Burnout Among Probation Officers: The Importance of Resilience

Abstract

Like many human service professionals, probation officers are subject to a great deal of stress as part of their occupational duties. This study examines occupational stress and burnout among probation officers through the lens of the Job Demands-Resources (JD-R) model. This model suggests that organizational characteristics have implications for individual outcomes related to stress. However, the model largely neglects the role of individual factors, such as dispositional resilience. This study tests a refined model using cross-sectional surveys from 289 members of a probation officer union in the eastern United States. Results from structural equation modeling support the general predictions of the JD-R model in that job demands and job resources are correlated with burnout and engagement, which in turn predict health issues and intention to quit. Moreover, resilience significantly predicts every latent variable in the model, both directly and indirectly through its effect on intervening factors. Implications for workplace policy and practice are discussed.

Keywords

probation job demands stress burnout structural equation modeling

The link between workplace stress and declines in productivity, health, and retention is well-documented across a variety of occupations (Gershon et al., 2009; Iwasaki et al., 2002; Waters & Ussery, 2007). While criminal justice professions are commonly considered stressful, most research has focused on the stressors faced by law enforcement and corrections officers; considerably fewer studies have examined probation and parole staff (e.g., Garner et al., 2007; Jin et al., 2018). Research suggests, for example, that police officers experience very high levels of stress and, because probation officers routinely encounter similar stressors, they probably also experience elevated stress levels (Lutze et al., 2012).

Furthermore, probation officers may experience several factors unique to their occupation (see Presley, 2017 for thorough overview). Compared to many other criminal justice professionals, probation officers typically supervise more cases at any given time, and that number has been increasing over time to unacceptable levels (DeMichele & Payne, 2007). As of 2019, there were an estimated 91,800 probation officers in the United States responsible for the nearly 3.5 million individuals under probation supervision (Bureau of Labor Statistics, 2020; Maruschak & Minton, 2020). Therefore, probation officers may be subjected to greater levels of overload and fatigue (Wells et al., 2006). This is likely due in part to the dual nature of the job, which focuses on surveilling and supervising clients on one hand and ensuring they receive treatment and rehabilitative services on the other. With this bifurcated occupational philosophy and frequent contact with clients at all levels of the criminal justice process (in court, at home, in the office, and occasionally even in jail), it is important to study probation officers and the sources of their stress (Finn & Kuck, 2003; O’Donnell & Stephens, 2001). If stress among probation officers has unique correlates, then it likely has unique solutions.

While the impacts of stress are well-documented for numerous professions, research on the causes of workplace stress typically focus on organizational factors. In many studies, individual-level demographics exert little influence on stress and are often overshadowed in impact by organizational traits (Gershon et al., 2009; Whitehead & Lindquist, 1985). The resulting conclusion is somewhat deterministic—that fault (and correction) lie in the hands of organizational leadership and that individuals are victim to the whims of macro-level processes. As reviewed later, however, several studies have explored the impact of individual-level resilience as a buffer against the deleterious impacts of stressors. Importantly, resilience might explain why some individuals in stressful situations do not experience burnout and other negative impacts of stress.

Despite a plausible link between resilience and the impacts of occupational stress, these branches of research often operate in seeming isolation from each other, examining either how to promote resilience in employees (Vogelvang et al., 2014), the impact of hardiness on stress and coping (Rossi, 2007), or the impact of workplace stressors on intention to quit (W.J. Lee et al., 2009). Rarely do studies consider the whole chain of events simultaneously. Additionally, we are aware of no studies that assess the modulating effects of dispositional factors (e.g., resilience) on workplace outcomes (e.g., burnout and intention to quit) among members of an occupation that reports very high levels of stress, especially compared to the general population (see, e.g., Finn & Kuck, 2003; Lutze et al., 2012).

In this paper, we argue that resilience is an important but underused concept in studies of workplace stress, particularly in studies relying on criminal justice samples. Drawing from research on the importance of resilience as a buffer against stressors, we examine whether and to what extent resilience mediates the impact of stress on burnout, engagement, health complaints, and intention to quit using a sample of probation officers in the eastern United States.

Literature Review

The Job Demands-Resources Model

The Job Demands-Resources Model (hereafter JD-R) was designed to study the processes leading to employee burnout and engagement and the subsequent implications for employee turnover and general well-being (Bakker & Demerouti, 2017; Schaufeli & Bakker, 2004). As illustrated in Figure 1, the model consists of two parallel processes: a health impairment process and a motivational process. Under the health impairment process, chronic demands—whether physical, social, or task-oriented—will lead to burnout, defined as a state of mental weariness, exhaustion, and cynicism about work. This chronic state of exhaustion contributes to declines in health status and absenteeism. Under the motivational process, resources, such as feedback and social support, motivate the employee to become engaged at work. In turn, engagement reduces the likelihood that the employee will want to quit. Support for the model is widespread across numerous fields (Crawford et al., 2010; Schaufeli & Taris, 2014).

