Burnout in Residential Support Workers: The Impact of Locus of Control and Perceived Supervisor Support

ABSTRACT

Child protection residential support workers experience physical and emotional workplace stressors that put them at high risk of experiencing burnout. Despite the demanding nature of residential support work, there has been limited research into the experience of Australian residential support workers and the factors that are linked to burnout in this environment. Johnson and Hall’s (1988) Job Demand-Control-Support model proposes that burnout occurs when individuals in high demand roles also have limited control and social support in their roles. The current study aimed to understand the relationship between locus of control and burnout, and perceptions of supervisor support and burnout in residential care environments. A cross-sectional online survey was completed by 50 Australian residential support workers, recruited through two resident homes, social media, and an industry association e-mail list. This survey included the Work Locus of Control Scale, the Survey of Perceived Organisational Support, and the Copenhagen Burnout Inventory. Data were analyzed using two separate hierarchical regression models. Findings indicated that a greater external locus of control was associated with higher work-related burnout. Additionally, a significant negative relationship was seen between perceived supervisor support and burnout. In developing practice, frameworks designed to prevent or reduce burnout, the impact of locus of control and perceptions of supervisor support should be considered. Further research is recommended to understand the complexity of these relationships along with other factors specific to residential support work that may also impact burnout.

Practice Implications

• Perceived supervisor support is associated with lower work-related burnout in Australian residential care workers

• Residential care workers who feel they have less control in their workplace are likely to experience higher work-related burnout

• Increased staff involvement in decisions related to working conditions in residential care may reduce the prevalence and impacts of burnout

KEYWORDS:

• Residential care workers
• locus of control
• burnout
• supervisor support

Introduction

In Australia, over 46,000 children are currently in some form of out of home care, most of whom have been removed from their parents or guardians as a result of maltreatment, neglect, or abuse (Australian Institute of Health and Welfare, 2021). While the majority of children are placed in foster care, approximately 3,000 are living in government funded residential care (Australian Institute of Health and Welfare, 2021). This does not include children who are placed into residential treatment or youth detention facilities if they have been sentenced to a period of detention due to criminal offending (Zelechoski et al., 2013). Given the increasing number of children who are being placed into residential care facilities, there is an associated increase in the need for residential care workers. However, there are a range of work-related factors that make the role of a residential care worker complex, including the psychological impact of the working environment (Purdy & Antle, 2022). Moreover, impacts on this worker populations may differ depending on the unique characteristics of certain residential care environments (e.g., privately run facilities, government-run facilities, level of family involvement). Occupational stressors in residential care work can have psychological impacts on workers, including burnout, secondary traumatic stress, depression, and anxiety (Santos et al., 2023; Seti, 2008; Sprang et al., 2011; Steinlin et al., 2017). Burnout can be particularly problematic for residential care workers, with high emotional demands and workloads resulting in feelings of mental and physical exhaustion, in addition to emotional exhaustion, depersonalization of care recipients, and a diminished sense of personal accomplishment (Seti, 2008). Both locus of control (i.e., the level of control a worker believes they have over their work and work environment) (Jorif, 2018; Schmitz et al., 2000) and perceived supervisor support (Rhoades & Eisenberger, 2002) can impact the degree of burnout experienced by a working population. As such, it is critical to understand the impact of these potential countermeasures in residential care workers.

Burnout in Residential Care Settings

While any role in child protection could be considered challenging, it is clear that work in a residential care environment is a highly demanding role (McLean, 2019). Ongoing stressors experienced by residential care workers can include vicarious trauma or secondary traumatic stress (i.e. hearing about the traumatic experiences of young people), as well as exposure to episodes of anti-social, threatening, or aggressive behaviors (i.e., by youth who are externalizing their trauma, anxiety, or depression) (Hughes, 2004; Li et al., 2019; Santos et al., 2023). Burnout can occur in response to chronic workplace stressors and includes feelings of emotional exhaustion (including feelings of emotional and physical depletion), depersonalization (negativity and detachment toward job), and a lack of accomplishment (general feelings of incompetence in the workplace) (Maslach et al., 1997). Emotional exhaustion is the most common aspect of burnout and occurs when an individual is faced with overwhelming job demands (Maslach et al., 2001). This is proposed to be the case for residential support workers, whereby emotional exhaustion depletes their ability to interact with clients in an emotionally responsive manner. Depersonalization develops as an attempt to create an emotional distance from clients that enables staff to protect themselves from intense emotional reactions (Maslach et al., 2001). It is thought that individuals experience a lack of accomplishment, the final core dimension of burnout, when their perception of effectiveness in their role is reduced because of feeling exhausted or indifferent (Maslach et al., 2001).

