Work engagement and voluntary absence: The moderating role of job resources

Abstract

The present study examined the moderating role of job resources, namely, organizational trust, the quality of employees’ relationship with their manager, and the motivating potential of jobs, on the negative relationship between work engagement and voluntary absence. Employee survey results and absence records collected from the Human Resources department of a construction and consultancy organization in the United Kingdom (n = 325) showed that work engagement was negatively related to voluntary absence, as measured by the Bradford Factor. Furthermore, the results showed that organizational trust and the quality of employees’ relationships with their line managers ameliorated the negative effect of relatively low levels of engagement on voluntary absence. Theoretical and practical implications of the findings are discussed.

Full text

Work engagement and voluntary absence: The moderating role of job
resources
Amanda Shantz
1
and Kerstin Alfes
2
1
IÉSEG School of Management, Université Catholique de Lille, Paris, France
2
Department of Human Resource Studies, Tilburg University, Tilburg, The Netherlands
The present study examined the moderating role of job resources, namely, organizational trust, the quality of employees’
relationship with their manager, and the motivating potential of jobs, on the negative relationship between work engagement
and voluntary absence. Employee survey results and absence records collected from the Human Resources department of a
construction and consultancy organization in the United Kingdom (n= 325) showed that work engagement was negatively
related to voluntary absence, as measured by the Bradford Factor. Furthermore, the results showed that organizational trust and
the quality of employees’relationships with their line managers ameliorated the negative effect of relatively low levels of
engagement on voluntary absence. Theoretical and practical implications of the findings are discussed.
Keywords: Engagement; Voluntary absence; Organizational trust; Leader-member exchange; Motivating potential score.
Research has produced staggering estimates of the cost
of employee absenteeism to organizations. For instance,
the cost of absence amounts to 12.1% of total annual
payroll expenses in Canada (Towers Watson, 2012), $74
billion annually in the United States (Conlin, 2007), and
approximately £600 per employee each year in the
United Kingdom (CIPD, 2010). Aside from the direct
cost implications of employee absenteeism for organiza-
tions, it also has indirect costs. For instance, absent
employees may jeopardize the completion of a project,
miss out on opportunities with clients or customers, and/
or detract from the effectiveness of others at work
(Sagie, Birati, & Tziner, 2002).
Although research has identified that a potential anti-
dote to absence is work engagement (e.g., Schaufeli,
Bakker, & Van Rhenen, 2009; Soane et al., 2013), no
research, to our knowledge, has analysed factors that
may influence this relationship. The main contribution
of the present study lies in the building of theoretical
arguments and an empirical examination of job resources
that moderate the relationship between engagement and
absence. Specifically, we argue that work engagement
and job resources produce a multiplicative effect such
that when employees experience high levels of both,
employee absenteeism is lowest. For employees who
are relatively disengaged, we argue that job resources
buffer the negative effect of low levels of engagement on
absence. In other words, when engagement is relatively
low, job resources may produce a compensatory effect,
causing absence levels to be lower than if engagement
and the job resource were both absent.
ABSENTEEISM
Although absence is a complex phenomenon involving
the interplay among societal, workplace, and personal
factors (e.g., Dekkers-Sánchez, Hoving, Sluiter, &
Frings-Dresen, 2008), absence can be classified as either
involuntary or voluntary. In other words, employees are
absent because they are either unable (involuntary
absence) or unwilling (voluntary absence) to attend
work (e.g., Chadwick-Jones, Nicholson, & Brown,
1982; Johns, 1997; Schaufeli et al., 2009).
In most studies, voluntary absence is measured as
absence frequency (i.e., the number of absence episodes)
and involuntary absence is measured as absence duration
(i.e., the number of days absent). Although some
research finds support for this dichotomy (e.g., Bakker,
Demerouti, de Boer, & Schaufeli, 2003; Chadwick-Jones
et al., 1982; Schaufeli et al., 2009), researchers’agree-
ment about the measurement of voluntary absence is not
unanimous; some argue that equating absence frequency
Correspondence should be addressed to Amanda Shantz, IÉSEG School of Management, Socle de la Grande Arche, 1 Parvis de la Défense, 92044
Paris, France. E-mail: a.shantz@ieseg.fr
European Journal of Work and Organizational Psychology, 2015
Vol. 24, No. 4, 530–543, http://dx.doi.org/10.1080/1359432X.2014.936392
© 2014 Taylor & Francis

with voluntary absence is artificial and misleading (e.g.,
Farrell & Stamm, 1988; Shapira-Lishchinsky &
Rosenblatt, 2009; ten Brummelhuis, ter Hoeven, de
Jong, & Peper, 2013). Moreover, the results of meta-
analyses (Farrell & Stamm, 1988; Hackett & Guion,
1985; Scott & Taylor, 1985) do not provide a strong
case for treating absence frequency as a measure of
voluntary absenteeism.