Figure 1.

The job demands-resources theoretical model.

Applying the JD-R model conceptually to probation officers, the potential linkages become evident. The number of probation officers has not kept pace with the change in correctional populations—particularly in the United States—subsequently leading to inflated caseloads (Finn & Kuck, 2003). Professionals who work in human services occupations, such as probation, parole, and social work, are often personally impacted by the trauma experienced by their clients (Lewis et al., 2013; Pearlman & Mac Ian, 1995). Additionally, internal conflicts resulting from disconnects between preferred and actual styles of supervision, such as a social worker versus a law enforcement mentality, may be an important source of stress (Steiner et al., 2004). Personnel who face excessive job demands like high caseloads with little perceived control over those demands may end up experiencing role overload (too many duties), role conflict (competing obligations), role ambiguity (unclear duties), and a heightened sense of danger (R. T. Lee & Ashforth, 1996; Neveu, 2007). Overworked staff may feel increasingly stressed, which then leads to burnout and subsequent turnover (Lambert et al., 2015). Thus, there exists ample reason to suspect that the JD-R model would be supported in studies of probation officers.

Resilience

While the JD-R model has been widely tested and supported, one criticism is that, as a theory of individual responses to organizational behavior, it considers organizational factors as the primary causes of negative work-related outcomes and downplays the potential impact of dispositional traits. Dispositional resilience or hardiness, is defined as a personality trait related to continued good health and performance under stress (Bartone, 2007; Bartone et al., 2006). Hardy individuals express a strong perception of control over their lives and welcome challenging situations (Kobasa, 1979). As a dispositional trait, hardiness develops early in life and is typically stable over the life course. The concept of dispositional resilience has been widely found to modulate the effects of stress among samples and topics ranging from college students (Sagone & De Caroli, 2014) and children of migrant farmworkers (Taylor & Ruiz, 2019) to physicians (Taku, 2014), prisoners (Lo et al., 2020), law enforcement officers (Williams et al., 2010), and women with delayed menstrual cycles (Palm Fischbachher & Ehlert, 2014). One meta-analysis of 180 studies found that hardiness is significantly and negatively related to stress and positively related to performance, and these effects hold net of other personality characteristics (Eschleman et al., 2010).

The Present Study

Based on a review of extant literature, the issue addressed by the current study is whether dispositional resilience affects other workplace indicators, such as those included in the JD-R model. If so, the concept could be used to explain why some individuals in stressful environments—such as probation work—do not eventually burn out and experience other negative impacts related to their work environment. Specifically, we ask the following questions:

Does dispositional resilience affect subjective perceptions of job demands and job resources? That is, do resilient individuals perceive greater resources and fewer demands compared to less resilient individuals?

Does dispositional resilience buffer against burnout and promote engagement in work? That is, does resilience exert direct effects on burnout and engagement and/or indirect effects through its impact on perceived job demands and job resources?

Does dispositional resilience protect against health concerns and desires to quit? That is, are resilient individuals more likely to report better health outcomes and a decreased desire to leave a stressful work environment?

Based on the above questions, we derive the following revised theoretical model (see Figure 2). We suspect that, as a dispositional characteristic, hardiness can influence both perceptions of resources and demands at work. For example, hardiness should impact “felt” stress regardless of objective stress levels. Hardiness also likely impacts burnout and job engagement among employees, since hardy individuals should be more likely to adapt to and perhaps even welcome stressful situations. Finally, we suspect that resilience affects intention to quit, because resilient employees are more likely to adapt to the stresses of the job, and health concerns, because resilient individuals are more likely to adapt to stressors.

Figure 2.

A modified job demands-resources model including hardiness.

Methods

Sample and Procedure

To address these questions, the researchers collected survey data from a sample of probation officers with the assistance of the Probation Association of New Jersey (PANJ), a statewide union for probation officers. The organization represents the 1,772 career probation officers and 642 first level supervisors spanning the entire state of New Jersey, of which 1,657 and 557 are dues-paying members, respectively. To recruit participants, the union president posted a link to the questionnaire on the organization’s website and emailed the link to all county chapter presidents, who in turn forwarded the link to their respective membership. The researchers also attended the union’s annual conference to promote the data collection effort and remind attendees to complete the survey. As an incentive, all respondents were offered a chance to win one of three $50 gift cards to a business of their choice. To maintain anonymity and prevent the linkage of contact information from the raffle to individual survey respondents, the raffle data were stored in a separate survey that launched only after the initial survey was submitted.

