Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Research article
The relationship between leader-member exchange and work engagement
in social work: A mediation analysis of job resources
Bettina Wagner
a
,
1
, Clemens Koob
b
,
*
,
1
a
Department of Social Work, Catholic University of Applied Sciences Munich, Munich, Bavaria, Germany
b
Department of Health and Nursing, Catholic University of Applied Sciences Munich, Munich, Bavaria, Germany
ARTICLE INFO
Keywords:
Leader-member exchange
Work engagement
Job resources
Team atmosphere
Work control
Social work
Leadership
ABSTRACT
Introduction: The working environment of social workers has become an important concern in research. Whereas
studies typically focus on negative aspects of these environments, this study investigates the possibility of creating
a positive environment by examining whether and how high-quality leader-member exchange (LMX) relation-
ships are related to social workers' work engagement. It is hypothesized that particularly relevant job resources
mediate the relation between LMX and social workers’work engagement.
Method: To identify those job resources that are particularly important to social workers, a qualitative preparatory
study was first conducted with social workers and social work executives. Based on that, study data were collected
by administering an online survey among social workers. Both parts of the study were realized at twenty different
advice centers for pregnancy issues of a leading social organization in Germany. The dataset for statistical ana-
lyses comprised 43 social workers in total. Regression analysis with parallel mediation was used to test the hy-
pothesized relationships.
Results: Social workers in high-quality LMX relationships operate in working environments with more abundant
social and structural job resources: they experience the team atmosphere more positively and have greater work
control. Further, it was found that experiencing a positive team atmosphere was associated with higher work
engagement, with team atmosphere fully mediating the relationship between LMX and work engagement.
Notably, high-quality LMX relationships do not seem to promote work engagement directly.
Discussion: The study underscores the relevance of high-quality LMX relationships in fostering a resourceful work
environment. This, in turn, contributes significantly to a high level of work engagement. Organizations and
managers in social work should therefore strengthen managerial relationships and create resourceful
environments.
1. Introduction
By solving individual and social problems, social work contributes to
the welfare of a society [1]. Hence, the quality of services provided by
social workers and social work organizations is of pivotal importance. It
is well-known from prior research in the services marketing domain that
service quality is essentially dependent on high-performing employees
[2]. In recent years, numerous studies in organizational psychology have
furthermore shown that employees' performance in delivering excellent
results to clients is in turn closely related to whether and to what degree
they have positive attitudes toward their jobs and are apt to engage in
behaviors to support the organization and its clients [3,4]. In other
words, the extent to which employees are actively engaged in their work
substantially influences the quality of services delivered. Work engage-
ment thereby denotes ‘a positive, fulfilling, work-related state of mind
that is characterized by vigor, dedication, and absorption’[5]. Vigor
refers to work-related energy and resilience, dedication to work-related
sentiments of pride and significance, and absorption to a particularly
focused and flow-like working mode in which disconnecting from
working appears difficult [5]. High work engagement implies the
simultaneous investment of personal physical, emotional, and cognitive
energies in working and thus contributes to high work performance [4].
Therefore, social work organizations should have an interest in
encouraging their employees' work engagement. In research on social
* Corresponding author.
E-mail address: clemens.koob@ksh-m.de (C. Koob).
1
These authors contributed equally to this work.
Contents lists available at ScienceDirect
Heliyon
journal homepage: www.cell.com/heliyon
https://doi.org/10.1016/j.heliyon.2022.e08793
Received 22 June 2021; Received in revised form 17 September 2021; Accepted 14 January 2022
2405-8440/©2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Heliyon 8 (2022) e08793
work, however, work engagement has so far received comparatively little
attention. One early exception is the investigation by Schaufeli and col-
leagues [6], which found that social workers experienced rather low
levels of work engagement in comparison to employees in other pro-
fessions such as education or police. Another and more recent exception
is Ravalier's [7] study that investigated the influence of work engagement
in social workers in England. The study demonstrated that higher work
engagement was significantly associated with a number of positive
work-related outcomes in social work, i.e., lower stress and turnover
intentions, less presenteeism and greater job satisfaction.
Regarding the antecedents of work engagement, previous research
in organizational psychology has demonstrated a pivotal role of em-
ployees' working environment [3,4,8]. The topic of ‘working envi-
ronment’has also become an important concern in research on social
work. Various studies examined the working environment of social
workers and came to the conclusion that social work is a very
demanding profession, characterized, e.g., by high psychological
workload and time pressure, role conflicts, workplace aggression, lack
of formal rewards, low social support, limited job autonomy and high
job insecurity, leading to high risks of developing burnout and other
mental and physical impairments of employees of social work organi-
zations [9,10,11,12,13]. However, social work research has so far
primarily focused on the negative aspects of working environments in
social work organizations and associated adverse effects and has paid
little attention to the possibilities of designing a working environment
that has a positive impact on the work engagement of employees of
social work organizations [14].
With work engagement being related to various positive outcomes,
a better understanding of the working environment factors contrib-
uting to work engagement in social work may help to further profes-
sionalize the management of social work organizations, and in turn to
improve the quality of services provided by those organizations. In
addition, a greater understanding of such factors may add to research
on working environments and work engagement in social work. Thus,
this study's aim is to investigate the possibility of creating a positive
working environment that nurtures employees' work engagement in
social work organizations. To do so, we draw on prior research in
organizational psychology that has confirmed that leaders are
commonly a critical element in organizations' working environments
[4]. Positive leadership has been related to work engagement [15,16,
17,18,19,20].
2. Theoretical background and research hypotheses
To investigate whether and how positive leadership is related to so-
cial workers' work engagement, this study draws on the leader-member
exchange (LMX) theory. This theoretical approach emphasizes that the
quality of dyadic relationships between leaders and employees is pivotal
[21]. Leaders are assumed to establish different exchange relationships
with their employees, potentially affecting how they exchange attention,
favors, and resources. If followers interpret the LMX relationship as
positive, this will unfold positive effects on employees’attitudes and
behaviors in general [21,22,23,24] and work engagement in particular
[25,26,27]. We expect this to hold for employees in social work orga-
nizations as well.