Some scholars have suggested that measures of
absence duration and frequency are intertwined
(Harvey & Nicholson, 1999; ten Brummelhuis et al.,
2013) and that rather than forming two distinct cate-
gories, the duration/frequency typology represents a con-
tinuum of degrees of voluntary employee absence
(Brooke, 1986). Hence, it is more useful to use a mea-
sure of voluntary absence that emphasizes frequency of
absence, but also incorporates its duration.
In the present study, we used the Bradford Factor
because it takes into consideration both the number of
incidences and the duration of each incident to compute
an absence score for each employee. In other words, the
Bradford Factor measures an employee’s irregularity of
attendance and is calculated as (number of absence epi-
sodes)
2
× number of days absent. For example, one
absence episode of 10 days in duration equals 10 points;
10 absence episodes of 1 day each equal 1,000 points.
This formula emphasizes the frequency with which
absence spells occur, but it does not ignore the duration
of the absence episodes. Hence, it allows a more holistic
snapshot of an individual’s absenteeism and is therefore
a suitable way to capture voluntary absence caused by a
lack of motivation (Taylor, 2008).
Absenteeism and work engagement
Employee engagement is a promising antecedent of
voluntary workplace attendance because it is central to
an energetic and motivational work-related process (e.g.,
Bakker & Demerouti, 2007). Engagement is defined as a
“positive, fulfilling work-related state of mind character-
ized by vigor, dedication and absorption”(Schaufeli,
Salanova, Gonzalez-Roma, & Bakker, 2002, p. 20).
Engaged employees exert a high level of energy and
persist in the face of difficulties (vigour); they are highly
involved in work and thereby experience a sense of pride
and enthusiasm for it (dedication), and they are fully
focused on work so that time appears to pass by quickly
(absorption).
Engagement is likely to be inversely related to absen-
teeism for a number of reasons. First, Johns (1997)
argued that voluntary, as opposed to involuntary
absence, is best explained by models that focus on psy-
chological job attitudes, such as engagement. Second,
engaged employees are self-determined to accomplish
tasks despite perceived obstacles. In the face of setbacks
at work, engaged employees are less likely to be volun-
tarily absent, and instead, they relish in challenges
presented to them at work. Third, engaged employees
find their work stimulating, which draws them to spend
more time at work. Finally, being engrossed in one’s
work also contributes to lower voluntary absence rates,
as employees who are fully absorbed in work experience
flow, that is, they find their work intrinsically enjoyable
and difficult to detach from (Csikszentmihalyi, 1990).
Three published studies, to our knowledge, have
examined the relationship between engagement and
absence. The results revealed that engagement is nega-
tively related to company-registered number of absence
days (absence duration) for support service workers in
the United Kingdom (Soane et al., 2013), self-report
absence days (absence duration) of coworkers of a
Dutch police organization (ten Brummelhuis, Bakker,
& Euwema, 2010) and the number of company-regis-
tered absence episodes (absence frequency) and absence
days (absence duration) of telecom managers in the
Netherlands (Schaufeli et al., 2009). In the present
study, we test the hypothesis that engagement is inver-
sely related to company-registered absence using the
Bradford Factor to compute voluntary absence:
Hypothesis 1: Engagement is negatively related to
voluntary absence.
JOB RESOURCES
Job resources refer to physical, social or organizational
aspects of the job that reduce the physiological or psy-
chological costs of job demands; are functional in
achieving work goals and/or stimulate personal growth
and learning (Bakker & Demerouti, 2007; Demerouti,
Nachreiner, Baker, & Schaufeli, 2001). Resources are
important in their own right (Hackman & Oldham,
1975,1976) and also as a means to achieve and protect
other valued resources (Hobfoll, 1989). In the present
study, we examine organizational trust (an organization-
level resource), employees’relationships with their lea-
der (an interpersonal-level resource) and the motivating
potential of jobs (a task-level resource) as job resources
(Bakker & Demerouti, 2007).