Surveys were administered in Qualtrics, a comprehensive software package for the creation, delivery, and storage of survey data. While we cannot confirm how many individuals viewed or received the link out of the universe of those who could possibly participate, analytics show that 406 started it. To minimize the potential for duplicate submissions, a setting was enabled in Qualtrics that restricted the same IP address from taking the survey multiple times. Upon opening the survey, respondents were assured that their information was anonymous. All responses were voluntary. Respondents gave passive consent by submitting a completed survey. Any survey attempts stored as “in progress,” which were not submitted on the final survey screen, were automatically discarded to minimize concerns about informed consent. Of the 406 people who started the survey, 289 progressed to the final question and submitted responses for an estimated response rate of roughly 17% of the population. The average completion time was 18 minutes. Responses were downloaded to a local encrypted computer for analysis. Of the 289 responses, 241 contained no missing data and 48 were missing some information. We detected no logical pattern among respondents with missing data, so we consider the data to be missing at random for modeling purposes and use an imputed sample of all 289 responses.

One potential concern with the data collection procedure (a mixed notification approach of digital links with in-person data collection at a statewide conference) is that respondents who attended the conference may be different from respondents who did not regarding attitudes toward their work. For example, individuals who attended the conference may have favorable attitudes toward their career (more engaged, less likely to feel burnt out less likely to quit, etc.). While we are unable gauge, overall, what percentage of respondents who completed the survey at any point attended the conference due to the anonymous nature of the survey, we collected only 34 responses in person from the conference; the remaining 255 were collected either before or after the conference. Anecdotally, nearly all the at-conference respondents noted in casual conversation that they had seen the e-mail link but kept forgetting to complete it (e.g., an e-mail sent during work hours while officers were meeting with clients). We conducted sensitivity analyses comparing the model results with and without the 34 at-conference respondents and found no meaningful differences between the models.

Measures

Unless otherwise noted, all measures were treated as latent variables based on observable indicators of the underlying concepts. Given the large number of these observed indicators (100 survey items plus three demographic variables), latent variables were modeled using single item parcels (Chen & Weng, 2019; Little et al., 2013; Yang et al., 2010). Under this approach, researchers create a single-item index (i.e., through averaging component items) and use the reliability of the scale to specify the fraction of variance not due to measurement error. Results substituting three parcels per latent variable and using all 100 observed indicators in the measurement model revealed similar substantive findings (results available upon request), though the models took significantly longer to converge and had worse model fit due to the large number of parameters and omitted pathways. For computational parsimony, and because the focus of the study is to understand the linkages between latent variables rather than the factor loadings of individual survey items, we opted to use single item parcels.

Resilience

Hardiness was measured using the 15-item Dispositional Resilience Scale (DRS-15; Bartone et al., 2006). Respondents were asked to reflect on their dispositional outlook. Responses ranged from not at all true (1) to completely true (4). Six negative items in the DRS-15 were reverse coded prior to scale creation for consistency in the scale interpretation. Sample questions include “My choices make a real difference in how things turn out in the end” and “Changes in routine are interesting to me.” The scale alpha was .818 with a mean score of 2.86.

Job demands-resources model

Job demands

Demands on probation officers were measured with a modified version of Simmons et al., (1997) stress scale for probation officers. Respondents were asked to identify how stressful they perceived a variety of job-related functions. The 26-item, 5-point Likert scale consisted of potential answers from not stressful (1) through very stressful (5). Sample items from this scale include “Visiting probationers’ homes,” “Excessive paperwork,” and “Duties and responsibilities not clearly defined.” Responses reflected respondents’ subjective levels of stress to these various obligations. The scale alpha was .910 with a mean score of 3.43.

Job resources

Job resources consisted of 26 items that capture perceived availability of a range of job resources, including autonomy (Bakker et al., 2004), training (Getahun et al., 2008), colleague support (Van Veldhoven et al., 1997), supervisor support (Graen & Uhl Bien, 1995), and other tangible benefits, like salary and vacation time. Responses reflected officers’ subjective satisfaction with a specific aspect of their job. All items were measured on a five point Likert scale. Sample items included “My job offers me opportunities to find out how well I do,” “I am adequately trained for my job,” and “I can count on my colleagues when I encounter difficulties in my work.” Responses ranged from 1 (strongly disagree) to 5 (strongly agree) with an average of 3.09. The reliability of the scale was .911.

Burnout

The Oldenburg Burnout Inventory (OLBI; Demerouti et al., 2010) is a 16-item scale that measures both burnout and engagement. Nine of the items reflect burnout, with sample items like “There are days when I feel tired before work” and “Lately, I tend to think less at work and do my job automatically.” Responses ranged from strongly disagree (1) to strongly agree (5). The average was a 3.19 with an alpha of .799.

Engagement

The OLBI was also used to measure engagement. Seven items in the questionnaire reflect work engagement. Responses ranged from strongly disagree (1) to strongly agree (5). Sample items include “When I work, I usually feel energized” and “I feel more and more engaged in my work.” The average was a 2.77 out of 5 with a reliability coefficient of .775.