Job Demand-Control-Support Model

The Job Demand-Control-Support (JDCS) model has been used to examine the underlying elements that contribute to burnout in the workplace (Johnson & Hall, 1988; Pinto et al., 2014). The JDCS proposes that burnout occurs when individuals in high demand roles also have limited control and social support in their own roles (Johnson & Hall, 1988). Several reviews of the literature have linked burnout in high demand-low control and high demand-low control-low support roles (Häusser et al., 2010; Van der Doef & Maes, 1999). Van der Doef and Maes (1999) found that high demand roles are more likely to be associated with job-related burnout. Additionally, individuals working in high demand-low control-low support roles experienced the lowest levels of wellbeing and higher levels of emotional exhaustion that is characteristic of burnout (Pinto et al., 2014; Van der Doef & Maes, 1999). In their review, Van der Doef and Maes (1999) identified that the way in which the individual experiences both control and support is likely to impact burnout.

Locus of Control

According to (Spector, 1998), level of control in the workplace must be considered in relation to objective control and perceived control. Objective control refers to the amount of control the worker is permitted by the role and the procedures that define the role, whereas perceived control refers to the individual’s beliefs in relation to the amount of choice they have in any given workplace scenario. This perceived control is largely referred to as “locus of control” (Rotter, 1966; Spector, 1998). Rotter (1966) describes locus of control as the degree to which individuals expect that an outcome of a situation is dependent on their own internal characteristics versus the outcome of external factors such as chance, luck, or other individuals or circumstances. It is argued that this is due to the individual’s expectation that reinforcements will be associated with particular behaviors or events (Rotter, 1966). Therefore, those with an internal locus of control are more likely to expect that achievements, rewards, and other events happen as a result of their own behavior rather than luck (Rotter, 1966). Maslach et al. (2001) highlights that this sense of achievement or reward for hard work is what creates a sense of efficacy in individuals, the opposite of which is considered an indicator of burnout. Therefore, individuals with external locus of control beliefs may be less likely to consider that rewards are a result of their own actions, and more likely to experience burnout as they miss opportunities to experience a sense of accomplishment from their efforts (Maslach et al., 2001).

Work Locus of Control and Burnout

Locus of control is considered a situational trait, whereby individuals’ perceptions of control are different in varying areas of their life (Rotter, 1966; Spector, 1998; Tong & Wang, 2012). As a result, Spector developed the concept of work locus of control, a dimension of Rotter’s (1966) theory which simply specifies locus of control in the workplace. Work locus of control is considered an unfixed trait, in that an individual’s locus of control beliefs may change depending on their long-term experiences in the workplace (Wilski et al., 2015). Furthermore, locus of control and employee’s psychological attachment to the workplace have been theorized to be linked – whereby perceptions of organizational membership, support and care for employees can increase an employee’s locus of control and attachment to the workplace. This in turn then increases intention to stay and decreases turnover of staff. Conversely, inflexible management or strict procedures may lead to the development of an external work locus of control, while an internal locus of control may develop if an individual’s workplace autonomy is supported (Wilski et al., 2015). As such, it would be expected that residential support workers have more external locus of control beliefs due to their experience of limited role autonomy (i.e., the degree of independence a worker has regarding their work tasks and/or conditions) and decision-making that stems from the requirement to abide by legislation enforced by statutory child protection services (McLean, 2019). Other impacts on work locus of control in residential care settings include the heavy workloads due to child to caregiver ratios and impacts of shift work (as a result of 24-h care) (Santos et al., 2023).