These three job resources were chosen for two rea-
sons. First, they have theoretical and empirical ties with
absenteeism (e.g., Colquitt, Scott, & LePine, 2007; Fried
& Ferris, 1987; van Dierendonck, Le Blanc, & van
Breukelen, 2002). Second, Kahn (1990) suggested that
elements of one’s work environment, such as organiza-
tional trust, leadership and task characteristics, are rele-
vant in understanding how engagement unfolds at work.
He implied that features of the work environment set
boundaries on the extent to which engagement is likely
to be fully expressed at work. Specifically, he stated that
when work situations are clear, consistent and predict-
able (i.e., high levels of organizational trust), when man-
agement is supportive and clarifying (e.g., high-quality
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 531

manager–employee relationships), and when work tasks
are organized such that employees have autonomy and
variety in their work (e.g., high levels of motivational
job design), employees feel safe and able to fully express
their engagement at work.
Organizational trust
Organizational trust refers to employee perceptions of
the employer’s integrity, motives, openness and beha-
vioural consistency (Dirks & Ferrin, 2001; Robinson &
Rousseau, 1994). Organizational trust, as a job resource,
is functional in achieving work goals, because it
increases an employee’s willingness to take risks and
increases employees’identification with the organization
(Edwards & Cable, 2009). Trust in the organization also
encourages employees to share knowledge amongst each
other (McEvily, Perrone, & Zaheer, 2003), thereby sti-
mulating personal growth and learning.
Leadership
Leader–member exchange (LMX) refers to the quality
of the relationship between a subordinate and his/her
leader. A high-quality LMX relationship is character-
ized by reciprocal trust, respect and the expectation of a
future positive relationship, whereas a low-quality rela-
tionship is characterized by just the opposite; it is based
solely on the fulfilment of the employment contract
(Graen & Uhl-Bien, 1995). Employees in high-quality
relationships receive increased communication, better
roles, have higher levels of emotional support and
greater access to organizational resources, compared
to those in lower-quality LMX relationships (Dienesch
&Liden,1986; Graen & Scandura, 1987;Wayne,
Shore, & Liden, 1997). Hence, a high-quality LMX
relationship is a job resource because it helps employ-
ees achieve important work goals, provides a support
mechanism to defend against job demands and spurs
employee development.
Job design
The third moderator under consideration is the motivat-
ing potential of jobs. Hackman and Oldham (1975)
proposed that five job characteristics have the potential
to intrinsically motivate employees: task variety, task
identity, task significance, autonomy and feedback from
the job. An overall motivating potential score (MPS) is
typically calculated from the core job dimensions. That
MPS is considered a job resource is consistent with
Hackman and Oldham’s(1980) contention that task
characteristics are motivational, assist in achieving
work goals and facilitate jobholders’learning and
development.
THE MODERATING ROLE OF JOB
RESOURCES
Research has shown that both engagement and job
resources are negatively related to absenteeism (e.g.,
Bakker et al., 2003; Schaufeli et al., 2009; Soane et al.,
2013). Despite this, the two variables have seldom been
considered concomitantly in a study on absenteeism. In
fact, only one study has simultaneously examined
engagement and job resources on absenteeism.
Schaufeli et al. (2009) found that increases in job
resources predicted work engagement, and engagement
was negatively related to absence duration and fre-
quency. This study, however, did not consider an inter-
action between engagement and job resources on
absenteeism. Although job resources have typically
been studied as antecedents of work engagement (e.g.,
Bakker & Demerouti, 2007), an influential review of the
absence literature suggested that absence is likely most
strongly influenced by combinations (interactions) of
variables (Johns, 1997). Ignoring the potential interac-
tion between engagement and job resources on absence
may be limiting, since there are good conceptual reasons
to posit that an interaction might take place.
There are at least three theoretical frameworks that
support the contention that absence is lowest for those
who are highly engaged and have access to job
resources. First, conservation of resources (COR; e.g.,
Hobfoll, 1989,2011) theory suggests that organizational
trust, LMX and MPS are job resources. Gorgievski-
Duijvesteijn and Hobfoll (2008) added that engagement
is an “intrinsic energetic resource”and that resources
operate synergistically to produce in employees an
even greater ability to overcome obstacles, perform at
higher levels and maintain a healthy sense of well-being.
