend incarcerated in both prison and jail. Jails are an intermediate step in the progression to the deep end of the criminal justice system (i.e., prison), and while some policies have decreased how much the deeper end of the system is utilized, some changes have arguably placed additional pressures on the other components of the system such as jails (Bernard, Paoline, & Pare, 2005; Caudill et al., 2014). Clearing prisons is likely to further complicate jail officers' attempts to navigate the stressful aspects of their work and adversely influence their job performance, well-being, and quality of life, all of which may indirect affect the well-being of individuals closest to them both in and outside of work.

Declaration of Competing Interest

None.

Scale

Item

Work stress

A lot of time my job makes me very frustrated or angry.

I am usually under a lot of pressure while at work.

When I am at work I often feel tense or uptight.

I am usually calm and at ease when I am working. (reverse coded)

There are a lot of aspects of my job that make me upset.

Most of the time when I'm at work, I don't feel like I have much to worry about. (reverse coded)

Role problems

I have to do things at work that should be done differently.

I work on unnecessary projects or tasks on the job.

This facility's policies and guidelines are not very realistic in practice.

I often have to ignore a rule or policy in order to carry out my tasks at work.

The policies and guidelines in this facility are often contradictory.

Autonomy

Most of the time I am able to choose how my tasks are completed on my shift.

I have a great deal of freedom as to how I do my job.

I do not have much opportunity to make my own decisions at work. (reverse coded)

Perceived cooperation from inmates

Inmates help me identify problems that exist in my work area.

Inmates help me identify those who pose a danger to themselves or others.

Inmates often tell me about rule violations committed by other inmates.

Inmates typically tell me the truth if I ask them something.

Coworker support

My coworkers often blame one another when things go wrong. (reverse coded)

I receive compliments from my coworkers when I have done my job well.

My coworkers encourage me to think of better ways of getting my work done.

I generally receive help from my coworkers when I ask for it.

Supervisor support

My immediate supervisor is helpful in getting me what I need to do my job effectively.

My immediate supervisor encourages me if I do my job well.

My immediate supervisor helps me resolve problems when they arise.

My immediate supervisor blames others when things go wrong. (reverse coded)

Family support

My family members understand how difficult my job can be.

I can turn to my family when my job upsets me.

My family helps me relax when I am worried about work.

There is really no one in my family that I can talk to about work. (reverse coded)

Perceived dangerousness

Threaten your safety?

Threaten your family's safety?

Injure you if you try to break up a fight between inmates?

Injure you as you walk around the housing unit?

Injure you while you escort inmates to and/or from housing units?