To investigate the role of the quality of leader-member relation-
ships in influencing social workers' work engagement, we build on the
‘leadership and work engagement’-framework developed by Decuy-
pere and Schaufeli [28]. Based on a review of well-established theories
and recent empirical research, this framework differentiates between
several mechanisms underlying the influence of leadership on work
engagement. It first proposes a direct interpersonal process by which
leaders directly influence work engagement through, among others, a
‘cognitive pathway’, i.e., social exchange. According to this social
exchange perspective [29], the LMX relationship between leader and
employee is an exchange relationship maintained by interdependence
and expected reciprocation of, e.g., favors, work, or support. Em-
ployees voluntarily reciprocate positive leadership behaviors, moti-
vated by the returns this is expected to bring and typically brings from
supervisors. Thus, when the LMX relationship is interpreted as positive
by employees, it is more likely that they will reciprocate with strong
work engagement in the form of vigor, dedication, and absorption
[28]. We expect this to hold for employees in social work organiza-
tions, too. Social work organizations' supervisors and employees
constitute interdependent dyads. When employees experience the
dyadic LMX relationship as positive, they reciprocate with work
engagement.
Hence, we propose:
H1.A high quality of leader-member exchange directly fosters work
engagement of employees in social work organizations.
The ‘leadership and work engagement’-framework further suggests
that leaders indirectly influence work engagement by, among other
things, a ‘material pathway’, i.e., via shaping work characteristics [28].
The job demands-resources (JD-R) theory [30,31], which is widely used
to explain the emergence of work engagement and has the advantage of
being flexible to be adapted to different work contexts [32], groups work
characteristics into two basic categories: job demands and job resources.
Demerouti and colleagues define job demands as attributes of the job that
require sustained physical, emotional, or cognitive effort and are thus
linked to physiological and psychological costs [33]. Job resources, on
the other hand, are defined as those physical, psychological, social, or
organizational job characteristics that are helpful in either achieving
work goals, reducing job demands and associated costs, or stimulating
personal growth, learning, and development [33]. Two different pro-
cesses are triggered by these two categories of work characteristics: one is
a process of health impairment, the other is a motivational process [30,
34]. In the former, excessive demands and inadequate resources can
cause burnout. In the latter, appropriate job resources can contribute to
work engagement. Decuypere and Schaufeli assume that leaders can
allocate job demands and resources to their employees, thereby alter
their employees' demands and resources situation, and in turn indirectly
affect employees' work engagement [28]. Based on that argumentation
and in line with prior research in the LMX domain [21,35], we expect
that better leader-member relationships should result in more job re-
sources being available to employees in social work organizations. More
abundant job resources should in turn positively influence the work
engagement of social work organizations' employees'. In this sense, Aiello
and Tesi [36], Mette et al. [37], Tesi [38,39] as well as Geisler and
colleagues [40] recently found that factors such as decision latitude,
meaning of work, coworkers' social support, group cohesion, quality of
communication processes with coworkers, social support from superiors
and general social community at work can act as job resources that foster
social workers' work engagement.
Taken together, we thus expect:
H2.The relation between LMX and social workers' work engagement is
mediated by particularly relevant job resources.
The proposed conceptual framework is summarized in Figure 1.
To test the proposed hypotheses, a cross-sectional mixed-methods-
design was chosen. Since, to our best knowledge, no previous studies
have systematically examined which job resources are particularly rele-
vant to social workers, a qualitative preparatory focus group study was
conducted first to identify those job resources. Based on this, the quan-
titative main study was realized to test the hypotheses. Both parts of the
study were realized at twenty different advice centers for pregnancy is-
sues run by a leading social service organization in the state of Bavaria in
Germany. The study was conducted in this domain because it belongs to
the area of ‘children and families’, which is traditionally one of the
largest fields of practice of social workers [41].
In the next section, we briefly review the ethical considerations
common to both parts of our study. Then, we describe the preparatory
focus group study. Subsequently, we turn to the main study.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
2
3. Ethical considerations
Prior to conducting the preparatory and the main study, a review of
the University Ethics Committee statutes indicated that no ethical review
of the research project was required. This decision was based on the fact
that the investigations did not involve any experimental manipulations
or vulnerable groups, that participants were assured of the confidenti-
ality of their participation and data, and that all results would be ano-
nymized. Additionally, the data were collected following the EU General
Data Protection Regulation. Before the study started, all participants
explicitly declared their informed consent, and participation was
completely voluntary.
4. Preparatory focus group study
4.1. Method
4.1.1. Procedure
First, the relevance of various job resources in social work was
investigated in a qualitative preparatory study using two focus groups
with social workers and social work executives. The purpose of the focus
groups was to identify the job resources that are particularly relevant in
social work and to distill the absolutely decisive ones from them. Focus
groups were used as methodology because we wanted to gain in-depth
insights into the importance of various categories of job resources in
social work and to obtain rich accounts of why they were rated as
important. Besides, it was reasoned that in a social work setting, where
actors are used to exchanging and reflecting, they might be willing to be
candid in their discussion.
Participants in the focus groups were selected from, in total, twenty
advice centers for pregnancy issues that belonged to a leading social
organization in the state of Bavaria in Germany. The centers offer
counseling services for pregnant women and families with children under
the age of three. Social workers are the only professional group
employed. They work as advisors and provide information and psycho-
social counseling for all pregnancy or family-related issues, e.g., fertility,
childlessness, abortion, social and financial help, prenatal diagnostics.
Some of them also hold sexual education workshops. Each counseling
center has one executive who is also a professional social worker and
counselor.
Before starting the study, all of the organization's executives and
employees were thoroughly informed about the research. An invitation
to participate in the focus groups was then sent by e-mail to all of the
social workers and executives of the organization. All participants
responded to this invitation and voluntarily took part in the focus groups.
Besides, all participants were promised that the results would be ano-
nymized and that their participation would be treated confidentially. The
sessions were characterized by extensive discussions, and each lasted an
hour.