Health concerns

We used a shortened version of the Calgary Symptoms of Stress Inventory (C-SOSI; Carlson & Thomas, 2007) to measure respondents’ self-reported health concerns. Respondents indicated how often they experienced 14 specific symptoms in the past month on a scale from never (1) to very often (5). Sample items include nausea, chest pains, rapid breathing, difficulty sleeping, and mood swings. The average was 2.38 out of 5 with an alpha of .933.

Turnover intention

Turnover intention was measured by three items (W. J. Lee et al., 2009). Two items included “As soon as I can find a better job, I will quit my current job” and “I often think about quitting my job,” and responses ranged from strongly disagree (1) to strongly agree (5). The third item asked “How likely are you to quit your job in the next year?” with responses ranging from not at all likely (1) to very likely (5). Because of the small number of items in this measure, we opted to include all of them in the measurement model to capture the latent concept of turnover intention rather than create an item parcel.

Control variables

Additionally, we included three demographic characteristics as exogenous control variables: employment length, gender, and race. Gender and race were operationalized as dichotomous variables, with females and white respondents as the focal categories. Employment length was operationalized as the number of years the respondent has been working in the same county probation department. We opted for this definition to avoid confounding current and past workplace dynamics, given that different agencies likely have different resources and stressors.

Analyses

Analyses were conducted using the SEM package in Stata version 14. Latent variables were measured as a function of the parceled items, such that each latent variable consisted of a single parcel and a variance component representing the reliability of the scale. The exception to this is turnover intention; all three observed variables were entered as indicators. This process allowed the measurement model for each latent variable to be just-identified.

We used the method(mlmv) option in Stata to estimate the models using a maximum likelihood for missing values algorithm. The process is like multiple imputation of missing data prior to analyses, but it skips the imputation step by using maximum likelihood estimation instead. Therefore, the reported models capture all 289 respondents.

Results

Descriptive Statistics

Table 1 presents descriptive statistics for the sample demographics and key indicators used in the study. Note that the means listed for the latent variables reflect average scores for the observed item parcels. The structural equation models used maximum likelihood estimation to handle missing information.

Table 1.

Descriptive Statistics.

Measure	N	M /PCT	SD	Minimum	Maximum
Control variables
Years on job	241	14.216	8.499	1	35
Female	260	65.8%	—	0	1
White	242	70.7%	—	0	1
JD-R model
Job demands	289	3.435	0.663	1	4.962
Job resources	289	3.091	0.613	1.423	4.551
Burnout	284	3.190	0.646	1.333	4.778
Engagement	284	2.771	0.667	1.143	4.571
Health concerns	282	2.379	0.868	1	5
Turnover intention	284	2.769	1.217	1	5
Hardiness	276	2.861	2.861	1.667	3.800

Structural Equation Models

We computed a series of structural equation models to test the impact of adding omitted pathways from the theoretical model. Model fit statistics are presented in Table 2. Model 1 is a baseline Job Demands-Resources model that does not include resilience. Model 2 adds resilience with full mediation. In this model, resilience is a predictor of job demands and job resources, and any effects of resilience on later latent variables occur indirectly through demands and resources. Model 3 is a partial mediation model that allows resilience to have direct effects on all subsequent components. Model 4, the “partial mediation plus” model, adds two additional cross-linked pathways to Model 3: a direct effect of job demands on engagement and a direct effect of engagement on health. Model 5 represents a no mediation model, where all measures in one “step” of the model are added to the equations for the next step. For example, job demands and job resources are modeled to have direct effects on turnover and health instead of purely indirect effects through those latent variables.

Table 2.

Model Fit Parameters.

	Model 1: JD-R	Model 2: FM	Model 3: PM	Model 4: PM+	Model 5: NM
χ ²	59.522	142.958	61.357	60.786	58.030
df	22	28	24	22	18
RMSEA	0.077	0.119	0.073	0.078	0.088
pclose	0.030	0.000	0.044	0.024	0.007
CFI	0.964	0.901	0.968	0.966	0.965
CD	0.238	0.299	0.324	0.323	0.322
AIC	7624.997	7546.828	7473.227	7476.656	7481.900

Note. N = 289. Model 1 contains the 6 latent variables in the JD-R model; Model 2 adds fully mediated effects of resilience; Model 3 adds resilience with partial mediation; Model 4 adds cross-linked pathways to Model 3; Model 5 adds all omitted pathways (no mediation); See text for detailed description.

