sustainability

Article
The Mediating Role of Job Satisfaction in the Relationship
between FWAs and Turnover Intentions
Nemanja Berber 1 , Dimitrije Gašić 1, * , Ivana Katić 2 and Jelena Borocki 2

1 Faculty of Economics in Subotica, University of Novi Sad, 21000 Novi Sad, Serbia;
[email protected]
2 Faculty of Technical Sciences, University of Novi Sad, 21000 Novi Sad, Serbia; [email protected] (I.K.);
[email protected] (J.B.)
* Correspondence: [email protected]; Tel.: +381-214852911

Abstract: The problem of employee turnover has been investigated in recent years because more
and more countries and organizations are faced with the lack of an adequate labor force. The
new generation of employees (Y and Z generations), contemporary political, social, and economic
challenges, and the COVID-19 pandemic have raised new issues in human resource management
(HRM), especially concerning turnover intentions. In such situations, companies need to create
working conditions that will attract, motivate, and retain employees. One possible response is the
usage of flexible working arrangements (FWAs) as a more flexible way of organizing traditional
jobs and working positions; these arrangements allow employees more possibilities to maintain
work–life balance. The main aim of this paper is to investigate the relationship between FWAs
and turnover intentions of employees, and the mediating effect of job satisfaction. The theoretical
research is based on the literature review. The empirical part of the paper is based on the analysis of a
sample of 219 employees from organizations in Serbia. The authors used descriptive statistics and





 the PLS-SEM method to investigate proposed relations. The results pointed to the positive effects
Citation: Berber, N.; Gašić, D.; Katić, of FWAs and job satisfaction on turnover intentions. In addition, there is an indirect effect of FWAs
I.; Borocki, J. The Mediating Role of on turnover intentions through job satisfaction. FWAs may contribute to increasing job satisfaction
Job Satisfaction in the Relationship and, in turn, job satisfaction contributes to decreasing turnover intentions. Job satisfaction mediates
between FWAs and Turnover this relationship, and employees that are offered FWAs may experience a lower level of turnover
Intentions. Sustainability 2022, 14, intentions when they are satisfied on the job.
4502. https://doi.org/10.3390/
su14084502 Keywords: flexible working arrangements; job satisfaction; turnover intentions; human resource
Academic Editors: Adnan ul Haque management; organizational behavior; Serbia
and Hyo Sun Jung

Received: 8 February 2022

Accepted: 6 April 2022
1. Introduction
Published: 10 April 2022
The problem of employee turnover has been investigated in recent years because
Publisher’s Note: MDPI stays neutral
more and more countries and organizations are faced with the lack of an adequate labor
with regard to jurisdictional claims in
force [1–3]. The most important managerial task related to human resource management
published maps and institutional affil-
(HRM) is the attraction of qualified human resources and the struggle for retention of talent.
iations.
Employee turnover is one of the biggest challenges for all organizations. It may cause
different economic, psychological, and organizational consequences [3] (p. 32), such as the
loss of human capital and the loss of institutional knowledge [2] (p. 1467).
Copyright: © 2022 by the authors.
Numerous challenging factors influence the business environment and labor mar-
Licensee MDPI, Basel, Switzerland. ket; however, the most important are the presence of the new generation of employees
This article is an open access article (Y and Z generations) on the labor market, who have different views and ideas about
distributed under the terms and jobs, careers, and authority; contemporary political, social, and economic challenges; the
conditions of the Creative Commons COVID-19 pandemic; and especially rapid technological development. In such conditions,
Attribution (CC BY) license (https:// many organizations struggle to maintain their competitiveness. In addition, several or-
creativecommons.org/licenses/by/ ganizational factors like managerial style, organizational culture, HR practices (staffing,
4.0/). compensation, training, career development, etc.), job satisfaction, and stress and conflict

Sustainability 2022, 14, 4502. https://doi.org/10.3390/su14084502 https://www.mdpi.com/journal/sustainability

Sustainability 2022, 14, 4502 2 of 14

management influence the organizational ability to be an attractive employer and to retain

employees. Companies need to create working conditions that will attract, motivate, and
retain employees.
According to previous studies, one possible response to many challenges in HRM is
the implementation of flexible working arrangements (FWAs) as a new and more flexible
way of organizing traditional jobs and working positions; this allows employees to have
more possibilities to maintain work–life balance and to choose how and when they perform
business activities [4,5]. FWAs can be defined as a “work option that provides a control
for employees in terms of ‘where’ and/or ‘when’ to perform their job tasks” [4] (p. 3).
This level of control over job tasks means greater perceived autonomy of an employee,
which could lead to a greater possibility for balancing work and private life and to the
preservation of time and energy of an employee [6]. This is one of the most important
benefits of FWAs.
Previous research pointed to the conclusion that FWAs are seen as HRM practices
that comprise greater productivity, organizational commitment, job satisfaction, lower
absenteeism, lower turnover intentions, etc. [5,6]. On the other hand, some of the limitations
of FWAs lie in the fact that not all types of FWAs would have such effects in organizations.
Potential costs for employees could be related to increased stress if the workload is in excess
of the reduced work time or when employees are required to attend to work issues on
their days off [5] (p. 734). This could harm employee performance and diminish FWAs’
positive effects.
The above-mentioned relations are derived from the social exchange theory, which
states that positive and benevolent behavior of one person (sender) to another (receiver) in
an interdependent relationship would create the potential for the receiver to feel obligated
to reciprocate with returned positive behavior [4]. It is important to emphasize that this
relation or exchange in the employment context is not a contract; it is more about the beliefs
and perceptions of both sides, that is, employees and employers. Namely, if employees
perceive the work environment and other HRM practices of employers as positive and
satisfactory, they could feel obligated to show positive work attitudes and behaviors. On the
contrary, if employees perceive HRM practices as not satisfactory, they will show negative
organizational behavior and attitudes. Based on the social exchange theory and previous
research, it is expected that FWAs would have positive effects on turnover intentions, as a
type of employee behavior, in terms of decreasing turnover [4–7]. In addition, FWAs are
seen as a driver of higher job satisfaction [8], one of the most important employee attitudes
related to the willingness to stay or to leave the organization [5].
Based on the above, the main aim of this paper is to investigate the relationship
between FWAs and turnover intentions of employees, and the mediating effect of job
satisfaction as an important factor related to both constructs. The methodology used in
the paper consists of theoretical and empirical research. The theoretical research is based
on a literature review, using the desk-research method. The empirical part of the paper
is based on the analysis of the sample of 219 employees from business organizations in
Serbia, who gave responses to questions on FWAs, turnover intentions, and job satisfaction.
The novelty of this paper lies in the mediation analysis, since in the previous research, all
three constructs were usually examined in isolation, without considering the mediating
effects. Several prior studies investigated similar relations, but without mediation analysis.
The paper consists of five sections. The Section 2 is dedicated to the theoretical devel-
opment of hypotheses and the main argument related to flexible working arrangements,
job satisfaction, and turnover intentions. The Section 3 presents the research methodology.
Results of the study are presented in the Section 4, while discussion and conclusions are
given in the Sections 5 and 6 of the paper, respectively.