Work locus of control and burnout have been the focus of several studies, which typically demonstrate a positive relationship between external locus of control and burnout (Jorif, 2018; Schmitz et al., 2000). Specifically, a study involving German nurses identified an association between higher levels of burnout and an external locus of control (Schmitz et al., 2000). Furthermore, an internal locus of control was considered positive for mental health professionals in the United States (Koeske & Kirk, 1995). A review by Seti (2008) also indicated individuals with an internal locus of control experienced lower levels of burnout and reported greater overall life satisfaction. Similarly, locus of control and burnout were found to be associated in a study of professional school support staff in the United States working with students who have emotional and behavioral disorders (Jorif, 2018). Support staff who had an internal locus of control had lower emotional exhaustion and depersonalization (Jorif, 2018). Qualitative findings in this study also highlighted that support staff with an internal locus of control were more likely to focus on controlling their stress, and by extension their potential for burnout, by utilizing personal coping strategies (Jorif, 2018). Therefore, the impact of an individual’s locus of control must be considered as equally relevant as objective control (factors the worker can control within their role as per organizational structures and procedures) in relation to burnout (McLean, 2019). Furthermore, both the locus of control and burnout may be directly related to worker retention (DePanfilis & Zlotnik, 2008; Ng & Butts, 2009). Within the residential care environment, worker retention is of critical importance due not only to the effects on workload for remaining staff but for the psychological wellbeing of children and youth in care (DePanfilis & Zlotnik, 2008).

Supervisor Support and Burnout

The JDCS model also highlights the importance of social support in reducing the impact of job demands (Johnson & Hall, 1988). Supervisor support is a key strategy implemented by many residential care organizations to support their employees (McLean, 2019; Van der Doef & Maes, 1999). According to a meta-analysis conducted by Rhoades and Eisenberger (2002), employees who feel highly supported by their supervisor, and by extension their organization, report higher job satisfaction, and experience less physical symptoms of stress, such as fatigue and burnout. They explain that as supervisors act as agents of the organization, having responsibility for directing and evaluating subordinates’ performance, employees view their supervisor’s favorable or unfavorable orientation toward them as indicative of the organization’s support (Rhoades & Eisenberger, 2002). According to organizational support theory, personification of an organization by an employee occurs by assigning human-like characteristics such as legal, moral, financial responsibility, policies and cultural norms, and the power that the organizations’ agents (such as a supervisor) exert over employees (Levinson, 2009). Lakin and colleagues (2008) discuss social support from coworkers and supervisors as a key coping strategy in social service environments, yet that often staff do not report feeling listened to or valued by their superiors. In their study conducted with residential inpatient treatment center staff in the United States, workers who felt that they received higher levels of support from their supervisors reported lower levels of burnout (Lakin et al., 2008). Similarly, Mack and Rhineberger-Dunn (2021) found that community corrections staff who reported higher levels of supervisor support reported less emotional exhaustion and depersonalization. Given the potential similarities between community corrections work and residential care work (i.e., managing challenging behaviors (Lakin et al., 2008; McLean, 2019), it is likely that supervisor support may also be an effective strategy to mitigate the risk of burnout in residential care workers.

The Present Study

Research involving individuals employed in complex care-giving fields such as mental health professionals (Seti, 2008), nurses (Schmitz et al., 2000), juvenile inpatient treatment center staff (Lakin et al., 2008), and community corrections officers (Mack & Rhineberger-Dunn, 2019) highlights that there is an important relationship between work locus of control, perceived supervisor support, and burnout. However, these relationships have not been investigated in residential support workers who provide care to children in an Australian context. As such, this study will look to answer the following question: How are work locus of control and perceived supervisor support associated with burnout in residential support workers in Australia?

The current literature regarding burnout, and the impact of locus of control, and supervisor support has informed the following two hypotheses:

• H1. Higher scores on the Work Locus of Control Scale (representative of external locus of control) will predict higher levels of work-related burnout scores in residential support workers.

• H2. Lower scores on the Survey of Perceived Organisational Support will predict higher levels of work-related burnout scores in residential support workers.

Method

Design

A cross-sectional online survey was undertaken to evaluate burnout in a sample of residential support workers. Data collection was undertaken via an anonymous survey created using Qualtrics (Provo, UT). Ethics approval was obtained from the Central Queensland University Human Research Ethics Committee (2021–046) prior to participant recruitment.

Participants

Participants (n = 50) were Australian child protection residential support workers. All participants were employed in a residential facility caring for young people at the time of survey completion. Individuals under the age of 18 years were excluded from participation. Two methods of recruitment were employed to seek participants. First, recruitment was undertaken through two Queensland-based non-for-profit residential organizations who sent study information to worker mailing lists. Individuals then chose whether or not to participate (regardless of this link being provided by their workplace) therefore the workplace did not have any indication if workers did or did not complete the survey (i.e., participation was completely voluntary). Second, participants were also recruited via social media, and an e-mail invitation was distributed through the relevant industry association mailing lists.