They theorized that there is a potential for an interaction
(i.e., moderation) between individual resources, such as
engagement, and job resources.
Second, according to Broaden and Build Theory
(Fredrickson, 1998,2001), a positive state such as
work engagement has the capacity to broaden an indivi-
dual’s momentary thought-action repertoire through
expanding the obtainable array of potential thoughts
and action that come to mind (Fredrickson & Branigan,
2005; Wright, 2005). Experiencing the positivity gener-
ated from work engagement may lead individuals to
thrive and flourish. Additional resources may strengthen
this effect (Fredrickson, 1998,2001). In other words,
Fredrickson’s model suggests that engagement may
interact with job resources because of their complemen-
tary broadening (engagement) and building (job
resources) effects.
Third, Blumberg and Pringle (1982) argued that in
order to understand employee behaviour, it is necessary
to examine not only the employees’ability and motiva-
tion but also the opportunity that they have to invest
themselves in their work. Employees have an
532 SHANTZ AND ALFES

opportunity to invest themselves in work if they have
adequate resources (e.g., high trust in the organization,
high-quality LMX relationship and an enriched job).
Importantly in the present discussion, they stressed that
motivation (i.e., engagement) and opportunity (i.e., job
resources) are not only additive, but they also interact,
such that those with the highest motivation and the most
opportunity are most likely to achieve the highest level
of performance. In the same way, employees who are
engaged are less absent in their role when they are
provided with a resource-laden environment (e.g., high
trust in their organization, high-quality LMX relationship
and enriched jobs). Taken together, these three theoreti-
cal models imply that job resources and engagement
produce a synergistic effect, such that absence is lowest
when both are high.
Job resources may not only strengthen the negative
relationship between engagement and absence for highly
engaged employees; research on the buffering hypothesis
(Caplan, 1974) suggests that job resources may also play
a buffering role for those who are relatively disengaged
at work. The theory was originally developed to explain
the effect of social support on the outcomes of stress. It
stipulates that support can ameliorate the deleterious
effects of stress on individuals’health and well-being
and that support has little impact on individuals who are
not stressed. Cohen and McKay (1984) elaborated upon
the buffering hypothesis in positing that resources can
buffer the negative consequences of stress if the
resources either provide the person with tangible tools
to overcome the stress, if they assist in cognitive reap-
praisal of the situation, and/or if they provide the person
with positive feedback or a sense of belonging.
The buffering hypothesis (Caplan, 1974) and empiri-
cal evidence supporting it (e.g., Miner, Settles, Pratt-
Hyatt, & Brady, 2012) provide theoretical support for
the interaction between engagement and job resources
specifically at low levels of engagement. Research on the
buffering role of job resources on engagement (Bakker,
Hakanen, Demerouti, & Xanthopoulou, 2007), burnout
(Xanthopoulou et al., 2007) and absence frequency
(Demerouti, Bouwman, & Sanz-Vergel, 2011) likewise
support the hypothesis that job resources play a buffering
role. Hence, employees with low levels of engagement
may nevertheless have higher levels of attendance if they
are provided with adequate job resources.
Hypothesis 2: Organizational trust moderates the
negative relationship between engagement and
voluntary absence.
Hypothesis 3: A high-quality LMX relationship
moderates the negative relationship between
engagement and voluntary absence.
Hypothesis 4: High levels of MPS (motivating
potential score) of jobs moderates the negative
relationship between engagement and voluntary
absence.
METHODS
Participants
The data for this study were sourced from a construction
and consultancy organization in the United Kingdom.
The organization offers integrated services across the
property and infrastructure life cycle. Their work
involves project management, construction delivery and
facilities management across a number of sectors includ-
ing arts and culture, education, energy and utilities,
hotels and sport, science and technology, and transport.
The organization has a relatively large Human Resources
(HR) department, which serves as an advisor on Human
Resource Management (HRM) issues to line managers.
They also liaise between employees and line
management.