2. Theoretical Background
Flexible working arrangements have gained great attention recently due to the great
negative impact of the COVID-19 pandemic all over the world. FWAs are seen as a solution
Sustainability 2022, 14, 4502 3 of 14

for many business issues, but are mainly related to work–life balance. FWAs could increase
work control, and are defined as constructs that possess two major forms of flexibility,
related to time (flex-time) and related to location (flexplace) [9]. FWAs are generally seen
as a “negotiated term of employment related to timing and/or place of work” [10]. FWAs
are also explained as “employer-provided benefits that permit employees some level of
control over when and where they work outside of the standard workday” [11,12] and are
usually related to home-based working, flex-time, reducing or extending contract hours,
or allowing overtime hours to support employees’ work–life balance and improve firm
performance [13] (p. 727).
Flexibility in work has a great effect on employees and employers. Employees are
more ready to “join an organization, be satisfied with the jobs they do and continue working
with the same employers. Employers have become aware of some outcomes, such as being
interested, motivated, and retaining their talented employees, having satisfied and numer-
ous engaged employees, along with improving employee effectiveness and success” [14].
Previous research investigated the effects of FWAs on employees’ behavior and attitudes,
such as turnover [12,15], engagement [16], job satisfaction [17], absenteeism [18], etc. In
most of these areas, FWAs showed a positive relation to the dependent variables, in terms
of increasing job satisfaction, engagement, or performance, while decreasing absenteeism
and turnover.
For this research, bearing in mind that more and more companies are struggling to
retain their best employees, the authors decided to investigate relations between FWAs,
job satisfaction, and turnover intentions of employees. The mentioned constructs are
important outcomes of organizational behavior strategies; organizational human resource
management practices influence organizations such that the outcomes can be measured by
employee performance, job satisfaction, absenteeism, and turnover [19].
Regarding turnover, the best way to investigate the turnover of employees is to
follow their turnover intentions, as a reliable predictor of actual turnover behavior [2].
Sandhya and Sulphey stated that turnover intentions can be defined as “a mental decision
prevailing between an individual’s approach with reference to work to continue or leave
the work”; it is “a measure for understanding turnover before employees actually quit
or leave organizations” [1] (p. 327). FWAs, in terms of gaining more control over the
job and the usage of different work types, like teleworking, flex-time, home-based work,
etc., should decrease the turnover intentions of employees. This idea is derived from
social exchange theory, explained in the previous section of the paper, in which positive
employer practices towards their employees should cause positive work behavior (i.e.,
decision to stay, reduction of turnover intentions, and turnover itself) and work attitudes
(commitment, engagement, job satisfaction, etc.). In most of the previous research, this
hypothesis has been proven [20], but the question is what is happening now, during the
COVID-19 pandemic, when more companies are using FWAs [21] and when different
external factors are moving organizations towards greater usage of FWAs.
FWAs like flex-time (which allows employees to vary the times when they start and
finish work), sabbaticals (paid leave from work), and home-based work (work from a
location outside workplace, in a household) increased job satisfaction, while sabbaticals
and home-based work decreased turnover intention of employees in a sample of German
employees [20]. In addition, Tsen et al. [4] investigated 16,920 respondents from 35 coun-
tries and found that perceived job independence significantly moderates the relationship
between FWAs and turnover intention. They found that employees who perceive their jobs
as highly independent have a lower turnover intention when they are using FWAs such
as flex-time, flex-leave, or home-based work, while more interdependent employees who
use home-based work and flex-time may have a greater intention to leave [4] (p. 1). In a
similar study, the authors found that flexible work hours had no significant effect on job
satisfaction, but they reduced staff turnover. In contrast, job satisfaction improved with job
sharing and flexible leave [5].
Sustainability 2022, 14, 4502 4 of 14

Masuda et al. [22] investigated the relationship between FWAs and job satisfaction,
turnover intentions, and work–family conflict in Latin American, Anglo, and Asian country
clusters in a sample of 3918 managers from 15 countries. The results showed that there is a
negative relationship between flex-time work and turnover intentions in the Anglo cluster
and that the relationship is not significant for the Latin American cluster. In addition, a
positive relationship was found between flex-time availability and job satisfaction for the
Anglo cluster, but not in the Latin American cluster.
De Sivatte and Guadamillas [23] also investigated similar relations in a sample of
480 employees in Spain and noted that FWAs have a negative relation to turnover intentions
(these practices reduce the intention of employees to leave the organization) and a positive
relation to employees’ commitment. Mullins et al. [24] found that FWAs are associated
with lower dissatisfaction of employees, but with small effects on employee intention to
leave the organization.
According to these research results, FWAs mostly have a positive effect on turnover
intentions in terms of decreasing it, and therefore these programs could be used as an
employee retention strategy. Since FWAs are understood as employee-friendly practices,
properly implementation could cause a positive perception of employees, and in exchange,
they could demonstrate more positive working behaviors by reducing their intention to
leave the organization.
Based on the aforementioned, the first hypothesis of the research is as follows:

Hypothesis 1 (H1): FWAs have negative relations to turnover intention.