Procedure

Upon opening the online survey, participants were provided an information sheet which described the study and advised the survey would take approximately 20 min to complete. Participants were informed that their responses would be completely anonymous and closing the web browser prior to completion of the survey would indicate their wish to have their partial data excluded from the final analysis. Participants were informed that they would not be able to withdraw their responses after survey completion as the data would be deidentified, and that progressing to the next stage of the survey would be an indication of informed consent.

Measures

Demographics

All participants were asked to confirm that they were over 18 years of age and were currently employed as a residential support worker working with young people. Those that responded no to either of these questions were excluded. Additional demographic questions included age, gender, and years of experience in their current role. As the research student involved in data collection was employed in one of the organizations involved in this study, demographic information regarding participant location was not collected to protect participant confidentiality. This limited the opportunity to consider the data on a state-by-state basis.

Work Locus of Control

Residential support workers’ locus of control in the workplace was assessed by the short version of Spector’s (1988) Work Locus of Control Scale. This scale is widely used to determine locus of control specific to the work environment and has been shown to relate, as expected, to feelings of control at work (Spector, 1988). Internal consistency is high for this scale, with a coefficient alpha of .81 for samples in Western countries including Australia. This scale comprises eight questions that allow responses on a 6-point scale including disagree very much, disagree moderately, disagree slightly, agree slightly, agree moderately, and agree very much (producing scores of 1–6, respectively). Four items on this scale represent internal locus of control such as “On most jobs, people can pretty much accomplish whatever they set out to accomplish”, and four statements reflect external locus of control such as “Promotions are usually a matter of good fortune”. In line with the standard use of this scale, participants were asked to consider their beliefs about jobs in general, not only their present job, while providing their responses. Questions addressing an internal locus of control were reverse scored and added to the scores from the externally worded questions to provide a total score between 8 and 48 (Spector, 1988). Higher total scores represent having an external work locus of control while lower total scores represent an internal work locus of control (Spector et al., 2002).

Perceived Organisational Support

Participants’ perceptions surrounding supervisor support were measured using an adapted short form of the 36-item Survey of Perceived Organizational Support (Eisenberger et al., 1986). This scale consisted of eight items, each on a 7-point Likert scale with possible responses ranging from strongly disagree to strongly agree. The term “organisation” was replaced with “supervisor” to measure employee perception of the support received from their supervisor, in line with previous research (Eisenberger et al., 1986). As identified in Rhoades and Eisenberger (2002), often employees personify the organization and the level of support received by their supervisor tends to reflect their overall view and perception of organizational support. Participants were instructed to consider their current supervisor for this scale. Items on the scale include statements such as “My supervisor really cares about my wellbeing”. Half of the statements are negatively worded, such as “Even if I did the best job possible, my supervisor would fail to notice”. Questions 17, 20, 22, and 24 are scored 0 (strongly disagree) to 6 (strongly agree) while questions 18, 19, 21, and 23 are scored 0 (strongly agree) to 6 (strongly disagree). Scores from each question are totaled to provide an overall score between 0 and 48; high scores represent high levels of supervisor support, while low scores represent low support (Eisenberger et al., 1986). The Survey of Perceived Organizational Support has high internal validity with Cronbach’s alpha at 0.95 (Shore & Tetrick, 1991).

Burnout

The Copenhagen Burnout Inventory was used to measure burnout (Kristensen et al., 2005). This scale consists of three subscales: personal burnout, work-related burnout, and client-related burnout. The scales for personal burnout and client-related burnout consisted of six items on a 5-point Likert scale ranging from always/to a very high degree to never/to a very low degree. The work-related burnout subscale consisted of seven items measured similarly. Example items from the personal, work-related, and client-related subscales include questions such as “How often do you feel tired?,” “Are you exhausted in the morning at the thought of another day at work?,” and “Do you find it hard to work with clients?,” respectively. Participants were instructed to consider their feelings in relation to their current role. A total score between 0 and 100 for each subscale was obtained by determining the mean score of the responses provided, with higher scores representing higher levels of burnout. Individuals are considered moderately burnt out with scores between 50 and 74, highly burnt out with scores between 75 and 90, and severely burnt out with scores of 91–100 (Creedy et al., 2017). Kristensen et al. (2005) determined a high internal validity for the Copenhagen Burnout Inventory, reporting that Cronbach’s alphas for the three subscales ranged from 0.85 to 0.87.