Considered as a package, the fit indices suggest acceptable model fit for several of the models. A cursory review of the fit statistics revealed that Model 3, the partial mediation model, fit the data best. Additionally, a likelihood ratio test found that Model 3 fit significantly better than Model 2 (χ² = 81.60, df = 4, p < .001). However, Models 4 and 5 did not represent a significant improvement over Model 3 (χ² = 0.57, df = 2, p = .75; χ² = 3.33, df = 6, p = .77). Therefore, remaining analyses focus on the effects in Model 3.

Figure 3 presents the standardized direct effects from Model 3, the partial mediation model. It is first worth mentioning the predictors of dispositional resilience or hardiness. In this study, we examined the influence of three demographic characteristics (years on the job, gender, and race) in the JD-R model. Compared to males and nonwhite respondents, females (B = 0.109, SE = 0.054, p < .05) and white (B = −0.194, SE = 0.058, p < .001) respondents, respectively, reported greater and lower resilience while years of service had no impact. Additionally, we examined the total effect of these predictors on other constructs in the model, but their impact was mostly non-significant. Females reported more engagement at work (B = 0.216, SE = 0.087, p < .05) and white respondents reported significantly lower intention to quit their job (B = −0.400, SE = 0.164, p < .05); none of the other total effects were significant. To the extent that demographics may have impacted perceived resources and demands, engagement and burnout, and health and turnover intention, the effects were primarily indirect through the variables’ influence on resilience.

Figure 3.

Direct effects in Model 3, the dispositional resilience model.

Resilience was a significant predictor of each latent variable in the JD-R model, and its effects were—with one exception—in the expected direction. Resilient individuals reported decreased perceptions of workplace demands, lower levels of burnout, and fewer concerns while also reporting a more positive perception of job resources and improved work engagement. However, the direct effect of resilience on turnover intention was opposite of that expected: respondents with higher self-reported hardiness also reported an increased intent to quit their job.

In addition to the direct effects shown in Figure 3, we investigated the indirect and total effects in the model. For parsimony, we present the indirect and total effects for the focal variable in this study: dispositional resilience. Resilience exerted a negative indirect effect on burnout (B = −0.310, SE = 0.065, p < .001) and a positive indirect effect on engagement (B = 0.247, SE = 0.063, p < .001) through its direct effects on demands and resources. The indirect effects of resilience on health concerns and turnover were both negative (B = −0.915, SE = 0.135, p < .001 and B = −1.495, SE = 0.238, p < .001, respectively). Taken together, these indirect effects are largely consistent with the direct effects: resilient individuals reported lower levels of burnout, health concerns, and intention to quit their job and higher levels of job engagement.

Not surprisingly, then, the total effects (indirect plus direct effects) reveal a similar pattern. Hardiness was related to decreased level of perceived job demands (B = −0.527, SE = 0.118, p < .001), burnout (B = −0.911, SE = 0.106, p < .001), health complaints (B = −1.321, SE = 0.139, p < .001), and intention to quit (B = −0.742, SE = 0.197, p < .001). Conversely, hardiness was related to higher levels of perceived job resources (B = 0.455, SE = 0.109, p < .001) and engagement in work (B = 0.823, SE = 0.111, p < .001). To summarize, dispositional resilience was a significant predictor for all components of the JD-R model, and its addition to the baseline model improved model fit.

Discussion

In a seminal comparative study of probation systems, Hamai et al. (1995) argued that systems of probation are in a constant state of flux as developed nations increasingly rely on probation as a means of social control. Many developing countries in Latin and South America and Eastern Europe have only recently begun to expand previously limited community supervision programs (Porporino, 2018). In nearly all European countries examined in one recent study, the number of people under community supervision has continuously expanded since the early 1990s (Aebi et al., 2015). The seemingly global shift to probation could reflect attempts to decarcerate prison populations or could inadvertently contribute to a net widening whereby more people are swept up by the criminal justice system. Regardless of the specific circumstances, probation officers in many nations are experiencing increasing caseloads. The United States is no exception, where a record number of individuals is under community supervision (Maruschak & Minton, 2020).

Accompanying this global expansion of probation is a growing body of literature suggesting that probation is a stressful occupation with high levels of burnout. While most of this research admittedly uses data from the United States, studies from diverse contexts with unique probation systems, such as the handful of studies conducted in Australia, China, Israel, Poland (Wirkus et al., 2021), and Turkey (Erdem et al., 2019), find remarkably similar patterns in the correlates and consequences of burnout (see Page & Robertson, 2022, for review). Exploring possible solutions to burnout is an important avenue for improving morale and retention in a growing field.

This study examined the role of dispositional resilience, or hardiness, in predicting elements of the Job Demands-Resources model. Results from structural equation models revealed that resilience exhibited moderate to strong effects compared to other predictors in the model, generally in the directions expected. Additionally, its inclusion improved the explained variation of the JD-R model by about 36% in this sample. Returning to the three research questions, all of them were answered affirmatively with one caveat.