In addition to behaviors, employees could respond to positive HRM practices with

their positive work attitudes, such as job satisfaction, job engagement, job commitment, etc.
However, in most of the mentioned research, one of the commonly explored variables was
job satisfaction, as employee attitude on the job, which consists of evaluative, cognitive, and
affective components [25], shows how much an employee is satisfied with the job. It has
been proven that employees with a high level of job satisfaction will have lower turnover
intentions [5,26]; therefore, job satisfaction is considered as a cause of employees’ turnover
intentions. In addition, FWAs are found to be positively related to job satisfaction [20,24],
and on that basis, job satisfaction can have a mediating on the relationship between FWAs
and turnover intentions [16]. According to the results of the research of Azar et al., job
satisfaction and work–life conflict mediate the impact of FWA use on turnover intentions.
The authors of this paper propose that job satisfaction will mediate this relationship and
that employees who are offered FWAs will experience a lower level of turnover intention
when they are satisfied on the job.

Hypothesis 2 (H2): Job satisfaction mediates the relationship between FWAs and turnover intentions.

3. Materials and Methods

3.1. The Questionnaire
A specially designed questionnaire was used in the process of researching the me-
diating role of job satisfaction in the relationship between FWA and turnover intentions.
Respondents had the opportunity to answer the questions asked via the electronic ques-
tionnaire created in “Google Forms”. The authors used a 5-point Likert scale to measure
attitudes (1 = strongly disagree, 5 = strongly agree) [27]. The first part of the questionnaire
consisted of control questions on gender, age, education, and position in the company. The
second part of the questionnaire included questions related to the independent variable
“Flexible work arrangements (FWA)”, where a standardized questionnaire was used with
eleven questions. All questions related to FWAs (from FWA1 to FWA10, including “FWAs
help me to balance life obligations”, “I can’t afford the loss of earnings associated with
most flexible work options (R)”, “FWAs do not suit me because they make me feel discon-
nected from the workplace (R)”, “Shorter working hours would negatively affect my career
Sustainability 2022, 14, 4502 5 of 14

progress in the company where I work (R)”, “Working with more flexible working hours
is essential for me to fulfill family obligations”) were derived from Albion [28]. The third
part of the questionnaire related to dependent variables was based on two standardized
questionnaires, with the first related to “Job satisfaction” (from JS1 to JS5, including “I like
the kind of work I do”, “I like my job better than the average worker does”, “I find real
enjoyment in my work”, “I feel a great sense of personal satisfaction when I do my job
well”, “Considering everything, I am satisfied with my job”), consisting of five questions,
derived from Morgeson and Humphrey [29]. The second part was related to “Turnover
intentions” and consisted of four questions (from TI1 to TI4, including “I often think about
leaving my current job”, “I probably won’t have a bright future if I stay in this company”,
“Maybe next year I will leave my current company and start working for someone else”, “I
plan to stay in this company for a longer time to develop my career (R)”), based on Chen
and Francesco [30].

3.2. The Sample

The questionnaire was primarily intended for employees from various business activi-
ties who were offered the opportunity to reorganize their way of doing business through the
application of flexible work arrangements, such as weekend work, overtime, flexible work-
ing hours, home-based work, teleworking, etc. These include sectors such as education,
finance, insurance, scientific and technical industries, trades, etc. A total of 219 employees
completed a questionnaire relating to the mediating role of job satisfaction in the relation-
ship between FWA and turnover intentions. Data collection was performed in the period
from the end of December 2020 until the mid-June 2021. The number of questionnaires
sent was 1000, but only those who stated that they use at least some form of FWA and gave
a response to their perception on these FWAs were taken into consideration for further
analyses. The number of valid questionnaires was 219, and the rate of response was 21.9%.
The authors used Google Forms to create the questionnaire and to collect the data. This
kind of data collection was appropriate since face-to-face contact in many organizations
was not possible due to social distancing. The identity of the respondents was secure, as all
questionnaires were sent with the invitation to participate without including any personal
details. The authors used the LinkedIn network as a source of respondents in Serbia.
Table 1 presents the structure of the sample regarding gender, age, education, and po-
sition of the employee in the company. The sample consisted mostly of female respondents
(almost 58%), younger employees (up to 35 years, about 57%), with bachelor’s or master’s
degree (about 70%), and in professional positions in a company (53%).

Table 1. Sample characteristics.

Sample Characteristics N Percentage of Sample

Gender 219 100
1. Male 92 42.01
2. Female 127 57.99
Age 219 100
1. Less than 25 42 19.18
2. 25–34 81 36.99
3. 35–44 45 20.55
4. 45–55 34 15.53
5. More than 55 17 7.76
Education 219 100
1. High School 19 8.68
2. Three-year vocational studies 6 2.74
Sustainability 2022, 14, 4502 6 of 14

Table 1. Cont.

Sample Characteristics N Percentage of Sample

3. Bachelor’s degree 77 35.16
4. Master’s study 79 36.07
5. Master’s degree (Serbian version Magistar) 13 5.94
6. Ph.D. study 25 11.42
Position in the company 219 100
1. Manager 60 27.4
2. Professional worker 117 53.42
3. Administrative worker 33 15.07
4. Manual worker 9 4.11
Source: Authors.

4. Results
The authors used the program “Smart PLS” to present the results of the research
on the mediating role of job satisfaction in the relationship between FWA and turnover
intentions. To investigate the proposed relations, PLS-SEM analysis was performed. First,
the authors calculated the measurement and structural model parameters and generated
the accompanying bootstrap estimates. The study was performed to assess the total and
direct effects of the FWA construct on the dependent variable (turnover intentions) and the
indirect effects via the mediator (job satisfaction). The analysis of the obtained data based
on the completed questionnaires was divided into two parts; the authors first investigated
the measurement model, and then the research hypotheses were tested. Table 2 shows the
descriptive statistics for each indicator in the study.

Table 2. Descriptive statistics for each indicator in the study.

N Min Max Mean Std. Deviation

FWA 219 1.91 4.82 3.64 0.69
Job satisfaction 219 1.8 5 4.50 0.67
Turnover intentions 219 1 5 1.98 1.00
Source: Authors.