4.1.2. Study participants
Five employees participated in the first focus group with social
workers, and six executives took part in the second focus group with
social work executives. All of the participants held a university degree in
social work. The participants had been working for the social organiza-
tion for between one and eighteen years, which made it possible to record
the perspectives of both very experienced actors and new employees. The
group facilitator was an academic with in-depth social work experiences.
4.1.3. Discussion guideline
The discussion guideline for the focus groups was developed in line
with the core questions regarding job resources and work engagement
proposed by Hakanen and Roodt [42]. Participating social workers were
asked to examine and discuss the aspects in their work that helped them
to succeed in work tasks, even in stressful circumstances. They were
requested to consider things that made them feel energetic, provided a
sense of pride and purpose, and enabled them to enjoy what they were
accomplishing. In other words, they were asked to recall and discuss all
that contributed to a sense of engagement. Finally, they were requested
to rank the identified job resources according to their importance for
work engagement. The same topics were discussed with the executives,
only that they were requested to evaluate them from the perspective of
the subordinate employees as the focus of this study was counselor's work
engagement.
4.1.4. Analytical strategy
Focus groups were recorded, the material was transcribed and anony-
mized, and thereafteranalyzed by qualitativecontent analysis as proposed
by Mayring [43]. During the analysis, categories were applied in a
deductive way to the material. The categories used were adopted from
Schaufeli [44], who, based on a comprehensive literature review, pre-
sented an array of 22 job resources, ranging from various social resources
(e.g., role clarity) to work resources (e.g., task variety), to organizational
resources (e.g., organizational justice), to developmental resources (e.g.,
performance feedback). To ensure that the application of categories was as
controlled as possible, all categories were described, and coding rules were
defined. After working through the focus groups’material, quantitative
steps of analysis were integrated. First, the frequency with which a cate-
gory occurred was recorded, as this is an additional indication of its rele-
vance. Second, the frequency analysis was supplemented by the ranking of
the job resources carried out in the focus groups.
4.2. Findings
The two focus group discussions univocally revealed that both social
job resources and structural job resources were regarded to be key for work
engagement. While social job resources relate to the supportive re-
lationships an individual has at work, structural job resources mainly
refer to job design aspects [45].
Figure 1. Conceptual framework.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
3
With respect to the first-mentioned main category of social job re-
sources, the focus groups' participants listed subcategories such as sup-
port from colleagues, support from supervisors, recognition, and respect.
However, particular importance as a key driver of work engagement was
attributed to team atmosphere. Detailed analysis of the focus group ma-
terial revealed that the key category ‘team atmosphere’was connected
with aspects such as mutual trust and understanding, constructive feed-
back, a fear-free environment, and collective willingness to help.
Regarding the latter group of job resources, i.e., the structural job
resources, the participants named, e.g., involvement in the decision-
making process, person-job-fit, availability of work equipment, or task
variety. However, work control was regarded as decisively contributing to
work engagement. With the key category ‘work control’, focus groups'
participants referred to the level of influence and freedom in one's work,
i.e., to aspects such as control over working time, the extent of influence
on work-related decisions, influence on what one does at work, or in-
fluence on how one accomplishes work tasks.
4.3. Interim conclusions
The findings of this preparatory focus group study are generally in
line with research on job resources which suggests that work engagement
is dependent on different categories of job resources such as social and
structural resources [44,45]. Results further suggest that within each of
these categories, team atmosphere and work control might be regarded as
job resources that are particularly relevant to social workers. Finally, the
results indicate that social workers’understanding of these two resources
largely corresponds to the aspects that are typically also referred to in the
literature [46,47].
The results of this preparatory study must, of course, be handled
prudently and carefully. Though focus groups allow for in-depth explo-
ration of responses, caution must be exercised regarding the generaliz-
ability of results. Future research should replicate this focus group study
in other social work settings and countries.
In conclusion, the purpose of this preparatory study as a precursor to
the main study was to identify job resources that are particularly relevant
in social work. Hence, in the subsequent quantitative main study, team
atmosphere and work control were considered as those job resources
potentially mediating the relationship between LMX and work
engagement.
5. Method of the main study
5.1. Procedure
Subsequent to the preparatory study, an online survey among social
workers was administered to gather the data needed to examine the hy-
pothesizedrelationships between LMX, prioritized job resources, and work
engagement. All social workers who worked as advisors in the twenty
pregnancy advice centers of the social organization with which this study
was carried out were eligible to participate. Social workers were asked to
provide a self-assessment of their work engagement and the exchange
relationship with their leader, as well as to rate the availability of those job
resources that were categorized as pivotal in the previous focus groups.
Before fielding the survey, all of the organization's executives and em-
ployees were thoroughly informed about the research. An invitation to
participate in the survey was then sent by e-mail to the 90 eligible social
workers of the organization. The invitation contained information about
the aim and procedure of the study and also includedthe link to the survey.
The questionnaires were filled out completely anonymously.
5.2. Study participants
The dataset for analyses comprised N ¼43 employees in total, cor-
responding to a response rate of 48%. The response rate was in line with
the level of participation that was expected based on meta-analyses of
response rates in organizational science [48]. The mean age of the par-
ticipants was 45.4 years (SD 9.67). The study participants were not asked
about their gender, as this question would have been problematic for male
participants from an anonymity point of view due to the great dominance
of female employees in the field under study. Regarding occupation, the
mean length of employment within the focal organization was 10.8 years
(SD 6.16). The duration of the working relationships between the social
workers and their supervisors ranged from 0 to 25 years, with a mean of
9.2 years (SD 7.31). The participating employees worked between 12 and
40 h per week with an average of 26.0 h (SD 6.75) per week.
5.3. Measurement of main variables
We employed a structured questionnaire and drew on measures from
prior research to collect primary data for the quantitative study.
To measure work engagement, the German version of the Utrecht Work
Engagement Scale (UWES-9) was used. The scale is composed of three
subscales for vigor, dedication, and absorption. These comprise three
items each, all measured on a 7-point Likert scale with a range from 1 ¼
never to 7 ¼always. The scale was considered appropriate for the present
study because of its high internal consistency and test-retest reliability
and discriminant, convergent, and construct validity in previous studies
[4,6,49]. We also found high internal consistency of the scale (
α
¼.90).
