Social workers offer support and counselling for various groups of clients with regard to the prevention, reduction, and coping with social problems. Their tasks mainly involve person-related services, such as consultancy, education, training, and representation [1
]. Two increasingly important areas of social work are refugee and homeless aid. The growing number of people being forcibly displaced represents a social problem with a worldwide scope. According to recent estimates, there are currently 25.9 million refugees worldwide [2
]. Additionally, it is estimated that about 100 million people are without a place to live [3
]. Although there are no official records available for Germany, data provided by the German Federal working group for homeless aid show that there were about 650,000 homeless people in Germany in 2017, of which about 375,000 people had been recognized refugees [4
]. Due to domestic migration and the influx of refugees, there are currently about 1.5 million refugees and asylum seekers in Germany [5
Both homeless people and refugees are groups of persons with increased vulnerability. As recent studies show, social workers serving homeless individuals and refugees experience several job demands, such as the confrontation with a bureaucratic system, high caseloads, experience of prejudices from other persons against their clients, as well as clients’ suffering from traumatic experiences. The latter can also become apparent during counselling services and thereby affect social workers’ daily working tasks or even cause secondary traumatization [6
]. However, social workers described several job resources, too. These primarily referred to the meaning of their work as well as social support from colleagues [7
The present study expands upon existing results by addressing the question how social workers’ job demands (quantitative and emotional demands) and resources (meaning of work and social support) as well as personal resources (resilience) are related to work-related outcomes (work engagement) and strain reactions (personal burnout). The variables were chosen based on previous research findings demonstrating their particular relevance for social workers in refugee and homeless aid [6
]. Specifically, a recent qualitative study was able to show that social workers perceived high quantitative and emotional demands and cited social support as a crucial job resource [8
Quantitative demands can be defined as the extensive and intensive demands inherent to one’s work (e.g., work hours, pace, and load). Positive associations between quantitative demands and workers’ stress levels were found for several occupations [9
]. With regard to emotional demands, Steinberg and Figart point out that emotional labour encompasses reading emotions of others as well as managing one’s own and others’ emotions [13
]. A meta-analysis by Hülsheger and Schewe revealed substantial associations of emotional labour with indicators of impaired well-being [14
]. For example, in one study, a positive association between surface acting (as a form of emotional labour) and emotional exhaustion among health care providers was revealed [15
]. The job resource meaning of work is composed of the job meaning, the role meaning and the self-meaning at work [16
]. Based on the framework of Rosso et al., the key dimensions for meaning of work include the approach and pursuit attached to work as well as direction of action towards the self or others [17
]. Social support as another important job resource describes the availability and quality of helping relationships [18
]. Evidence suggests positive associations between social support and workers’ health [19
] as well as direct positive effects of social support on employees’ strain [22
]. Two categories of support systems can be assumed for social workers: formal support via management and supervision as well as informal support provided by colleagues or friends [24
]. Work engagement represents a work-related outcome characterized by a positive, fulfilling, work-related state of mind [25
]. It is a distinct motivational construct that is related to contextual and individual factors, such as task variety and significance [26
]. It consists of three key components (vigour, identification, and absorption) and can be understood as the opposite of burnout [25
]. In contrast, burnout has been defined as “a state of physical, emotional and mental exhaustion that results from long-term involvement in work situations that are emotionally demanding” [27
] p. 501. Kristensen and colleagues further describe fatigue and emotional exhaustion as the core of burnout [28
Social work is often performed in settings characterised by regular changes, e.g., from recently instituted legislation, policies, and practices [24
], and has been found to be highly stressful [30
]. In view of this, resilience as the ability to adapt to stressors and bounce back from negative experiences plays an important role in social work. In line with the original concept presented by Wagnild and Young, resilience is defined as a trait and personal resource enabling flexible adaption to adverse situations and moderating the effects on negative emotions and stress [31
]. Resilience can be understood in terms of two dimensions: (1) personal competences such as self-reliance and independence as well as (2) acceptance of self and life with attributes such as adaptability and flexibility [31
1.1. Theoretical Background
The Job Demands–Resources (JD-R) model by Bakker and Demerouti [33
] may be used in order to explain the underlying mechanisms with regard to the links between social workers’ working conditions, personal, and work-related outcomes. According to the JD-R model, aspects of the job that require physical or mental effort can be considered as job demands, which are positively related to the depletion of health (health impairment process). In contrast, job resources are assumed to reduce job demands and their adverse effects, and to play an important role in increasing employees’ motivation and work engagement (motivational process). The proposed effects of the JD-R model have been empirically proven. Job demands were found to be related to strain reactions and negatively affected employees’ mental and physical health, whereas job resources were associated with positive health effects and employees’ work engagement [35
1.2. Study Aims
The aim of our study was to examine the links between social workers’ job demands (quantitative and emotional demands), job resources (meaning of work and social support), personal burnout, and work engagement. Moreover, we aimed to assess the role of resilience as a personal resource in the stressor–strain relationship.
Based on the empirical evidence and the JD-R model, the following hypotheses were tested:
Hypothesis 1 (H1).
Quantitative and emotional demands are positively related to social workers’ personal burnout (1a) and negatively related to social workers’ work engagement (1b).
Hypothesis 2 (H2).
Meaning of work and social support are negatively related to social workers’ personal burnout (2a) and positively related to social workers’ work engagement (2b).
Hypothesis 3 (H3).
Resilience is negatively related to social workers’ personal burnout (3a) and positively related to social workers’ work engagement (3b).
Hypothesis 4 (H4).