For the first step, the authors loaded reflective indicator loadings, internal consistency
reliability, convergent validity, and discriminant validity. This type of measurement is
proposed for reflective constructions in the model [31]. In addition to the mentioned
methods, the CMB—common method bias—test was also performed. The authors Berber,
Slavić, and Aleksić [32] state in their paper that the lowest eligibility limit for factor load is
0.708. Load factors between 0.4 and 0.7 should be retained only if their removal does not
have an impact on AVE and Composite Reliability [15,33]. Certain items had to be removed
from further analysis because their loads had very low values. Based on the above, Figure 1
presents the retained items with loadings above 0.708 (see Figure 1).
The following table shows the indicator reliability and construct reliability and validity.
Table 3 shows the reliability test, performed by calculating Cronbach’s Alpha, Compos-
ite Reliability, and Average Variance Extracted. Based on the data obtained by calculating
Cronbach’s Alpha, we noticed that the values ranged from 0.768 (FWA), to 0.851 (turnover
intentions), to the highest value, recorded for job satisfaction, 0.904. Some of the authors
recommend the lowest acceptable limit of Cronbach’s Alpha be 0.6 [34–36].
Sustainability 2022,
Sustainability 2022, 14,
14, 4502
x FOR PEER REVIEW 77 of 14
of 14

Figure 1.
Figure 1. Path
Path coefficient
coefficient estimates.
estimates.

TableThe following
3. Indicator table and
reliability shows the indicator
construct reliabilityreliability and construct reliability and va-
and validity.
lidity.
Cronbach’s
Table 3 shows the reliability test, Alpha
performed by rho_A CR
calculating Cronbach’s AVE Com-
Alpha,
FWA
posite Reliability, 0.768
and Average Variance Extracted.0.828
Based on the0.833
data obtained0.500
by calcu-
latingJob
Cronbach’s
satisfactionAlpha, we noticed
0.904 that the values ranged
0.922 from 0.768
0.929 (FWA), to 0.851
0.726
(turnover intentions), to the highest
Turnover intentions
value, recorded0.855
0.851
for job satisfaction,
0.900
0.904. Some
0.694
of the
authors recommend
Source: Authors.
the lowest acceptable limit of Cronbach’s Alpha be 0.6 [34–36].

TableThe value of
3. Indicator composite
reliability reliability
and construct of constructs
reliability ranges from 0.833 (FWA), to 0.9
and validity.
(turnover intentions), to the highest value, recorded for job satisfaction, 0.929. Some
Cronbach’s Alpha rho_A CR AVE
of the authors recommend that the lowest limit of acceptability of CR is 0.7 [37,38]. Based
FWA 0.768 0.828
on the results presented in the table above, we conclude that the CR criterion was met. 0.833 0.500
Job satisfaction
Composite reliability can be used as0.904 an alternative, because0.922 CR values 0.929
are slightly0.726
higher
Turnover intentions 0.851 0.855
than Cronbach Alpha values, but this difference is relatively insignificant [39]. 0.900 0.694
Source: Authors. validity was assessed by testing Average Variance Extracted (AVE). The
Convergent
table above shows the AVE values, and they range from 0.5 (FWA), to 0.694 (turnover
The value
intentions), of composite
to the highest value, reliability
recordedof constructs ranges from
for job satisfaction, 0.833is(FWA),
which 0.726. to
The0.9lowest
(turn-
over intentions),
acceptable limit oftoAVEthe highest value,Based
is 0.5 [40,41]. recorded
on thefordata
job satisfaction,
shown in the0.929.
table Some
above,ofthethelimit
au-
thors
of recommend
acceptability wasthat
met,theandlowest limitwe
therefore of conclude
acceptability
that of CR is 0.7 validity
Convergent [37,38]. Based on the
was satisfied
results
in presented
all three in the table above, we conclude that the CR criterion was met. Compo-
constructs.
site reliability
Discriminant can validity
be usedcan as an alternative,
be assessed by because CR values are
using Cross-loadings slightly higher
indicators, than
the Fornell
Cronbach
and Larcker Alpha values,
criterion, andbut this difference is relatively
heterotrait–monotrait insignificant
correlation [39].
ratios [42]. Table 4 shows the
discriminant
Convergent validity crosswas
validity loadings.
assessed by testing Average Variance Extracted (AVE). The
tableIfabove
the loadshowsof the
the indicator
AVE values, for its
andconstructive
they range structure is greater
from 0.5 (FWA), than any
to 0.694 other
(turnover
construction,
intentions), tothe the measurement
highest value, model recorded hasfor
a job
corresponding
satisfaction, discriminant
which is 0.726. validity
The lowest[43].
Based on the data obtained in the table above, the results show that a
acceptable limit of AVE is 0.5 [40,41]. Based on the data shown in the table above, the limitload of each block
is
ofgreater than awas
acceptability load of any
met, and other block
therefore weinconclude
the samethat column and row,
Convergent clearly
validity separating
was satisfied
each
in alllatent variable. The cross-loading output confirms the discriminant validity of the
three constructs.
measurement
Discriminant model.validity can be assessed by using Cross-loadings indicators, the Fornell
Table 5 displays
and Larcker criterion,discriminant validity via thecorrelation
and heterotrait–monotrait Fornell–Lacker
ratioscriterion.
[42]. Table 4 shows the
Based on Fornell–Lacker
discriminant validity cross loadings. criteria, the root of the AVE latent variable must have a value
that is greater than the value of all correlations with the latent variable [40]. Based on the
obtained data, we conclude
Table 4. Discriminant that discriminant
validity—cross loadings. validity is satisfied because the value of the
AVE root on the diagonal is greater than all the values shown below for each variable.
Table 6 shows discriminant FWA validityJob of Satisfaction
the heterotrait–monotrait Turnover Intentions
(HTMT).
FWA10R 0.737 0.223 −0.283
Sustainability 2022, 14, 4502 8 of 14

Table 4. Discriminant validity—cross loadings.

FWA Job Satisfaction Turnover Intentions

FWA10R 0.737 0.223 −0.283
FWA5 0.645 0.204 −0.177
FWA6 0.625 0.202 −0.109
FWA8R 0.776 0.402 −0.416
FWA9R 0.741 0.171 −0.259
JS1 0.284 0.889 −0.559
JS2 0.315 0.851 −0.425
JS3 0.351 0.903 −0.509
JS4 0.243 0.713 −0.396
JS5 0.372 0.890 −0.617
TI1 −0.305 −0.561 0.902
TI2 −0.274 −0.497 0.865
TI3R −0.369 −0.490 0.750
TI4 −0.363 −0.433 0.805
Source: Authors.