Leader-member exchange was measured using the validated German
version of the LMX7 scale that was developed by Schyns and Paul [50]
based on the English original from Graen and Uhl-Bien [23]. The LMX7
scale describes the relationship between manager and employee with
seven items that are rated on a 5-point Likert type scale. Prior research
demonstrated very satisfying psychometrical properties of the scale [50].
The internal consistency of the scale was very satisfactory in our study,
too (
α
¼.94).
Team atmosphere was found in the focus groups to be one of the two
pivotal job resources in social work. In the quantitative study, team at-
mosphere was recorded using the team atmosphere scale developed by
Z
arraga and Bonache [46]. Active empathy, lenience of judgment,
members’courage, mutual trust, and access to help in the work team
were measured with ten items on a 7-point Likert type scale (ranging
from 1 ¼totally disagree to 7 ¼totally agree). In previous research, the
scale was attributed good validity and acceptable reliability (
α
¼.85)
[46]. Since the scale items were only available in English, we translated
them into the German language in accordance with the guidelines of the
European Social Survey program for the translation of questionnaires
[51]. Compared to previous research, we found a slightly lower but still
acceptable internal consistency (
α
¼.75).
Work control emerged in the focus groups to be the second key job
resource in social work. In the quantitative study, job control was
measured with the influence at work and degree of freedom at work sub-
scales of the Copenhagen Psychosocial Questionnaire (COPSOQ) [47].
The measure consisted of 14 questions that were to be answered on a
5-point Likert type scale (ranging from 1 ¼never to 5 ¼always). We
applied this tool because the COPSOQ concept proved to be valid and
reliable in prior studies [47]. We also found sufficient internal consis-
tency of the scale (
α
¼.83).
Beyond the variables mentioned above, we also included control
variables in our statistical analyses. We followed the literature's guide-
lines for using control variables, recommending a focused approach so
that available degrees of freedom and statistical power are not unnec-
essarily lost [52,53]. Hence, age (years) and weekly working hours (binary
coded, <25 h/25 þhours) were incorporated as controls in the analyses,
as it is well established that these factors may be related to the variables
being investigated [6], potentially resulting in a ‘mixing of effects’.
5.4. Analytical strategy
All statistical analyses were performed using IBM SPSS Statistics 24.
We first calculated means and standard deviations, as well as Cronbach's
B. Wagner, C. Koob Heliyon 8 (2022) e08793
4
alpha coefficients for multi-item measures and bivariate correlations. To
test the hypothesized relationships between leader-member exchange, the
two key job resources team atmosphere and work control, and work
engagement, we then used regression analyses with parallel mediation.
Mediation analyses were conducted using Hayes' PROCESS Macro
version 3.5, model 4 [54], which employs ordinary least squares
regression and provides unstandardized path coefficients for the total,
direct, and indirect effects. Confidence intervals and inferential statistics
were computed using bootstrapping with 100000 samples and hetero-
scedasticity consistent standard errors. Effects were considered signifi-
cant if the p-value was <.05 or if bootstrap confidence intervals did not
include zero.
6. Results of the main study
6.1. Descriptive statistics and correlations
The means, standard deviations, Pearson's correlations, and Cron-
bach's alphas of the focal and control variables are reported in Table 1.
Contrary to expectations, work engagement was not significantly
related to LMX (r ¼.15, p ¼.34) and work control (r ¼.25, p ¼.10). As
expected, work engagement related positively to team atmosphere (r ¼
.40, p <.01), with the correlation coefficient indicating a moderate
relation [55] between the variables.
In addition, and as expected, LMX showed a strong correlation with
team atmosphere (r ¼.51, p <.01) and a moderate correlation with work
control (r ¼.32, p <.05). Besides, team atmosphere and work control
were positively related (r ¼.45, p <.01), with the correlation coefficient
indicating a moderate relation.
6.2. Hypotheses testing
We assumed a positive association between LMX and work engage-
ment in social work. Additionally, we predicted the key job resources
team atmosphere and work control to mediate the relationship between
LMX and work engagement of social workers. Figure 2 presents the
findings of the parallel mediation analysis.
The parallel mediation model accounted for a substantial proportion
of the variance in work engagement (R
2
¼.18). LMX was positively
related to team atmosphere (a
1
¼.37, p <.01), which, in turn, was
positively associated with work engagement (b
1
¼.59, p <.05). Further,
the analysis revealed a positive relationship between LMX and work
control (a
2
¼.24, p <.05), but no significant relationship between work
control and work engagement (b
2
¼.16, p ¼.55). As previously stated,
the significance of the indirect effects was determined via bootstrapping
[54], with 100000 bootstrapped samples and a 95 percent confidence
interval. The analysis revealed a positive, significant indirect effect of
LMX on work engagement through team atmosphere of a
1
xb
1
¼.22 with
a 95 percent confidence interval from .002 to .461, but no significant
indirect effect on work engagement via work control (a
2
xb
2
¼.04, 95
percent confidence interval from -.094 to .172). The total effect of LMX
on work engagement (c ¼.16, p ¼.46) and the residual direct effect of
Table 1. Descriptive statistics, Pearson's correlations, and Cronbach's alphas of study variables.
Variables M (range) SD Items 1 2 3 4 5 6
1. Work engagement 5.16 (1–7) .88 9 .90
2. Team atmosphere 5.63 (1–7) .58 10 .40 .75
3. Work control 3.49 (1–5) .58 14 .25 .45 .83
4. LMX 3.96 (1–5) .79 7 .15 .51 .32 .94
5. Age 45.44 (24–64) 9.67 1 -.04 .03 .17 -.06 –
6. Weekly working hours
a
.51 (0–1) .51 1 -.11 -.13 -.18 -.02 -.33 –
Notes:
a
Dummy coded. All |r| >.30 are significant at p <.05, all |r| >.38, p <.01. Cronbach's alphas for multi-item measures are depicted in italics on the diagonal.