The positive and significant direct link between resilience and turnover intention was unexpected. At the same time, the indirect and total effects of resilience on turnover were in the expected direction in that more resilient people reported reduced intent to quit. We can only speculate with the available data. Perhaps resilient respondents, due to their adaptability, were less deterred by the prospects of finding a new place of employment (with potentially new, unknown sources of stress). It is also possible that one or more variables directly predicting turnover intention serve as confounders and/or suppressors. Of course, turnover intention does not necessarily mean actual turnover; resilient employees may entertain the thought of finding a new job but be more likely to endure current conditions. Future research may seek to replicate and explain this dynamic further, but since the total and indirect effects of resilience were in the expected direction, we make this recommendation cautiously.

More recent research has suggested that hardiness may be more dynamic than originally conceptualized (Shochet et al., 2011); that is, hardiness can be developed through programmatic interventions. As Bartone (2006) posited, “if the factors/pathways that lead to human resiliency under stress were better understood, perhaps some of these resiliency factors could be developed or amplified in those who are initially low in resilience and more vulnerable to stress” (p. 132). Researchers must be careful, in their operational definitions, not to conflate dispositional resilience, like hardiness, with situational, or temporary forms of resilience (Jex et al., 2013). Longitudinal data and clear operational definitions are likely necessary to determine whether resilience is fluid over time.

A longitudinal data collection effort would also enable the disentangling of effects and better assess the placement of resilience in the model. We assumed, guided by existing research, that dispositional resilience is a mostly stable personal characteristic and, therefore, placed it with other personal characteristics exogenous to the JD-R model. However, if resilience is more dynamic, it may change in response to perceived workplace stressors and resources. Therefore, longitudinal data would also permit the investigation of resilience as impacted by, for example, a perception of overwhelming job demands. With cross-sectional data, any conclusions about cause and effect, reciprocal effects, or feedback loops are merely tentative.

Like most examinations of the JD-R model, this study relies on subjective self-report data. As Bakker and Demerouti (2017) suggest, however, relationships may be inflated because of common source bias; that is, because the focal employee provides all information, it is difficult to extract objective relationships from the data. Future research would benefit from including alternative sources of information, such as job demands reported by the supervisor or quantifiable information like caseload, objective health screenings, and other sources in addition to the focal respondent.

Future research would also benefit from a multi-level investigation of these processes. Multi-level analyses could include day-by-day variations in demands rather than a global average, nest respondents within a departmental, organizational, or other higher-order context to examine group effects (e.g., whether a particular unit impacts individuals more than another), or examine group-level outcomes like performance and objective turnover rates.

If the findings of this study hold under replication, including in other jurisdictions in the United States and around the world, the implications for workplace policy and practice are important to consider. Probation departments, and likely other agencies, would benefit from seeking out and/or creating resilient employees. If dispositional resilience is generally set early in life, static over the life course, and cannot be changed through workplace programing, then employers may consider incorporating resilience as a desirable characteristic in the hiring process. However, if hardiness is dynamic rather than a static personality trait, employers may consider investment in resilience development programs a worthwhile investment for workplace productivity, morale, and employee well-being.

Footnotes

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) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: This research was supported by the William Paterson University College of Humanities and Social Sciences Summer Research Program.

ORCID iD

Andrew S. Gladfelter

References

Aebi

M. F.

Delgrande

Marguet

(2015). Have community sanctions and measures widened the net of the European criminal justice systems? Punishment & Society, 17(5), 575–597.

Bakker

A. B.

Demerouti

(2017). Job demands-resources theory: Taking stock and looking forward. Journal of Occupational Health Psychology, 22(3), 273–285.

Bakker

A. B.

Demerouti

E. E.

Verbeke

(2004). Using the job demands-resources model to predict burnout and performance. Human Resource Management, 43(1), 83–104.

Bartone

P. T

. (2006). Resilience under military operational stress: Can leaders influence hardiness? Military Psychology, 18, 131–148.

Bartone

P.T.

(2007). Test-retest reliability of the Dispositional Resilience Scale-15, a brief hardiness scale. Psychological Reports, 101(3), 943–944.

Bartone

P. T.

Johnsen

B. H.

Eid

Laberg

J. C.

Thayer

Sommerfeldt Pettersen

(2006). International adaptation of a brief human resiliency (hardiness) scale [Paper presented]. Annual meeting of the Association of Military Surgeons of the US.

Bureau of Labor Statistics. (2020). Probation officers and correctional treatment specialists [Occupational outlook handbook]. Author. https://www.bls.gov/ooh/community-and-social-service/probation-officers-and-correctional-treatment-specialists.htm

Carlson

L. E.

Thomas

B. C.

(2007). Development of the calgary symptoms of stress inventory (C-SOSI). International Journal of Behavioral Medicine, 14(4), 249–256.