Table 5. Discriminant validity: Fornell–Lacker criterion.

FWA Job Satisfaction Turnover Intentions

FWA 0.707
Job satisfaction 0.371 0.852
Turnover intentions −0.394 −0.598 0.833
Source: Authors.

Table 6. Discriminant validity: heterotrait–monotrait (HTMT).

FWA Job Satisfaction Turnover Intentions

FWA
Job satisfaction 0.397
Turnover intentions 0.428 0.670
Source: Authors.

All values of HTMT below 0.9 show that the components differ from each other to a
satisfactory extent, which means that they describe different phenomena [44]. Based on
the obtained data shown in the table above, we conclude that the criterion of discriminant
validity according to HTMT is met because all the obtained values are below 0.9.
The common method bias (CMB) was investigated using the full collinearity ap-
proach [45]. The threshold value of VIF factors is 3 [44], while some authors accept VIF
values less than 5 or even 10, indicating the harmfulness of collinearity [46–48]. Based on
the data shown in Table 7, the multicollinearity analysis shows that VIF values are in most
cases below 3 but also that there are values such as (JS1, JS3, TI1, and TI2) that record values
slightly above 3; however, these are accepted based on the indicators of authors who accept
VIF values up to 5.
Sustainability 2022, 14, 4502 9 of 14

Table 7. Collinearity statistics.

VIF
FWA10R 1.854
FWA5 2.551
FWA6 2.515
FWA8R 1.258
FWA9R 1.900
JS1 3.162
JS2 2.673
JS3 3.739
JS4 1.593
JS5 2.824
TI1 3.528
TI2 3.068
TI3R 1.451
TI4 1.804
Source: Authors.

The final step is to analyze the relationship between the independent variable (FWA)
and the dependent variables of job satisfaction and turnover intentions, as well as the
mediating role of job satisfaction in the relationship between FWA and turnover intentions.
R2 (R-squared), as a statistical measure of the proportion of the variance for a dependent
variable that is explained by an independent variable, shows that for “Job satisfaction” it is
13.8%, while in the case of “Turnover intentions” it is 39.2%, explained by the independent
variable “FWA” in the model.
Table 8 presents the means, standard deviations, T-statistics, and p-values of the
variables in the model.

Table 8. Mean, Standard deviation, T-statistics, and p-values.

Original Standard
SampleMean T Statistics p Values Results
Sample Deviation
FWA -> Job satisfaction 0.371 0.384 0.059 6.297 0.000 Accepted
FWA -> Turnover intentions −0.198 −0.206 0.048 4.119 0.000 Accepted
Job satisfaction ->
−0.525 −0.523 0.062 8.415 0.000 Accepted
Turnover intentions
FWA -> Job satisfaction ->
−0.195 −0.198 0.033 5.932 0.000 Accepted
Turnover intentions
Source: Authors.

Based on the Table 8, we conclude that there is a positive and statistically significant
relationship between FWA and job satisfaction (β = 0.371; T = 6.297; p = 0.000), a negative
and statistically significant relationship between FWA and turnover intentions (β = −0.198;
T = 4.119; p = 0.000) and a negative statistically significant relation between job satisfaction
and turnover intentions (β = −0.525; T = 8.415; p = 0.000). Regarding the mediating role
of job satisfaction in explaining the relationship between FWA and turnover intentions, a
negative mediation relationship was found, since the indirect effect of FWA on turnover
intentions through job satisfaction was significant (β = −0.195; T = 5.932; p = 0.000). These
relations are presented in Figure 2.
 and turnover intentions (β = −0.525; T = 8.415; p = 0.000). Regarding the mediating role of
job satisfaction in explaining the relationship between FWA and turnover intentions, a
negative mediation relationship was found, since the indirect effect of FWA on turnover
Sustainability 2022, 14, 4502 10 of 14
intentions through job satisfaction was significant (β = −0.195; T = 5.932; p = 0.000). These
relations are presented in Figure 2.

Figure
Figure2.
2.The
Thepath
pathmodel
modelwith
withbootstrapping
bootstrappingresults.
results.