Data Analysis

Data were analyzed using IBM SPSS Statistics (Version 28). To be included in the analysis, participants were required to have completed all questions. Two separate two-step hierarchical regression analyses were performed to identify the relationships between the independent variables and work-related burnout (i.e., the work-related burnout subscale was utilized for the regression analyses). Age and gender were included in step one of both analyses to control for their potential impact. Years of experience were not included in the analysis as these data were colinear with age. Linearity, normality, and homoscedasticity of residuals as well as the Durbin–Watson statistic were tested to assess the basic assumptions of a multiple regression analysis. Statistical significance was set at p ≤ .05.

Results

Demographics

In total, 93 participants accessed the anonymous survey link. Thirty (32.3%) participants were screened out of the dataset as they were not over 18 years old, they were not currently employed as a residential support worker, or they failed to respond to the screening questions. Of the 63 individuals who qualified to participate in the survey, a further 13 failed to complete all questions. This resulted in a total of 50 responses included in the final analysis. The participants were aged between 19 and 56 years (M = 37.4, SD = 11.4 years). Residential support workers’ level of experience ranged from 1 month to 15 years (M = 3.2, SD = 3.3 years) in their current role. Respondents included 37 (74%) females and 13 (26%) males. Figure 3 outlines the process from participant recruitment to final data analysis.

Power Analysis

A post hoc power analysis was conducted using GPower software (Faul et al., 2009). A post hoc analysis was chosen instead of a priori to estimate power based on actual effect size. Analyses were conducted with a participant sample size of 50, and 3 predictors (age, gender, and work locus of control/perceived supervisor support). The alpha level used was p = .05, and the effect size used was f² = 0.21 (perceived supervisor support) and f² = 0.31 (work locus of control). The post hoc analysis demonstrated that the analysis was adequately powered (0.84) for the effect size of 0.31 identified in the regression analysis of work locus of control scores and burnout (Cohen, 1992). It was identified that the regression analysis for perceived supervisor support and burnout was similarly adequately powered (0.97) for the effect size of 0.21 identified (Cohen, 1992).

Burnout

Responses to the Copenhagen Burnout Inventory (Kristensen et al., 2005) were separated into three subscales: work-related burnout, client-related burnout, and personal burnout. The participants reported a mean total score of 63.3 (SD = 23.8) out of a possible score of 100 for the work-related burnout subscale. Mean scores for client-related and personal burnout subscales were 44.9 (SD = 24.0) and 59.5 (SD = 24.0), respectively.

Work Locus of Control and Burnout

It was hypothesized that higher scores on the Work Locus of Control Scale (i.e., external locus of control) would predict higher work-related burnout scores for the residential support workers. Residential support workers reported a mean score of 21.6 (SD = 6.4) on the Work Locus of Control Scale, indicating that responses were slightly skewed toward internal locus of control. Figure 1 presents a summary of the work locus of control scores in relation to work-related burnout scores.

Figure 1. Relationship between work-related burnout scores and work locus of control scores.

The relationship between work locus of control scores and work-related burnout scores was assessed by a hierarchical regression analysis. Variables were tested in the following order: age and gender, then work locus of control score. Work locus of control scores had a statistically significant positive relationship with work-related burnout scores after controlling for age and gender, F(1, 43) = 15.84, p = .001. Work locus of control scores accounted for 25.7% of the adjusted variance in work-related burnout scores. See Table 1 for the full model.

Table 1. Hierarchical multiple regression of variables of work-related burnout: work locus of control.

	b	SE B	β	p
Step 1
Constant	71.11 (29.77, 118.86)	22.04		<.001
Gender	4.28 (−14.36, 20.91)	8.70	0.08	.62
Age	−0.38 (−0.95, 0.26)	0.30	−0.19	.19
Step 2
Constant	35.55 (−7.69, 86.93)	23.99		.16
Gender	−0.09 (−16.71, 15.48)	8.00	−0.00	.99
Age	−0.31 (−0.84,0.24)	0.27	−0.16	.26
Work Locus of Control Score	1.88 (0.93, 2.84)	0.47	0.52	.002

With 95% bias corrected confidence intervals reported in parentheses. Confidence intervals and standard errors based on 100 bootstrap samples.