Figure 2. Parallel mediation model.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
5
LMX on work engagement (c’¼-.10, p ¼.67) were also not significant.
Therefore, results from our parallel mediation analysis with age and
weekly working hours as covariates indicated that LMX was indirectly
related to social workers' work engagement through its relationship with
team atmosphere. However, work control did not mediate the relation
between LMX and work engagement. Further, LMX did not directly
promote work engagement but only indirectly via team atmosphere.
Hence, we found partial support for our Hypotheses. Table 2 summarizes
the results.
7. Discussion
7.1. Theoretical implications
The present study advances research on working environments in
social work. While previous research typically has focused on negative
aspects of these environments and associated adverse effects, we inves-
tigated the possibility of creating a positive working environment in
social work organizations that fosters social workers’work engagement.
In this regard, and with respect to the role of leadership, our study
indicated that social workers in high-quality leader-member exchange
relationships work in more resourceful work environments. This finding
is in line with the ‘leadership and work engagement’-framework's theo-
retical proposition that positive leadership is apt to improve employees'
resources situation [28] and with prior empirical studies [35,56]in
occupational fields other than social work, and thus substantiates the
pivotal role of leadership for establishing a beneficial working environ-
ment also for the domain of social work. Our results further indicated that
sound leadership relations are capable of positively impacting social job
resources as well as structural job resources, since social workers in
high-quality leader-member exchange relationships were found to
experience a better team atmosphere and also higher levels of work
control. Considering that we were only able to investigate this for one
selected resource in each of these resource categories, further research is
warranted in this regard, that examines the potential effects of good
leader-member exchange relations on other social and structural job
resources.
Second, we found high-quality leader-member exchange relation-
ships not only to contribute to more resourceful work environments but
also to higher work engagement through a better team atmosphere,
which is characterized by aspects such as mutual trust and understand-
ing, constructive feedback, a fear-free environment, and collective will-
ingness to help. This finding supports the line of reasoning in prior
research in other occupational fields than social work, that the avail-
ability of job resources is generally apt to foster work engagement [4,
30]. It also corroborates previous research in the social work domain
which concluded that social job resources foster social workers' work
engagement [36,40]. Yet, we did not find a positive indirect effect of
high-quality leader-member exchange relationships on work engagement
through higher levels of work control. At first sight, this seems to deviate
from the findings of several empirical studies in occupational fields other
than social work that have demonstrated the motivating capacity of work
control for work engagement (reviews can be found, e.g., at Christian
et al. [4] or Halbesleben [57]). However, prior investigations have found
that the association between work control and work engagement varies
substantially in strength, being sometimes rather weak [4] or even
negative [58]. Hence, our study's finding could suggest that in social
work social job resources (such as team atmosphere) could be more
important for work engagement than structural resources (such as
work-control). Future research could shed further light on this issue, for
example, by analyzing more closely whether and how different di-
mensions of work control (e.g., regarding method, scheduling, time, or
place of work) [59] relate to social workers' work engagement.
Third, and differently than expected, we did not find any evidence
that high-quality leader-member exchange relationships foster work
engagement directly. It seems to be the case that in social work, unlike in
other work areas, in which studies [4] have found a direct influence of
leader-member exchange relations on work engagement, beneficial
leader-member relationships themselves do not act directly
engagement-promoting for employees, but as an important enabler of
resource-rich work environments. This result is nevertheless compatible
with some other studies in the LMX domain which also showed only
indirect effects of leader-member relationships on workplace outcomes
[60,61]. It could imply that pursuing beneficial LMX relationships may
not be sufficient to foster engagement in social work, but that a necessary
condition is that beneficial leader-member relationships must also have
an impact on social job resources such as a better team atmosphere to
unfold engaging effects. Since some previous studies have also suggested
nonlinear relationships between LMX and workplace outcomes [62,63],
it could also be that there could be a nonlinear, U-shaped direct rela-
tionship between LMX and work engagement: social workers in
low-quality relationships with their superiors could be particularly
engaged to receive superiors’recognition, and social workers in excep-
tionally good relationships could also be particularly engaged. However,
a post hoc analysis that we conducted and that included a quadratic term
for LMX in the regression model did not provide any evidence for this
kind of relationship.
Another possible explanation for the non-significant tests regarding a
direct effect of LMX on work engagement and an indirect effect of LMX on
work engagement through work control could be that we did not have
sufficient power. In this respect, a post hoc power analysis with G-Power
(V3.1) in terms of the joint significance approach indicated that our
sample was sufficiently powered for detecting comparatively large in-
direct effects (f
2
¼.35, alpha ¼.05, n ¼43, path a with 3 predictors, path
b with 5 predictors, power (a x b) ¼.9324), but that larger sample size
would have been required to achieve power of .80 for smaller effects (f
2
¼.15, power (a x b) ¼.4855). Results of simulation studies also point in
this direction [64]. Still, the fact remains that we did not find an obvious
direct link between LMX and work engagement or an indirect link be-
tween LMX and work engagement through work control.
7.2. Practical implications
Additionally, this study has important implications for management
practice. The identified positive relationship between team atmosphere
as a social job resource and social workers' work engagement suggests
that the design of resourceful working environments is a promising op-
tion for social organizations to nurture engagement and thereby enhance
social workers' working performance, organizational citizenship
behavior, and finally the quality of services provided to clients. Conse-
quently, we advise practitioners to be attentive to and embrace this
positive relationship and conduct systematic and regular assessments of a
social organization's working environments in terms of their endowment
with job resources in general and team atmosphere in particular. Further,
we recommend implementing targeted work environment redesign ini-
tiatives, including, e.g., redesign workshops [65], in which current work
environments could be rated and changes aimed at improving job re-
sources and engagement could be developed. In addition to activities
where the organization transforms the work environment, we
Table 2. Summary of hypothesized results.
Hypotheses Supported
H1: A high quality of leader-member exchange directly fosters
work engagement of employees in social work organizations.