Within the HRM department, there is an Absence
Partner who takes calls from absent employees and
reports them to the absent employee’s line manager and
project coordinator. The Occupational Health Advisor
periodically reviews absence information and follows
up with employees who are frequently absent via tele-
phone calls. Notwithstanding these efforts, employee
absence has been a growing concern for the HR
Director of the organization, as absence rates have stea-
dily increased from approximately 2.5 days per
employee per year in 2007 to 5.2 days per employee
per year in 2012. The HR Director acknowledged that
although there is little she can do to remedy involuntary
absenteeism (e.g., sickness), she is interested in finding
ways to reduce voluntary absenteeism. Hence, the cur-
rent study was of particular interest to this organization.
Electronic surveys were sent to 671 employees of this
organization. The respondents were encouraged to parti-
cipate in the survey within 2 weeks. All employees were
given time to complete the survey at work. Employees
were informed of the purpose of the data collection (i.e.,
to gather opinions of their work) and its confidentiality.
Four hundred and fourteen surveys were completed.
Eighty-nine respondents were excluded due to incom-
plete absence data. These employees worked in overseas
offices where absence data are not collected by the HR
department. Hence, the response rate was 48%.
Of the 325 remaining participants, 60.6% were male;
the mean age of participants was 40.12 (SD = 11.55); the
proportion of married individuals was 46%.
Approximately 54% of the sample had a university
degree or above, 10% had other higher (e.g., college)
education, 12% had A-levels (preuniversity/college cred-
its) or equivalent, 15% had GCSE (high school diploma)
or equivalent, 7% had other job-related qualifications
(e.g., UK National Vocational Qualifications include
job-related sector courses, such as Basic Plumbing
Studies or Business Administration) and 2% had no
qualifications. Approximately 35% of the respondents
were engineering and construction professionals, 20%
held administrative roles, 24% were managers, 5% held
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 533

customer service roles, 2% were general labourers and
the remaining employees indicated “other type of work”.
Tests were conducted to determine whether there were
significant differences between the 89 excluded employ-
ees (due to missing absence data) and 325 included
employees in terms of gender, age, marital status, type
of work and education level. All tests were not signifi-
cant. Absence records were obtained from the HR
department 4 months subsequent to the completion of
the survey.
Measures
Scale reliabilities are found in Table 1. All items for all
scales were scored on a 1 (strongly disagree) to 7
(strongly agree) scale, unless otherwise noted.
Engagement. Engagement was measured using Schaufeli
et al.’s(2002) self-report questionnaire consisting of 17
items, which capture the three dimensions of engagement,
namely, vigour (e.g., When I get up in the morning, I feel
like going to work), dedication (e.g., I am enthusiastic about
my job) and absorption (e.g., Time flies when I am work-
ing). The items were scored on a 1 (never) to 7 (always)
scale. A mean score was calculated from the three compo-
nents to reflect an overall measure of engagement.
Organizational trust. Robinson and Rousseau’s(1994)
scale captures the extent to which employees trust their
organization (e.g., I believe my employer has high
integrity).
Leadership. The quality of the relationship with one’s
leader was measured using the LMX scale developed by
Graen and Uhl-Bien (1995; e.g., My working relation-
ship with my leader is effective).
Job design. Job design was operationalized as a MPS.
The MPS was calculated by taking the average of five
job characteristics using scales developed by Morgeson
and Humphrey (2006), namely, decision-making auton-
omy (three items, e.g., The job provides me with sig-
nificant autonomy in making decisions), task variety
(four items, e.g., The job requires the performance of a
wide range of tasks), task significance (four items; e.g.,
The job itself is very significant and important in the
broader scheme of things), task identity (e.g., four items;
The job is arranged so that I can do an entire piece of
work from beginning to end) and feedback from the job
(three items; e.g., The job itself provides feedback on my
performance).
Voluntary absence. The number of recorded absence
episodes and duration over 4 months subsequent to the
completion of the survey were collected from organizational
records. Each participant was assigned a Bradford Factor
score calculated as (number of absence episodes)
2
×
number of days absent.
Control variables. Age, gender (female = 1) and
whether an employee managed others (manages
others = 1) were used as control variables in all analyses,
because meta-analyses suggest that they are related to
absence (e.g., Darr & Johns, 2008; Martocchio, 1989).
All of the analyses presented later were conducted twice,
once with and once without control variables. The results
were consistent across the analyses (Becker, 2005).