5.Discussion
5. Discussion
Theresults
The resultsofof this
this study
study pointed
pointedto tothe
thepositive
positiveeffects
effectsofofFWAs
FWAs and and job
job satisfaction
satisfaction
on turnover intentions. Job satisfaction mediates the relationship
on turnover intentions. Job satisfaction mediates the relationship between FWAs between FWAs and and
turnover intentions. There is an indirect effect of FWAs on turnover
turnover intentions. There is an indirect effect of FWAs on turnover intentions through intentions through job
satisfaction. The implementation of FWAs increases job satisfaction
job satisfaction. The implementation of FWAs increases job satisfaction and decreases and decreases turnover
intentions,
turnover hence providing
intentions, possibilities
hence providing for companies
possibilities to retain
for companies to their
retainbest
theiremployees
best em-
and create a modern working environment that will
ployees and create a modern working environment that will enable employees enable employees to have more
to have
possibilities
more and more
possibilities and more safety in the
safety in contemporary
the contemporary business
businessenvironment,
environment, in the
in thecontext
con-
of the COVID − 19 pandemic. The present study results showed
text of the COVID−19 pandemic. The present study results showed the expected relations, the expected relations,
as FWAs
as FWAshave havepositive
positiverelations
relationstoto jobjob satisfaction
satisfaction andand negative
negative relations
relations to turnover
to turnover in-
tentions [5,16]. Greater usage of flexible working arrangements leads to higherhigher
intentions [5,16]. Greater usage of flexible working arrangements leads to job
job satis-
satisfaction and lower turnover intentions, and based on these results, both proposed
faction and lower turnover intentions, and based on these results, both proposed hypoth-
hypotheses are confirmed. A positive and statistically significant relationship between
eses are confirmed. A positive and statistically significant relationship between FWA and
FWA and job satisfaction (β = 0.371; T = 6.297), a negative and statistically significant
job satisfaction (β = 0.371; T = 6.297), a negative and statistically significant relationship
relationship between FWA and turnover intentions (β = −0.198; T = 4.119) and a negative
between FWA and turnover intentions (β = −0.198; T = 4.119) and a negative statistically
statistically significant relation between job satisfaction and turnover intentions (β = −0.525;
significant relation between job satisfaction and turnover intentions (β = −0.525; T = 8.415;
T = 8.415; p = 0.000) were found. Regarding the effect of job satisfaction on the relationship
p = 0.000) were found. Regarding the effect of job satisfaction on the relationship between
between FWA and turnover intentions, a negative mediation relationship was found, where
FWA and turnover intentions, a negative mediation relationship was found, where FWA
FWA had a negative impact on turnover intentions through job satisfaction (β = −0.195;
had a negative impact on turnover intentions through job satisfaction (β = −0.195; T =
T = 5.932), which means that employees who are offered FWAs and who are more satisfied
5.932), which means that employees who are offered FWAs and who are more satisfied
with their job express lower turnover intentions. In addition, the results of the tests of the
with their job express lower turnover intentions. In addition, the results of the tests of the
questionnaire and the data showed a high level of validity and reliability; therefore, these
questionnaire
instruments could and thebe data
usedshowed a high level ofofvalidity
for the investigation and reliability;
the mentioned therefore,
constructs these
in a specific
instruments could be used for the investigation of the mentioned constructs in a specific
business environment.
business environment.
These results are in the line with previously reported results, where FWAs also had
These impact
a positive results are
on injobthe line with previously
satisfaction [8,12,17,24]reported results, impact
and a negative where FWAs also had
on turnover in-
atentions
positive[4–7].
impact on job satisfaction [8,12,17,24] and a negative impact
Employees that perceived FWAs as offering the potential for maintaining on turnover inten-
tions [4–7].
balance Employees
between workthat andperceived
private life FWAs as offering
expressed the potential
a higher level of job forsatisfaction
maintainingand bal-a
ance
lowerbetween
level of work and to
intention private
leave life expressed
(turnover a higher
intention). levelwere
They of job satisfaction
more satisfied,and
andaon lower
that
level of intention to leave (turnover intention). They were more satisfied,
basis, they could show more positive work behaviors, such as reduced turnover intentions. and on that basis,
The results of the mediation analysis, which was the main goal of the research, are in
line with other researchers, especially Azar et al. [16], who also confirmed mediation. Job
satisfaction is the mediator in the relationship between FWAs and turnover intentions. Ad-
ditionally, the results are in the line with the social exchange theory [5]. With the respect to
FWAs, the results of this research found employees who understand and perceive FWAs as
positive and helpful HR practices offered by their employers, felt more satisfied on the job,
Sustainability 2022, 14, 4502 11 of 14

because they had more autonomy and increased decision-making possibilities regarding
their work-related issues, and they reciprocated positively toward their employers. This
positive reciprocation response is usually understood as a will to stay in an organization
and a reduced intention to leave.
Therefore, FWAs should be carefully planned, monitored, controlled, and enhanced by
managers to provide better working conditions and try to retain talent in an organization.
An important role in this process should be dedicated to leaders, because they are seen as a
key factor in building a positive psychological climate in an organization [49], especially
in times of crisis, such as the pandemic. The results can be interpreted in line with the
pandemic. Since the main issue related to the pandemic is the fear of becoming infected
with the COVID-19 virus, the decision to implement FWAs, especially home-based work,
teleworking, or flex-time, can be very stimulative for employees in terms of their satisfaction
related to the preservation of their health and security. Additionally, when employees
see that the company and their managers are struggling to help them to stay healthy and
offer such beneficial working programs, their intention to leave can be decreased. On the
other hand, lower turnover intention can occur because of the fear of employees that they
would lose their job due to the negative effects of the pandemic on the business. This
reason is not related to the implementation of FWAs, and therefore this area also needs to
be investigated.

6. Conclusions
This paper has practical implications that lie in the potential of implementation of
FWAs in business to decrease the turnover intentions of employees. This is important,
since almost all companies today noted that lack of talent, higher levels of occupational
stress [50], and high levels of employee turnover are the most important HRM issues.
Allowing employees greater control over their jobs in terms of how, when, and where the
job would be done makes them more satisfied, with reduced stress, and less willingness
to leave their organizations. Of course, FWAs are not enough to reach this goal. Some
other factors, such as strong organizational culture and climate, adequate leadership style,
other HRM practices, are important for both job satisfaction and turnover of employees.
Therefore, companies should analyze, plan, and prepare the FWAs that can be offered to
employees as a part of a broader strategy for the retention of employees. This is especially
important in times of new challenges, such as the COVID-19 pandemic, which has increased
work in the virtual environment [51], made digital business and digital strategies even
more important [52], and greatly increased teleworking. This research is one of the first to
deal with the issues of FWAs, job satisfaction, and turnover intentions in Serbia, using this
specific methodology and data analysis. Therefore, the results of this paper can serve as a
starting point for the creation of strategies and actions of flexible working programs, as
well as their adequate implementation, to lead to HR outcomes like job satisfaction and the
decision to stay at a company.
The important theoretical implications lie in the increased understanding of the ef-
fects of job satisfaction on the relationship between the perception of employees regard-
ing offered flexible working arrangements and their turnover intentions. According to
Azar et al. [16], all three constructs are usually investigated separately, without mediation
analysis. This study tested the mediation effects, and the results pointed to the conclusion
that job satisfaction is a mediator in the proposed relationship, which is in the line with
previous research [16] and with the theory of social exchange. The study adds to the accu-
mulating body of knowledge on the impact of FWAs on both job satisfaction and turnover
intentions, but also on specific relations between these three constructs.
The present study has potential limitations. One of the most important is the size of
the sample. Although 219 employees can be seen as a small sample, the usage of the Smart
PLS program proved that the number of respondents in the sample was adequate to obtain
results and for testing the validity of the data. The 219 responses were acceptable for the
PLS-SEM procedure, because this number met the requirement related to the ten-times rule:
Sustainability 2022, 14, 4502 12 of 14

“ten times the largest number of inner model paths directed at a particular construct in the
inner model” [37].
In addition, the authors of this research did not use controls for the proposed relations,
like gender, age, or sector of business; this might provide interesting results. Therefore,
the inclusion of age, marital status, family status, gender, and other control variables as
moderators in the analysis can be valuable and is recommended for future research. This
will further elucidate the practical implications for managers implementing FWAs on an
organizational level.