No
H2: The relation between LMX and social workers' work
engagement is mediated by particularly relevant job resources.
Mediator: team atmosphere Yes
Mediator: work control No
Notes: Yes ¼the hypothesis was supported. No ¼the hypothesis was not
supported.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
6
recommend encouraging social workers to use opportunities to design
their working environments themselves. One possibility in this regard
could be initiating job crafting interventions to increase social workers'
awareness regarding ways in which they can change the level of job re-
sources themselves in order to experience more engagement [66]. Job
crafting workshops could familiarize social workers with the concept of
job crafting and enable them to set a personal crafting plan, and
accompanying measures such as crafting logbooks and reflection meet-
ings could permit them to keep track of and discuss improvements,
problems, and solutions [66].
In addition, we highly recommend establishing, maintaining, and
nurturing high-quality relationships between supervisors and social
workers in social organizations. To do so, offering specifically tailored
workshops and trainings for managers and employees in social organi-
zations would seem to be beneficial. Since they often do not have a
management education background, but come from social work, such
workshops could make managers and employees more knowledgeable
about leader-member exchange theory, provide an understanding of the
value of LMX relationships, and illustrate how to improve them. Carefully
planned trainings could also further strengthen managers' and em-
ployees’active listening capabilities and their competencies in
exchanging mutual expectations, which are critical for building and
maintaining beneficial relationships [67].
7.3. Limitations and future research
Like any empirical investigation, this study is not free of shortcom-
ings. The first limitation results from its cross-sectional design. Though
this research employed leader-member exchange relationships as a pre-
dictor variable and work engagement as a dependent variable, cross-
sectional data in principle allow for reverse causation. Work engage-
ment could very well have an influence on the quality of relationships
between supervisors and subordinates. One could conceive, for instance,
that highly engaged social workers eagerly take on role-making oppor-
tunities and thus work towards establishing high-quality relationships
with their supervisors. Thus, while the directions of causality indicated in
the present study are likely based on the theoretical reasoning offered,
we must exercise caution in inferring causal, unidirectional relations.
Future longitudinal or experimental studies may thus establish an even
more solid foundation for the direction of the link between LMX and
work engagement.
Another limitation is that all participants in the study were employed
at in total twenty advice centers for pregnancy issues that belonged to
one social organization in the state of Bavaria in Germany. Thus, re-
searchers could scrutinize the suggested relationships (a) in the context
of other governing bodies, (b) in areas of social work other than preg-
nancy counseling, and (c) in other countries to generalize the actual re-
sults further.
Third, while we relied on established approaches from past research,
the study variables' measurement could be a potential limitation of our
work. Particularly, this study's reliance on social workers' self-rated and
hence subjective assessments of the quality of leader-member exchange
relationships, job resources, and work engagement could be a limiting
factor. For reasons of research economy, we had to refrain from incor-
porating measures that do not rely on self-reporting (e.g., supervisor
ratings, peer ratings, or objective measures). The only exception to this
was that we had the quality of the leader-member exchange relationships
assessed by both employees and executives. Our analysis of the matched
pairs of supervisors' and social workers' ratings found that they rated
their leader-member exchange consistently. Nevertheless, researchers
might validate our findings incorporating non-self-report measures [68]
in future research efforts.
Fourth, as noted before, the sample size might have been so small that
we did not have sufficient power in our tests to detect effects even if they
were present. Hence, a larger study with more statistical power is
warranted.
Finally, this study could only consider a limited number of potential
confounders. Inadequate consideration of potential confounding vari-
ables may bias research findings and result in erroneous conclusions.
Thus, it would be an advancement if future studies considered other
potential confounders to exclude any alternative explanations for the
observed relationships.
Declarations
Author contribution statement
Bettina Wagner: Conceived and designed the analyses; Collected the
data; Analyzed and interpreted the data; Contributed reagents, materials,
analysis tools or data; Wrote the paper.
Clemens Koob: Conceived and designed the analyses; Analyzed and
interpreted the data; Contributed reagents, materials, analysis tools or
data; Wrote the paper.
Funding statement
This research did not receive any specific grant from funding agencies
in the public, commercial, or not-for-profit sectors.
Data availability statement
Data included in article/supplementary material/referenced in
article.
Declaration of interests statement
The authors declare no conflict of interest.
Additional information
Supplementary content related to this article has been published
online at https://doi.org/10.1016/j.heliyon.2022.e08793.
References
[1] O. Autrata, B. Scheu, Soziale Arbeit: Eine Paradigmatische Bestimmung, VS Verlag
für Sozialwissenschaften, 2008. Epub ahead of print.
[2] J. Hogreve, A. Iseke, K. Derfuss, et al., The service–profit chain: a meta-analytic test
of a comprehensive theoretical framework, J. Market. 81 (2017) 41–61.
[3] R. García-Sierra, J. Fern
andez-Castro, F. Martínez-Zaragoza, Work engagement in
nursing: an integrative review of the literature, J. Nurs. Manag. 24 (2016)
E101–E111.
[4] M.S. Christian, A.S. Garza, J.E. Slaughter, Work engagement: a quantitative review
and test of its relations with task and contextual performance, Person. Psychol. 64
(2011) 89–136.
[5] W.B. Schaufeli, M. Salanova, V. Gonz
alez-rom
a, et al., The measurement of
engagement and burnout: a two sample confirmatory factor Analytic approach,
J. Happiness Stud. 3 (2002) 71–92.
[6] W.B. Schaufeli, A.B. Bakker, M. Salanova, The measurement of work engagement
with a short questionnaire: a cross-national study, Educ. Psychol. Meas. 66 (2006)
701–716.
[7] J.M. Ravalier, The influence of work engagement in social workers in England,
Occup. Med. 68 (2018) 399–404.
[8] W.H. Macey, B. Schneider, The meaning of employee engagement, Ind. Organ.
Psychol. 1 (2008) 3–30.
[9] C. Lloyd, R. King, L. Chenoweth, Social work, stress and burnout: a review, J. Ment.
Health 11 (2002) 255–266.