Chen

Y. L.

Weng

L. J.

(2019). The nature of multidimensional constructs represented by item parcels in structural equation modeling. Chinese Journal of Psychology, 61(1), 25–50.

10.

Crawford

E. R.

LePine

J. A.

Rich

B. L.

(2010). Linking job demands and resources to employee engagement and burnout: A theoretical extension and meta-analytic test. Journal of Applied Psychology, 95(5), 834–848.

11.

Demerouti

Mostert

Bakker

A. B.

(2010). Burnout and work engagement: A thorough investigation of the independency of both constructs. Journal of Occupational Health Psychology, 15(3), 209–222.

12.

DeMichele

Payne

B. K.

(2007). Probation and parole officers speak out-Caseload and workload allocation. Federal Probation, 71, 30–35.

13.

Erdem

Tuncer

A. E.

Safi

O. A.

Çankaya

Ergin

Aydoğan

(2019). The professional experiences and training needs of probation officers in Turkey. Journal of Social Work, 19(5), 664–686.

14.

Eschleman

K. J.

Bowling

N. A.

Alarcon

G. M.

(2010). A meta-analytic examination of hardiness. International Journal of Stress Management, 17(4), 277–307.

15.

Finn

Kuck

(2003). Addressing probation and parole officer stress. Abt Associates.

16.

Garner

B. R.

Knight

Simpson

D. D.

(2007). Burnout among corrections-based drug treatment staff: Impact of individual and organizational factors. International Journal of Offender Therapy and Comparative Criminology, 51(5), 510–522.

17.

Gershon

R. R.

Barocas

Canton

A. N.

Vlahov

(2009). Mental, physical, and behavioral outcomes associated with perceived work stress in police officers. Criminal Justice and Behavior, 36(3), 275–289.

18.

Getahun

Sims

Hummer

(2008). Job satisfaction and organizational commitment among probation and parole officers: A case study. Professional Issues, in Criminal Justice, 3(1), 1–16.

19.

Graen

G. B.

Uhl Bien

(1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2), 219–247.

20.

Hamai

(1995). Probation round the world: A comparative study. Routledge.

21.

Iwasaki

Mannell

R. C.

Smale

B. J.

Butcher

(2002). A short-term longitudinal analysis of leisure coping used by police and emergency response service workers. Journal of Leisure Research, 34(3), 311–339.

22.

Jex

S. M.

Kain

Park

(2013). Situational factors and resilience: Facilitating adaptation to military stressors. In Sinclair

R. R.

Britt

T. W.

(Eds.), Building psychological resilience in military personnel: Theory and practice (pp. 67–84). American Psychological Association.

23.

Jin

Sun

I. Y.

Jiang

Wang

Wen

(2018). The impact of job characteristics on burnout among Chinese correctional workers. International Journal of Offender Therapy and Comparative Criminology, 62(2), 551–570.

24.

Kobasa

S. C.

(1979). Stressful life events, personality, and health: An inquiry into hardiness. Journal of Personality and Social Psychology, 37(1), 1–11.

25.

Lambert

E. G.

Hogan

N. L.

Kelley

(2015). The correctional staff burnout literature. Criminal Justice Studies, 28(4), 397–443.

26.

Lee

R. T.

Ashforth

B. E.

(1996). A meta-analytic examination of the correlates of the three dimensions of job burnout. Journal of Applied Psychology, 81(2), 123–133.

27.

Lee

W. J.

Phelps

J. R.

Beto

D. R.

(2009). Turnover intention among probation officers and direct care staff: A statewide study. Federal Probation, 73(3), 28–40.

28.

Lewis

K. R.

Lewis

L. S.

Garby

T. M.

(2013). Surviving the trenches: The personal impact of the job on probation officers. American Journal of Criminal Justice, 38(1), 67–84.

29.

Little

T. D.

Rhemtulla

Gibson

Schoemann

A. M.

(2013). Why the items versus parcels controversy needn’t be one. Psychology Methods, 18(3), 285–300.

30.

Iasiello

Carey

Van Agteren

(2020). Improving the wellbeing of female prisoners via psychological skills training: A feasibility study. International Journal of Offender Therapy and Comparative Criminology, 64(15), 1571–1586. https://doi.org/10.1177/0306624X20928029

31.

Lutze

Johnson

Clear

Latessa

Slate

(2012). The future of community corrections is now: Stop dreaming and take action. Journal of Contemporary Criminal Justice, 28(1), 42–59.

32.

Maruschak

L. M.

Minton

T. D.

(2020). Correctional populations in the United States, 2017-2018. Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/cpus1718.pdf

33.

Neveu

J. P.

(2007). Jailed resources: Conservation of resources theory as applied to burnout among prison guards. Journal of Organizational Behavior, 28(1), 21–42.

34.