Hence, we are confident that the inclusion of the control
variables did not alter our main findings. The results
discussed in the following present the analyses that
include the control variables.
RESULTS
Descriptive statistics and preliminary
analyses
SPSS (version 21) was used to calculate the descriptive
statistics and to test the hypotheses. Table 1 presents the
scale reliabilities, means, standard deviations and corre-
lations amongst the variables.
Tests of discriminant validity. As the measures for
engagement, organizational trust, leadership and job
design were collected at one point of time, a series of
confirmatory factor analyses were carried out in AMOS
TABLE 1
Descriptive statistics
Alpha Mean SD 1 2 3 4 5 6 7
1. Gender
2. Age 40.12 11.55 −.13*
3. Manages others .15** −.22**
4. Engagement .91 4.96 .73 −.06 .21** .28**
5. MPS .81 5.15 .99 −.05 −.01 .21** .57**
6. Leadership .94 5.17 1.30 −.02 −.01 .10 .27** .31**
7. Trust .97 4.80 1.31 −.10 .01 .17** .44** .48** .52**
8. Absence 1.39 4.65 −.01 .02 −.01 −.19** −.13* −.11* −.16**
*p< .05, **p< .01; female = 1, male = 0; manages others = 1, does not manage others = 0.
534 SHANTZ AND ALFES

(version 21) to assess the discriminant validity of the
measures. A full measurement model was initially tested,
in which the three facets of engagement loaded onto a
general engagement factor, the five facets of job design
loaded onto a general MPS factor and all indicators for
organizational trust and leadership were allowed to load
onto their respective factors. All factors were allowed to
correlate. Five fit indices were calculated to determine
how the model fitted the data (Hair, Black, Babin, &
Anderson, 2009). For the χ
2
/df, values less than 2.5
indicate a good fit and values around 5.0 an acceptable
fit (Arbuckle, 2006). For the Comparative Fit Index
(CFI) and Tucker–Lewis coefficient (TLI), values
above .90 are recommended as an indication of good
model fit (Bentler, 1990; Bentler & Bonett, 1980). For
the Root Mean Square Error of Approximation
(RMSEA) and Standardized Root Mean Square
Residual (SRMR), values less than .06 indicate a good
model fit and values less than .08 an acceptable fit
(Browne & Cudeck, 1993; Hu & Bentler, 1998). The
four-factor model showed a good model fit(χ
2
= 592;
df = 202; CFI = .93; TLI = .92; RMSEA = .077;
SRMR = .055). All factor loadings were above the
suggested threshold of .5 (between .63 and .95) and
significant at the p< .001 level. Next, sequential χ
2
difference tests were carried out. Specifically, the full
measurement model was compared to six alternative
nested models, as shown in Table 2. Three models (A,
B and C) were created to assess the distinctiveness of the
independent variable from each moderator variable.
Model D was created to assess the distinctiveness of
trust and LMX, given the possibility that employee rat-
ings of these constructs are affected by halo error, such
that employees who provide a high rating to their leader
may also provide a high rating to their organization
because they view their leader as an agent of the orga-
nization. Model E was created to assess whether the
three different job resources were distinct. Finally, a
single-factor model was tested (Model F). Results of
the measurement model comparison revealed that the
model fit of the alternative models was significantly
worse compared to the full measurement model (all at
p< .001). This suggests that the variables in this study
are distinct.
Moreover, we carried out a number of tests to ensure
that each hierarchical moderated regression model met
the assumptions of regression (Field, 2013; Stevens,
2002). Relevant test statistics are available by request
from the first author.
Hypothesis testing
Hypothesis 1 predicted that engagement is negatively
related to absence. The results of a multiple regression
analysis indicated that the model was significant
(F= 2.76, p< .05, R
2
= .04), and after controlling for
age, gender and managerial role, engagement was sig-
nificantly and negatively related to absence (β=−.21,
SE =.36,t=−3.22, p< .01). Hypothesis 1 was
supported.