[10] H. Kim, M. Stoner, Burnout and turnover intention among social workers: effects of
role stress, job autonomy and social support, Adm. Soc. Work 32 (2008) 5–25.
[11] G.M. Acker, Burnout among mental health care providers, J. Soc. Work 12 (2012)
475–490.
[12] E. S
anchez-Moreno, I.-N. de La Fuente Rold
an, L.P. Gallardo-Peralta, et al., Burnout,
informal social support and psychological distress among social workers, Br. J. Soc.
Work 45 (2015) 2368–2386.
[13] S. Winstanley, L. Hales, A preliminary study of burnout in residential social workers
experiencing workplace aggression: might it Be cyclical? Br. J. Soc. Work 45 (2015)
24–33.
[14] A. Tesi, A. Aiello, E. Giannetti, The work-related well-being of social workers:
framing job demands, psychological well-being, and work engagement, J. Soc.
Work 19 (2019) 121–141.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
7
[15] M. Salanova, L. Lorente, M.J. Chambel, et al., Linking transformational leadership
to nurses’extra-role performance: the mediating role of self-efficacy and work
engagement, J. Adv. Nurs. 67 (2011) 2256–2266.
[16] Q. Mehmood, S. Nawab, M.R.W. Hamstra, Does authentic leadership predict
employee work engagement and in-role performance? J. Person. Psychol. 15 (2016)
139–142.
[17] A. Chughtai, M. Byrne, B. Flood, Linking ethical leadership to employee well-being:
the role of trust in supervisor, J. Bus. Ethics 128 (2015) 653–663.
[18] V.G. Rahmadani, W.B. Schaufeli, J. Stouten, How engaging leaders foster
employees’work engagement, Leader. Organ. Dev. J. 41 (2020) 1155–1169.
[19] M.M. Khan, M.S. Mubarik, S.S. Ahmed, et al., My meaning is my engagement:
exploring the mediating role of meaning between servant leadership and work
engagement, Leader. Organ. Dev. J. 42 (2021) 926–941.
[20] R. Niswaty, H. Wirawan, H. Akib, et al., Investigating the effect of authentic
leadership and employees’psychological capital on work engagement: evidence
from Indonesia, Heliyon 7 (2021), e06992.
[21] C.R. Gerstner, D.V. Day, Meta-Analytic review of leader–member exchange theory:
correlates and construct issues, J. Appl. Psychol. 82 (1997) 827–844.
[22] T.A. Scandura, G.B. Graen, M.A. Novak, When managers decide not to decide
autocratically: an investigation of leader–member exchange and decision influence,
J. Appl. Psychol. 71 (1986) 579–584.
[23] G.B. Graen, M. Uhl-Bien, Relationship-based approach to leadership: development
of leader-member exchange (LMX) theory of leadership over 25 years: applying a
multi-level multi-domain perspective, Leader. Q. 6 (1995) 219–247.
[24] R.K. Gottfredson, H. Aguinis, Leadership behaviors and follower performance:
deductive and inductive examination of theoretical rationales and underlying
mechanisms, J. Organ. Behav. 38 (2017) 558–591.
[25] J.R. de Villiers, M.W. Stander, Psychological empowerment, work engagement and
turnover intention: the role of leader relations and role clarity in a financial
institution, J. Psychol. Afr. 21 (2011) 405–412.
[26] K. Breevaart, A.B. Bakker, E. Demerouti, et al., Leader-member exchange, work
engagement, and job performance, J. Manag. Psychol. 30 (2015) 754–770.
[27] X.M. Bezuijen, K. van Dam, P. van den Berg, et al., How leaders stimulate employee
learning: a leader–member exchange approach, J. Occup. Organ. Psychol. 83
(2010) 673–692.
[28] A. Decuypere, W. Schaufeli, Leadership and work engagement: exploring
explanatory mechanisms, German J. Human Res. Manag. 34 (2020) 69–95.
[29] L.M. Shore, L.E. Tetrick, P. Lynch, et al., Social and economic exchange: construct
development and validation, J. Appl. Soc. Psychol. 36 (2006) 837–867.
[30] A.B. Bakker, E. Demerouti, The Job Demands-Resources model: state of the art,
J. Manag. Psychol. 22 (2007) 309–328.
[31] A.B. Bakker, E. Demerouti, A.I. Sanz-Vergel, Burnout and work engagement: the
JD–R approach, Ann. Rev. Organ. Psychol. Organ. Behav. 1 (2014) 389–411.
[32] A.B. Bakker, E. Demerouti, Job demands-resources theory: taking stock and looking
forward, J. Occup. Health Psychol. 22 (2017) 273–285.
[33] E. Demerouti, A.B. Bakker, F. Nachreiner, et al., The job demands-resources model
of burnout, J. Appl. Psychol. 86 (2001) 499–512.
[34] W.B. Schaufeli, A.B. Bakker, Job demands, job resources, and their relationship
with burnout and engagement: a multi-sample study, J. Organ. Behav. 25 (2004)
293–315.
[35] R. Martin, O. Epitropaki, G. Thomas, et al., A review of leader-member exchange
research: future prospects and directions, in: International Review of Industrial and
Organizational Psychology 2010 25, Wiley Blackwell, 2010, pp. 35–88.
[36] A. Aiello, A. Tesi, Psychological well-being and work engagement among Italian
social workers: examining the mediational role of job resources, Soc. Work. Res. 41
(2017) 73–84.
[37] J. Mette, S. Robelski, T. Wirth, et al., “Engaged, burned out, or both?”A structural
equation model testing risk and protective factors for social workers in refugee and
homeless aid, Int. J. Environ. Res. Publ. Health 17 (2020) 583.
[38] A. Tesi, A dual path model of work-related well-being in healthcare and social work
settings: the interweaving between trait emotional intelligence, end-user job
demands, coworkers related job resources, burnout, and work engagement, Front.
Psychol. 12 (2021) 2333.
[39] A. Tesi, Moving forward to social workers’work engagement: framing the
regulatory modes and the job demands-resources model, J. Workplace Behav.