Perceived Supervisor Support and Burnout

The second hypothesis of the study stated that lower scores on the Survey of Perceived Organisational Support scale would predict higher work-related burnout scores for the residential support workers. Residential support workers reported a mean score of 29.7 (SD = 13.1) for perceived supervisor support. Figure 2 presents the relationship between perceived supervisor support scores and work-related burnout scores.

Figure 2. Relationship between work-related burnout scores and supervisor support scores.

Figure 3. Flowchart from participant recruitment to final data analysis.

A hierarchical regression analysis was utilized to evaluate the relationship between perceived supervisor support scores and work-related burnout scores. There was a negative relationship between perceived supervisor support and work-related burnout, F(1, 43) = 8.60, p = .005. After controlling for the effects of age and gender, perceived supervisor support scores accounted for 15.2% of the adjusted variance in work-related burnout scores. See Table 2 for the full analysis.

Table 2. Hierarchical multiple regression of variables of work-related burnout: perceived supervisor support.

	b	SE B	β	p
Step 1
Constant	71.11 (29.98, 112.24)	23.92		.008
Gender	4.28 (−12.03, 20.60)	9.17	0.08	.65
Age	−0.38 (−0.99, 0.23)	0.31	−0.19	.23
Step 2
Constant	96.60 (54.75, 138.45)	21.41		<.001
Gender	0.50 (−14.79, 15.80)	8.15	0.01	.96
Age	−0.32 (−0.88, 0.25)	0.31	−0.16	.30
Supervisor Support Score	−0.82 (−1.38, −0.26)	0.29	−0.41	.01

With 95% bias corrected confidence intervals reported in parentheses. Confidence intervals and standard errors based on 100 bootstrap samples.

Discussion

The aim of this study was to understand the association between locus of control, perceived supervisor support and burnout in Australian child protection residential support workers. The findings identified that residential support workers with a more external work locus of control are more likely to experience burnout, supporting hypothesis one. The findings also indicated that residential support workers who feel more supported by their supervisor are less likely to experience burnout than those who feel less supported, supporting hypothesis two. Additionally, the results demonstrated that over half (72%) of the residential support workers who participated were experiencing moderate-to-severe levels of work-related burnout (Creedy et al., 2017).

The finding that having an external locus of control is associated with a greater degree of burnout in residential support workers is consistent with previous research involving other high-stress professions such as mental health professionals (Seti, 2008), nurses (Schmitz et al., 2000), and school support staff (Jorif, 2018). Given that the most common goal for residential facilities is to work with high-risk youth to help in their recovery from trauma (McLean, 2019), it is not surprising that residential support workers experience burnout. However, these results demonstrate that residential support workers who display external locus of control beliefs may be more vulnerable to the workplace stress that contributes to burnout (Schmitz et al., 2000; Wilski et al., 2015). Specifically, individuals with an external locus of control may be more likely to perceive that stressors are outside of their control (Tong & Wang, 2012). Therefore, these individuals may perceive that they are unable to change their experience (regardless of whether this is actually the case), increasing the likelihood that they will not attempt to address these stressors (Tong & Wang, 2012). Stressors can become chronic in nature when unaddressed and can further contribute to the development of burnout (Maslach et al., 2001; Tong & Wang, 2012). Furthermore, individuals with external locus of control may be less able to foresee the rewards or outcomes of their behaviors in the workplace, and therefore may be less likely to pursue the behaviors that provide a sense of achievement. Despite the trend toward having an internal locus of control in our sample, 72% (n = 36) participants had scores that indicated their work-related burnout was at or above moderate levels (Creedy et al., 2017). Our findings therefore highlight that while those with external work locus of control beliefs are more likely to experience higher levels of burnout, internal work locus of control beliefs does not completely prevent residential support workers from developing some level of burnout.