O’Donnell

Stephens

(2001). The impact of organizational, social environmental and job content stressors on the work-related strains of probation officers. Australian & New Zealand Journal of Criminology, 34(2), 193–203.

35.

Page

Robertson

(2022). Extent and predictors of work-related distress in community correction officers: A systematic review. Psychiatry, Psychology, and Law, 29(2), 155–182. https://doi.org/10.1080/13218719.2021.1894259

36.

Palm Fischbacher

Ehlert

. (2014). Dispositional resilience as a moderator of the relationship between chronic stress and irregular menstrual cycle. Journal of Psychosomatic Obstetrics & Gynecology, 35(2), 42–50.

37.

Pearlman

L. A.

Mac Ian

P. S.

(1995). Vicarious traumatization: An empirical study of the effects of trauma work on trauma therapists. Professional Psychology: Research and Practice, 26(6), 558–565.

38.

Porporino

F. J.

(2018). Developments and challenges in probation practice: Is there a way forward for establishing effective and sustainable probation systems? European Journal of Probation, 10(1), 76–95.

39.

Presley

(2017). Probation officer productivity: Using the effort-reward imbalance model. [Doctoral dissertation, University of Central Florida]. STARS. https://stars.library.ucf.edu/cgi/viewcontent.cgi?article=6483&context=etd

40.

Rossi

N. E.

(2007). The role of dispositional resilience in regaining life satisfaction after the loss of a spouse. Death Studies, 31(10), 863–883.

41.

Sagone

De Caroli

M. E.

(2014). A correlational study on dispositional resilience, psychological well-being, and coping strategies in university students. American Journal of Educational Research, 2(7), 463–471.

42.

Schaufeli

W. B.

Bakker

A. B.

(2004). Job demands, job resources, and their relationship with burnout and engagement: A multi-sample study. Journal of Organizational Behavior, 25(3), 293–315.

43.

Schaufeli

W. B.

Taris

T. W.

(2014). A critical review of the job demands-resources model: Implications for improving work and health. In Bauer

G. F.

Hammig

(Eds.), Bridging occupational, organizational, and public health: A transdisciplinary approach (pp. 43–68). Springer.

44.

Shochet

Shakespeare Finch

Craig

Roos

Wurfl

Hoge

Young

M. C.

(2011). The development and implementation of the promoting resilient officers (PRO) program. Traumatology, 17(4), 43–51.

45.

Simmons

Cochran

J. K.

Blount

W. R.

(1997). The effects of job-related stress and job satisfaction on probation officers’ inclinations to quit. American Journal of Criminal Justice, 21(2), 213–229.

46.

Steiner

Purkiss

Kifer

Roberts

Hemmens

(2004). Legally prescribed functions of adult and juvenile probation officers: Worlds apart? Journal of Offender Rehabilitation, 39(4), 47–67.

47.

Taku

(2014). Relationships among perceived psychological growth, resilience and burnout in physicians. Personality and Individual Differences, 59, 120–123.

48.

Taylor

Z. E.

Ruiz

(2019). Executive function, dispositional resilience, and cognitive engagement in Latinx children of migrant farmworkers. Children and Youth Services Review, 100, 57–63.

49.

Van Veldhoven

Meijman

T. F.

Broersen

J. P. J

Fortuin

R. J

. (1997). Research into the experience of psychosocial work load and work stress using the questionnaire experience and assessment of labor. Retrieved August 15, 2020, from https://www.researchgate.net/publication/254786754

50.

Vogelvang

Clarke

Welland

A. S.

Vosters

Button

(2014). Resilience of Dutch probation officers: A critical need for a critical profession. European Journal of Probation, 6(2), 126–146.

51.

Waters

J. A.

Ussery

(2007). Police stress: History, contributing factors, symptoms, and interventions. Policing: An International Journal of Police Strategies and Management, 30(2), 169–188.

52.

Wells

Colbert

Slate

R. N.

(2006). Gender matters: Differences in state probation officer stress. Journal of Contemporary Criminal Justice, 22(1), 63–79.

53.

Whitehead

Lindquist

(1985). Job stress and burnout among probation/parole officers: Perceptions and causal factors. International Journal of Offender Therapy and Comparative Criminology, 29(2), 109–119.

54.

Williams

Ciarrochi

Deane

F. P.

(2010). On being mindful, emotionally aware, and more resilient: Longitudinal pilot study of police recruits. Australian Psychologist, 4 5(4), 274–282 .

55.

Wirkus

A. B.

Opora

Stasiak

. Burnout among probation officers in Poland and the role of preferred styles of coping with stress. (2021). International Journal of Environmental Research and Public Health, 18(1), 355.

56.

Yang

Nay

Hoyle

R. H.

(2010). Three approaches to using lengthy ordinal scales in structural equation models: Parceling, latent scoring, and shortening scales. Applied Psychological Measurement, 34(2), 122–142.