Health 36 (2021) 146–158.
[40] M. Geisler, H. Berthelsen, T. Muhonen, Retaining social workers: the role of quality
of work and psychosocial safety climate for work engagement, job satisfaction, and
organizational commitment, Human Serv. Organ.: Manag. Leader. Govern. 43
(2019) 1–15.
[41] George Washington University Health Workforce Institute and School of Nursing,
Profile of the Social Work Workforce. A Report to Council on Social Work Education
and National Workforce Initiative Steering Committee, Text, Washington, DC,
2017. https://www.socialserviceworkforce.org/resources/profile-social-work-work
force. (Accessed 17 September 2021).
[42] J.J. Hakanen, G. Roodt, Using the job demands-resources model to predict
engagement: analysing a conceptual model, in: Work Engagement: A Handbook of
Essential Theory and Research, Psychology Press, New York, NY, US, 2010,
pp. 85–101.
[43] P. Mayring, Qualitative content analysis: theoretical background and procedures,
in: A. Bikner-Ahsbahs, C. Knipping, N. Presmeg (Eds.), Approaches to Qualitative
Research in Mathematics Education: Examples of Methodology and Methods,
Springer Netherlands, Dordrecht, 2015.
[44] W.B. Schaufeli, Applying the Job Demands-Resources model: a ‘how to’guide to
measuring and tackling work engagement and burnout, Organ. Dynam. 46 (2017)
120–132.
[45] J van Wingerden, A.B. Bakker, D. Derks, The longitudinal impact of a job crafting
intervention, Eur. J. Work. Organ. Psychol. 26 (2017) 107–119.
[46] C. Z
arraga, J. Bonache, The impact of team Atmosphere on knowledge outcomes in
self-managed teams, Organ. Stud. 26 (2005) 661–681.
[47] T.S. Kristensen, H. Hannerz, A. Høgh, et al., The Copenhagen Psychosocial
Questionnaire–a tool for the assessment and improvement of the psychosocial work
environment, Scand. J. Work. Environ. Health 31 (2005) 438–449.
[48] F. Anseel, F. Lievens, E. Schollaert, et al., Response rates in organizational science,
1995–2008: a meta-analytic review and guidelines for survey researchers, J. Bus.
Psychol. 25 (2010) 335–349.
[49] A. Shantz, K. Alfes, C. Truss, et al., The role of employee engagement in the
relationship between job design and task performance, citizenship and deviant
behaviours, Int. J. Hum. Resour. Manag. 24 (2013) 2608–2627.
[50] B. Schyns, T. Paul, Leader-Member Exchange Skala (LMX7). Zusammenstellung
sozialwissenschaftlicher Items und Skalen (ZIS), 2002.
[51] B. Dorer, ESS Round 9 Translation Guidelines, in: European Social Survey (Ed.), ESS
ERIC, London, 2018.
[52] M. Klarmann, S. Feurer, Control variables in marketing research 40, 2018,
pp. 26–40.
[53] P.E. Spector, M.T. Brannick, Methodological urban legends: the misuse of statistical
control variables, Organ. Res. Methods 14 (2011) 287–305.
[54] A.F. Hayes, Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach, second ed., Guilford Press, New York,
2017.
[55] J. Cohen, Statistical Power Analysis for the Behavioral Sciences, L. Erlbaum
Associates, Hillsdale, N.J., 1988.
[56] W. van Breukelen, B. Schyns, P. Le Blanc, Leader-member exchange theory and
research: accomplishments and future challenges, Leadership 2 (2006) 295–316.
[57] J.R.B. Halbesleben, A meta-analysis of work engagement: relationships with
burnout, demands, resources, and consequences, in: A.B. Bakker, M.P. Leiter (Eds.),
Work Engagement: A Handbook of Essential Theory and Research, Psychology
Press, New York, 2010.
[58] P. Sepp€
al€
a, A. M€
akikangas, J.J. Hakanen, et al., Is autonomy always beneficial for
work engagement? A six-year four-wave follow-up study, J. Pers. Orient. Res. 6
(2020) 16–27.
[59] S.D. Spiegelaere, G.V. Gyes, G.V. Hootegem, Not all autonomy is the same. Different
dimensions of job autonomy and their relation to work engagement &innovative
work behavior, Human Fact. Ergon. Manufac. Serv. Indust. 26 (2016) 515–527.
[60] J. Mascare~
no, E. Rietzschel, B. Wisse, Leader-Member Exchange (LMX) and
innovation: a test of competing hypotheses, Creativ. Innovat. Manag. 29 (2020)
495–511.
[61] M.Z. Carter, A. Jones-Farmer, A.A. Armenakis, et al., Transformational leadership
and followers’performance: joint mediating effects of leader-member exchange and
interactional justice, Proceedings 2009 (2009) 1–6.
[62] W. Hochwarter, LMX and job tension: linear and non-linear effects and affectivity,
J. Bus. Psychol. 19 (2005) 505–520.
[63] A.-F. Ionescu, D. Iliescu, LMX, organizational justice and performance: curvilinear
relationships, J. Manag. Psychol. 36 (2021) 197–211.
[64] M.S. Fritz, D.P. MacKinnon, Required sample size to detect the mediated effect,
Psychol. Sci. 18 (2007) 233–239.
[65] D. Holman, C. Axtell, Can job redesign interventions influence a broad range of
employee outcomes by changing multiple job characteristics? A quasi-experimental
study, J. Occup. Health Psychol. 21 (2016) 284–295.
[66] E. Demerouti, Design your own job through job crafting, Eur. Psychol. 19 (2014)
237–247.
[67] J. Mayfield, M. Mayfield, Increasing worker outcomes by improving leader follower
relations, J. Leader. Stud. 5 (1998) 72–81.
[68] F.Z. Xu, Y. Zhang, H. Yang, et al., Sustainable HRM through improving the
measurement of employee work engagement: third-person rating method,
Sustainability 12 (2020) 7100.
B. Wagner, C. Koob Heliyon 8 (2022) e08793
8