Lower perceived supervisor support scores predicted higher work-related burnout scores, whereby individuals who felt less supported by their supervisors reported higher levels of burnout. These results are consistent with previous studies involving other high-stress professions such as residential inpatient treatment center staff and community corrections officers (Lakin et al., 2008; Mack & Rhineberger-Dunn, 2019). Residential support workers are tasked with providing care to young people experiencing complex mental health support needs and challenging behaviors that pose a risk of harm to staff (McLean, 2019), and as such reflect a similarly stressful working environment. Our findings therefore suggest that particularly within challenging or stressful work environments, supervisor support may play a role in reducing worker burnout. This may include reducing or preventing some of the negativity and detachment that residential support workers report feeling because of intense emotional arousal and physical threats (McLean, 2019). In their meta-analysis, Slemp et al. (2018) highlight that supervisor support that encourages workers’ sense of autonomy is correlated with positive outcomes considered contraindicators of burnout (Maslach et al., 2001), including: employee wellbeing, work engagement, job attitudes, job performance, and reduced turnover intentions. An increased sense of autonomy resulting from supervisor support may explain why residential support workers who feel supported by their supervisors report lower levels of burnout.

The findings of the present study are consistent with the current literature surrounding burnout, work locus of control, and supervisor support (Maslach et al., 2001). The results build on existing evidence that individuals in high demand roles (in this case child protection residential support workers), with external locus of control beliefs and limited supervisor support, are more likely to report higher levels of burnout (Johnson & Hall, 1988). Our findings should be considered when aiming to reduce burnout in residential support workers. Firstly, residential organizations should focus on worker wellbeing initiatives that involve creating role autonomy or a greater sense of control in their workplace (McLean, 2019). Working conditions that support the development of an internal locus of control through involvement in decision-making, practice support, and reflective coaching (McLean, 2019) should be encouraged (Wilski et al., 2015). Furthermore, organizations should encourage supportive relationships between staff and supervisors, including the development of targeted supervisory training programs. Supportive supervisory relationships may also be used to identify staff who are at risk of developing burnout. Early identification and intervention with increased supervisory support, potentially in the form of debriefing, reflective practice discussions, and/or increased supervision meetings, may reduce burnout (Maslach et al., 2001; McLean, 2019).

While the present study has provided important information on burnout in Australian residential support workers, there are some limitations that must be considered. In particular, the moderate sample size may limit generalizability of findings. However, given that statistical analyses were appropriately powered, it is likely that these findings could be applicable to the broader residential support worker community. Despite this, future research would ideally include a greater sample size. Another limitation of this study concerns the broad method of recruitment of participants across the country. In Australia, child protection laws are the responsibility of state governments to legislate (Australian Institute of Health and Welfare, 2021). It is important to recognize that confounding variables such as child protection regulations specific to each state may potentially have an underlying impact on the relationship between the variables involved in this study. As residential support workers have reported feeling that legislation surrounding legal authority was one of the factors that limited their sense of control in the workplace, it is possible that different state legislation allows different degrees of perceived control in decision-making. The method of recruitment may also have resulted in some oversampling from certain organizations, and limited sampling from the general residential support worker population. Increased and more specific sampling of participants from different states is therefore recommended for future research to understand the impact of state-based factors on supervisory support, locus of control, and burnout. We did not collect demographic data on cultural background, this has limited our ability to determine the representation of cultural backgrounds in our sample. This may be a consideration for future research. Furthermore, the range of experience of residential care workers was quite vast in our sample (ranging from 1 month to 15 years). Further research into the rates of burnout in this population could be conducted, however may be difficult to collect as the turnover rate of staff is so high in residential care – many workers may indeed be exiting the workforce due to burnout.

Conclusion

The need for Australian child protection residential facilities is increasing (Australian Institute of Health and Welfare, 2021). It is therefore critical to understand how different internal and external factors contribute to burnout in residential support workers, to both support mental health and worker retention. This study established that both work locus of control and perceived supervisor support are associated with burnout in these workers. Participants with an external locus of control are more likely to experience burnout, while participants who do not feel supported by their supervisor are also more likely to report high burnout. These findings indicate that residential organizations may be able to implement strategies to reduce burnout by externalizing workers’ locus of control and improving supervisor support. Furthermore, encouraging supervisors to develop practices that provide their staff with a sense of emotional and instructional support may improve the wellbeing of those in the residential care environment. Future research should look to provide more insight into the development of locus of control, coping styles, and supervisor support styles, and how knowledge of these factors can be used to help organizations reduce or prevent burnout in this work environment. Reducing staff burnout in a vital industry that is aimed at protecting vulnerable young people has countless benefits for the individuals, their colleagues and employers, and the young people in their care.

Acknowledgments

We would like to acknowledge all the participants who completed the survey.

Disclosure statement

No potential conflict of interest was reported by the author(s).