Resilience moderates the relationship between social workers’ quantitative and emotional demands and personal burnout.
summarizes the relationships between the variables in a conceptual model.
To our knowledge, this study is the first to assess the job demands and resources of social workers serving refugees and homeless individuals by testing their predictive value for social workers’ burnout and work engagement in a structural equation model. We were able to gain important new insights into these topics based on reliable statistical methods.
4.1. Working Conditions of Social Workers in Refugee and Homeless Aid
In comparison with a norm sample of social and educational professions in Germany [42
] as well as with the results of a recent interview study with social workers [8
], respondents of the present study described similar patterns with regard to the measured variables. Substantial correlations between social workers’ job demands, resources, burnout, and work engagement were revealed, which are consistent with the propositions of the JD-R model [33
]. Some results of the structural model provided support for the JD-R model as well, although not all of the proposed hypotheses were verified.
An important finding is that resilience as a personal resource had direct effects both on personal burnout and work engagement. The result is consistent with previous studies in which resilience showed negative effects on burnout and positive effects on work engagement [43
]. In general, resilient people believe that they are able to solve problems on their own, which strengthens their perceived ability to act and contributes to increased motivation and engagement. Referring back to its definition, resilience enables flexible adaption to stressors and adverse situations [31
]. Thus, possessing this personal resource may be particularly helpful in settings characterised by frequent changes and daily stressors, such as social work institutions [8
]. However, no moderating effects of resilience could be found, meaning that, regardless of social workers’ levels of resilience, quantitative and emotional demands were associated with burnout. The findings thereby confirm the health impairment process suggested by the JD-R model [33
]. Contrary to our hypotheses, the job demands did not affect work engagement.
In accordance with a qualitative study showing that employees in social work reported high demands but simultaneously showed a strong motivation and commitment to their work [8
], the present data confirmed meaning of work as an important job resource with a strong positive effect on work engagement. The finding is consistent with the preposition of the JD-R model, stating that job resources play an important role in fostering employees’ work engagement [33
]. The phenomenon of simultaneously experiencing high demands and a strong sense of commitment could particularly apply to social workers and should be taken into consideration for the development of health promotion measures for this target group.
In contrast to previous results from varying occupations [21
], social support did not show any direct effects on the studied outcomes in the structural model (although significant associations on a correlation level were found). In accordance with this, Kim and Stoner also found no evidence for a significant negative relationship of social support and burnout [47
]. Several reasons for the non-significant effect of social support in the structural model could be discussed, which may partly relate to methodological issues. For example, the finding could be linked to the variable’s operationalization: While the other variables in the structural model refer to personal or inner evaluations of work and demands imposed on the individual, social support more strongly represents interactions with others. Thus, it could be assumed that social support rather has a moderating influence in buffering the impact of job demands on the workers’ health, which was also shown in previous research [46
]. Moreover, the result might be linked to the difference in terms of how workers felt supported by their colleagues as opposed to their supervisors (the social support scale encompasses items for both groups). Respondents in our sample rated the support from their colleagues (M
= 79.01) significantly higher than the support from their supervisors (M
= 64.32). However, we did not assess the reasons why employees felt more or less supported by these entities. In future research, it would be interesting to examine how social support provided by workers at different hierarchical levels (supervisors and colleagues) impacts differently on health outcomes.
4.2. Strengths and Limitations
The strengths of our study are the systematic recruitment process, good rate of completed surveys from the participants, and use of advanced statistical modelling techniques. SEM combines the strengths of regression analysis, path analysis, and confirmatory factor analysis. Moreover, we used well-validated instruments that have previously shown strong validity and high internal consistency, as well as performed thorough reliability and validity checks prior to testing our models.
However, the cross-sectional study design does not allow to draw causal conclusions, and reverse causality in the relationships between the variables cannot be completely ruled out. Furthermore, data was assessed by self-report measures only, which may favour bias to the data. Another limitation could be that several institutions refused to participate in the study, while only few of them provided explanations for non-participation (e.g., not enough time due to high workloads). Since we were not able to assess potential differences between institutions who agreed and refused to participate, a potential non-response bias cannot be ruled out completely. Although a broad range of social workers (job location, type, and sponsor of institution) has been studied, representativeness across all social workers serving refugees and homeless individuals cannot be assumed.
4.3. Implications for Future Research and Practice
With respect to research implications, future studies should apply longitudinal research designs in order to examine short- and long-term dynamics between the variables studied and establish causality regarding their relationships. Furthermore, the influence of social support could be examined more closely and with regard to its moderating effects.
In terms of practical implications, our results indicate a need for the development of health promotion measures for social workers in homeless and refugee aid. It is recommended to implement both behavioural and structural workplace health promotion measures. Structural approaches should specifically target the reduction of employees’ job demands in order to diminish their potentially health-depleting effects. Since both emotional and quantitative demands had direct effects on burnout, which were not buffered by personal resource resilience, implies that employees’ job demands need to be targeted directly and reduced by appropriate measures (e.g., in terms of diminished workloads, changes in work tasks and hours, and additional staff). In addition, behavioural measures should be taken to foster employees’ resources. It seems especially valuable to strengthen social workers’ meaning of work and resilience, since both resources showed beneficial effects on employees’ work engagement and were negatively related to burnout. This could be done, for example, by offering tailored education, training, and counselling to the workers on different topics, such as resilience and self-care. Moreover, thematic workshops within or across teams could provide employees with a platform to reflect and become aware of their professional achievements, which could further strengthen their meaning of work. This could also help the workers to visualize the societal relevance of their activities and promote their job motivation and commitment.