Hierarchical moderated regressions were used to test
the remaining hypotheses. Variables entered into moder-
ated regressions were standardized (Aiken & West,
1991). Table 3 presents the results. In the first column,
labelled Baseline model, absence was regressed on gen-
der, age and managerial role only. In the column labelled
Moderating role of organizational trust, the first subcol-
umn shows the results of the main effect model (con-
trols, engagement and organizational trust) and the
second subcolumn shows the moderated regression
model (controls, engagement, organizational trust and
the interaction between engagement and organizational
trust) on absenteeism. The columns that represent the
TABLE 2
Fit statistics from measurement model comparison
Models χ
2
(df) CFI TLI RMSEA SRMR χ2
diff dfdiff
Full measurement model 592 (202) .930 .920 .077 .055
Model A
a
944 (205) .867 .850 .105 .088 352 3***
Model B
b
1043 (205) .849 .830 .112 .121 451 3***
Model C
c
698 (205) .911 .900 .086 .065 106 3***
Model D
d
1923 (205) .690 .651 .161 .131 1331 3***
Model E
e
2214 (207) .638 .596 .173 .149 1622 5***
Model F
f
(Harman’s Single-Factor Test) 2479 (208) .590 .545 .184 .155 1887 6***
***p< .001; χ
2
= chi-square discrepancy, df = degrees of freedom; CFI = Comparative Fit Index; TLI = Tucker–Lewis Coefficient; RMSEA = Root
Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual; χ2
diff = difference in chi-square, dfdiff = difference in
degrees of freedom. All models are compared to the full measurement model.
a
Engagement and organizational trust combined into a single factor.
b
Engagement and leadership combined into a single factor.
c
Engagement and motivating potential score combined into a single factor.
d
Organizational trust and leadership combined into a single factor.
e
Organizational trust, leadership and motivating potential score combined into a single factor.
f
All factors combined into a single factor.
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 535

moderating role of leadership (LMX) and job design
(MPS) are similarly presented.
The second hypothesis predicted that organizational
trust would moderate the relationship between engage-
ment and absence. The interaction between engagement
and trust was significant (column 3 of Table 3). The
relationship is plotted in Figure 1.
Figure 1 reveals that employees with relatively low
levels of both trust and engagement showed the highest
level of voluntary absence. Simple slope analyses indi-
cated that at relatively low levels of organizational trust,
engagement was negatively related to absence (β=−.25,
SE = .36, t=−3.16, p< .05), whereas at relatively high
levels of organizational trust, engagement was not
related to absence (β=−.03, SE = .41, t=−.29,
p= n.s.). Hypothesis 2 was partially supported.
We also examined whether the end points of trust at
high versus low levels of engagement (see Figure 1)
were different from one another. To do so, we swapped
the independent variable (engagement) and moderator
(trust) and conducted an additional simple slopes test
(Dawson, 2014). The results showed that the slope of
the line at high levels of engagement was not signifi-
cantly different from zero (β=−.49, SE = .40, t= .23,
p= n.s.), whereas the slope of the line for those with
lower levels of engagement was negative and signifi-
cantly different from zero (β=−.88, SE = .33,
t=−2.66, p< .01). This implies that the two points on
Figure 1 at lower levels of engagement are significantly
different from one another. Hence, trust makes a differ-
ence at lower levels of engagement.
The second hypothesis predicted that the quality of
the relationship employees have with their leader mod-
erates the negative relationship between engagement and
absence. Column 5 of Table 3 presents the results of this
hierarchical moderated regression. The interaction term
was positive and significant. The relationship is plotted
in Figure 2.
TABLE 3
Hierarchical moderated regressions on employee absenteeism
Baseline
model
a
Moderating role of
organizational trust
b
Moderating role of
leadership (LMX)
Moderating role of job
design (MPS)
β(SE) β(SE) β(SE) β(SE) β(SE) β(SE) β(SE)
Step 1 Gender .01 (.55)
c
Age .02 (.03)
Manager .01 (.55)
Step 2 Gender .02 (.57) −.01 (.57) −.01 (.57)
Age .04 (.03) .05 (.03) .05 (.03)
Manager .05 (.58) .05 (.58) .05 (.59)
Engagement −.17 (.32)* −.19 (.30)** −.19 (.35)*
Trust −.10 (.31)
Leadership −.07 (.28)
MPS −.04 (.34)
Step 3 Gender .01 (.56) −.02 (.56) .01 (.55)
Age .04 (.02) .05 (.02) .04 (.02)
Manager .06 (.58) .05 (.58) .06 (.57)
Engagement −.15 (.32)* −.19 (.29)** −.18 (.34)*
Trust −.08 (.31) −.04 (.28) .06 (.34)
Engage × Trust .13 (.23)*
Engage × LMX .15 (.26)**
Engage × MPS .28 (.23)**
Adj R
2
(Δ
d
) 0 .03** .05* .03** .05** .02** .09**
F.06 2.84** 3.22** 2.60* 3.31** 2.40* 5.56**
a
The baseline model includes the control variables (gender, age and management responsibilities) only.
b
The first subcolumn reports the main effects of controls, engagement and organizational trust on absenteeism; the second column reports the
moderated regression results.
c
Standardized regression coefficients (standard error).
d
Indicates whether R
2
was significantly improved from previous step.