Author Contributions: Conceptualization, N.B.; original draft preparation, review, and editing, I.K.
and J.B.; data analysis and methodology, D.G. All authors have read and agreed to the published
version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The data are available upon request from the author.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Sandhya, S.M.; Sulphey, M.M. Influence of empowerment, psychological contract and employee engagement on voluntary
turnover intentions. Int. J. Prod. Perform. Manag. 2020, 70, 325–349. [CrossRef]
2. Verma, B.K.; Kesari, B. Does the Morale Impact on Employee Turnover Intention? An Empirical Investigation in the Indian Steel
Industry. Glob. Bus. Rev. 2020, 21, 1466–1488. [CrossRef]
3. Poór, J.; Slavić, A.; Nikolić, M.; Berber, N. The managerial implications of the labor market and workplace shortage in Central
Eastern Europe. Strat. Manag. 2021, 26, 31–41. [CrossRef]
4. Tsen, M.K.; Gu, M.; Tan, C.M.; Goh, S.K. Effect of Flexible Work Arrangements on Turnover Intention: Does Job Independence
Matter? Int. J. Sociol. 2021, 51, 451–472. [CrossRef]
5. Kotey, B.A.; Sharma, B. Pathways from flexible work arrangements to financial performance. Pers. Rev. 2019, 48, 731–747.
[CrossRef]
6. Onken-Menke, G.; Nüesch, S.; Kröll, C. Are you attracted? Do you remain? Meta-analytic evidence on flexible work practices.
Bus. Res. 2017, 11, 239–277. [CrossRef]
7. Berber, N.; Morley, M.J.; Slavić, A.; Poór, J. Management compensation systems in Central and Eastern Europe: A comparative
analysis. Int. J. Hum. Resour. Manag. 2017, 28, 1661–1689. [CrossRef]
8. Wheatley, D. Employee satisfaction and use of flexible working arrangements. Work. Employ. Soc. 2017, 31, 567–585. [CrossRef]
9. Allen, T.D.; Johnson, R.C.; Kiburz, K.M.; Shockley, K.M. Work-Family Conflict and Flexible Work Arrangements: Deconstructing
Flexibility. Pers. Psychol. 2013, 66, 345–376. [CrossRef]
10. Allen, T.D.; Shockley, K. Flexible work arrangements: Help or hype. In Handbook of Families and Work: Interdisciplinary Perspectives;
Crane, D.R., Hill, E.J., Eds.; University Press of America: Plymouth, MA, USA, 2009; pp. 265–284.
11. Lambert, A.D.; Marler, J.H.; Gueutal, H.G. Individual differences: Factors affecting employee utilization of flexible work
arrangements. J. Vocat. Behav. 2008, 73, 107–117. [CrossRef]
12. McNall, L.A.; Masuda, A.D.; Nicklin, J.M. Flexible Work Arrangements, Job Satisfaction, and Turnover Intentions: The Mediating
Role of Work-to-Family Enrichment. J. Psychol. 2009, 144, 61–81. [CrossRef] [PubMed]
13. Groen, B.A.; van Triest, S.; Coers, M.; Wtenweerde, N. Managing flexible work arrangements: Teleworking and output controls.
Eur. Manag. J. 2018, 36, 727–735. [CrossRef]
14. Karsili, H.; Yesiltas, M.; Berberoglu, A. Workplace Flexibility for Sustainable Career Satisfaction: Case of Handling in the Aviation
Sector in North Cyprus. Sustainability 2021, 13, 6878. [CrossRef]
15. Gašić, D.; Berber, N. The Influence of Flexible Work Arrangement on Employee Behavior During the COVID-19 Pandemic in the
Republic of Serbia. Manag. Sustain. Bus. Manag. Solut. Emerg. Econ. 2021, 26, 73–88. [CrossRef]
16. Azar, S.; Khan, A.; Van Eerde, W. Modelling linkages between flexible work arrangements’ use and organizational outcomes.
J. Bus. Res. 2018, 91, 134–143. [CrossRef]
17. Neirotti, P.; Raguseo, E.; Gastaldi, L. Designing flexible work practices for job satisfaction: The relation between job characteristics
and work disaggregation in different types of work arrangements. New Technol. Work Employ. 2019, 34, 116–138. [CrossRef]
18. Peretz, H.; Fried, Y.; Levi, A. Flexible work arrangements, national culture, organisational characteristics, and organisational
outcomes: A study across 21 countries. Hum. Resour. Manag. J. 2018, 28, 182–200. [CrossRef]
19. Štangl Šušnjar, G.; Slavić, A.; Berber, N. Organizational Behavior (Organizaciono Ponašanje); Ekonomski Fakultet u Subotici: Subotica,
Serbia, 2021.
Sustainability 2022, 14, 4502 13 of 14

20. Kröll, C.; Nüesch, S. The effects of flexible work practices on employee attitudes: Evidence from a large-scale panel study in
Germany. Int. J. Hum. Resour. Manag. 2019, 30, 1505–1525. [CrossRef]
21. Vasić, M. Challenges of teleworking during the COVID-19 pandemic. Anal. Èkon. Fak. U Subotici 2020, 56, 63–79. [CrossRef]
22. Masuda, A.D.; Poelmans, S.A.; Allen, T.D.; Spector, P.E.; Lapierre, L.M.; Cooper, C.L.; Abarca, N.; Brough, P.; Ferreiro, P.; Fraile,
G.; et al. Flexible Work Arrangements Availability and their Relationship with Work-to-Family Conflict, Job Satisfaction, and
Turnover Intentions: A Comparison of Three Country Clusters. Appl. Psychol. 2012, 61, 1–29. [CrossRef]
23. de Sivatte, I.; Guadamillas, F. Antecedents and outcomes of implementing flexibility policies in organizations. Int. J. Hum.
Resour. Manag. 2013, 24, 1327–1345. [CrossRef]
24. Mullins, L.B.; Charbonneau, É.; Riccucci, N.M. The effects of family responsibilities discrimination on public em-ployees’
satisfaction and turnover intentions: Can flexible work arrangements help? Rev. Public Pers. Adm. 2021, 41, 384–410. [CrossRef]
25. Weiss, H.M. Deconstructing job satisfaction: Separating evaluations, beliefs and affective experiences. Hum. Resour. Manag. Rev.
2002, 12, 173–194. [CrossRef]
26. Eliyana, A.; Ma’Arif, S. Muzakki Job satisfaction and organizational commitment effect in the transformational leadership
towards employee performance. Eur. Res. Manag. Bus. Econ. 2019, 25, 144–150. [CrossRef]
27. Boone, H.N.; Boone, D.A. Analyzing likert data. J. Ext. 2012, 50, 1–5.
28. Albion, M.J. A Measure of Attitudes Towards Flexible Work Options. Aust. J. Manag. 2004, 29, 275–294. [CrossRef]
29. Morgeson, F.P.; Humphrey, S.E. The Work Design Questionnaire (WDQ): Developing and validating a comprehensive measure
for assessing job design and the nature of work. J. Appl. Psychol. 2006, 91, 1321–1339. [CrossRef]
30. Chen, Z.X.; Francesco, A.M. Employee Demography, Organizational Commitment, and Turnover Intentions in China: Do Cultural
Differences Matter? Hum. Relat. 2000, 53, 869–887. [CrossRef]
31. Grubor, A.; Berber, N.; Aleksić, M.; Bjekić, R. The influence of corporate social responsibility on organizational performance: A
research in AP Vojvodina. Anal. Èkon. Fak. U Subotici 2020, 43, 3–13. [CrossRef]
32. Berber, N.; Slavić, A.; Aleksić, M. Relationship between Perceived Teamwork Effectiveness and Team Performance in Banking
Sector of Serbia. Sustainability 2020, 12, 8753. [CrossRef]
33. Hair, J.F.; Hult, G.T.M.; Ringle, C.M.; Sarstedt, M. A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM); Sage:
Thousand Oaks, CA, USA, 2017.
34. Rahimnia, F.; Hassanzadeh, J.F. The impact of website content dimension and e-trust on e-marketing effectiveness: The case of
Iranian commercial saffron corporations. Inf. Manag. 2013, 50, 240–247. [CrossRef]
35. Taber, K.S. The Use of Cronbach’s Alpha When Developing and Reporting Research Instruments in Science Education.
Res. Sci. Educ. 2018, 48, 1273–1296. [CrossRef]
36. Bjekić, R.; Jelača, M.S.; Berber, N.; Aleksić, M. Factors Affecting Entrepreneurial Intentions of Faculty Students. Manag. Sustain.
Bus. Manag. Solut. Emerg. Econ. 2020, 26, 1–14. [CrossRef]
37. Hair, J.F.; Hult, G.T.M.; Ringle, C.M.; Sarstedt, M.; Kuppelwieser, V.G. A Partial least squares structural equation modeling
(PLS-SEM): An emerging tool in business research. Eur. Bus. Rev. 2014, 26, 106–121. [CrossRef]
38. Sabi, H.M.; Uzoka, F.-M.E.; Langmia, K.; Njeh, F.N. Conceptualizing a model for adoption of cloud computing in education. Int. J.
Inf. Manag. 2016, 36, 183–191. [CrossRef]
39. Peterson, R.A.; Kim, Y. On the relationship between coefficient alpha and composite reliability. J. Appl. Psychol. 2013, 98, 194–198.
[CrossRef] [PubMed]
40. Fornell, C.; Larcker, D.F. Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res.
1981, 18, 39. [CrossRef]
41. Rouf, A.; Akhtaruddin, M. Factors affecting the voluntary disclosure: A study by using smart PLS-SEM approach.
Int. J. Law Manag. 2018, 60, 1498–1508. [CrossRef]
42. Ab Hamid, M.; Sami, W.; Sidek, M.M. Discriminant validity assessment: Use of Fornell & Larcker criterion versus HTMT criterion.
In Proceedings of the Journal of Physics: Conference Series, Kuantan, Malaysia, 8–10 August 2017; p. 012163.
43. Chin, W.W. The partial least squares approach to structural equation modeling. Mod. Methods Bus. Res. 1998, 295, 295–336.
44. Hair, J.F.; Risher, J.J.; Sarstedt, M.; Ringle, C.M. When to use and how to report the results of PLS-SEM. Eur. Bus. Rev. 2019, 31,
2–24. [CrossRef]
45. Kock, N. Common Method Bias in PLS-SEM. Int. J. E-Collab. 2015, 11, 1–10. [CrossRef]
46. O’Brien, R.M. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Qual. Quant. 2007, 41, 673–690. [CrossRef]
47. Wong, K.K.K. Partial least squares structural equation modeling (PLS-SEM) techniques using SmartPLS. Mark. Bull. 2013, 24,
1–32.
48. Iqbal, S.; Martins, J.M.; Mata, M.N.; Naz, S.; Akhtar, S.; Abreu, A. Linking Entrepreneurial Orientation with Innovation
Performance in SMEs; the Role of Organizational Commitment and Transformational Leadership Using Smart PLS-SEM.
Sustainability 2021, 13, 4361. [CrossRef]
49. Bjekić, R.; Rodić, M.; Aleksić, M.; Gašić, D. Relationship between social competences of manager and leadership outcomes.
Èkonomika 2021, 67, 47–57. [CrossRef]
50. Haque, A.U.; Yamoah, F.A. The Role of Ethical Leadership in Managing Occupational Stress to Promote Innovative Work
Behaviour: A Cross-Cultural Management Perspective. Sustainability 2021, 13, 9608. [CrossRef]
Sustainability 2022, 14, 4502 14 of 14

51. Lukić, J.M.; Vračar, M.M. Building and nurturing trust among members in virtual project teams. Strat. Manag. 2018, 23, 10–16.
[CrossRef]
52. Kontić, L.; Vidicki, Ð. Strategy for digital organization: Testing a measurement tool for digital transformation. Strat. Manag. 2018,
23, 29–35. [CrossRef]