*p< .05, **p< .01.
–2
–1.5
–1
–0.5
0
0.5
1
1.5
2
2.5
3
Absence
Low Trust
High Trust
Low Engagement High Engagement
Figure 1. Organizational trust as a moderator on the relationship
between engagement and absence. Variables are plotted at one standard
deviation above and below the mean.
536 SHANTZ AND ALFES

Simple slope analyses indicated that at relatively
lower-quality LMX relationships, engagement was nega-
tively related to absence (β=−.33, SE = .39, t=−3.91,
p< .05), whereas when employees had relatively higher-
quality LMX relationships, engagement was not related
to absence (β=−.04, SE = .39, t=−.42, p= n.s.).
To examine whether there is a significant difference
between the end points of LMX at high levels of engage-
ment (see Figure 2), we swapped the independent vari-
able (engagement) and moderator (LMX) and conducted
an additional simple slopes test (Dawson, 2014). The
results showed that at lower levels of engagement, the
slope of the line is significantly different from zero
(β=−.81, SE = .34, t=−2.38, p< .05). Conversely,
the negative relationship between LMX and absence was
not significant at high levels of engagement (β=−.49,
SE = .39, t= 1.27, p= n.s.). This suggests that the two
points at the far side of Figure 2 are not significantly
different from one another. Hypothesis 3 was partially
supported.
The fourth hypothesis predicted that perceptions of
job design would moderate the negative relationship
between engagement and absence (see column 7 of
Table 3). The interaction term was positive and
significant.
Simple slope analyses indicated that for employees
who reported that their jobs contain relatively low levels
of motivating potential, engagement was negatively
related to absence (β=−.40, SE = .40, t=−4.62,
p< .05), whereas for those who report relatively high
levels of motivating potential in their jobs, engagement
was not related to absence (β=−.06, SE = .41, t=−.54,
p=n.s.).
To investigate whether the end points of MPS at high
and low levels of engagement are significantly different
from one another (see Figure 3), we swapped the inde-
pendent (engagement) and moderator (MPS score) and
conducted a second simple slopes test (Dawson, 2014).
The negative relationship between MPS and absence was
significantly different from zero for employees with rela-
tively higher levels of engagement (β= .99, SE = .45,
t=−2.18 p< .05), but not at lower levels of engagement
(β=−.25, SE = .35, t=−.72 p= n.s.). This implies that
the end points that represent employees at high levels of
engagement and high versus low levels of MPS are
significantly different from one another. In summary,
although there was a significant interaction between
engagement and MPS, the additional analyses suggest
that high levels of MPS do not influence the relationship
between engagement and absence. This unanticipated
finding is discussed later.
Additional test. In order to further validate our results,
we tested each of our hypotheses using moderated struc-
tural equation modelling (MSEM; Cortina, Chen, &
Dunlap, 2001; Mathieu, Tannenbaum, & Salas, 1992;
ten Brummelhuis, ter Hoeven, Bakker, & Peper, 2011).
The MSEM results and resulting plots mirror those pro-
duced by our hierarchical moderated regression analyses.
The details of the MSEM procedure and results are
available by request from the first author.
DISCUSSION
Although the buffering hypothesis (Caplan, 1974) was
originally developed to explain the impact of support on
the outcomes of stress, the theory and its associated
empirical research show a striking similarity with the
pattern of results found in this study, that is, at lower
levels of engagement, organizational trust and LMX
buffered the negative relationship between work engage-
ment and voluntary absence. These findings caution
against arguments that disengaged employees are neces-
sarily a cost to organizations (Wollard, 2011). Indeed, the
results of the present study show that work engagement
and some job resources compensate for one another.
The results also showed that for relatively highly
engaged employees, job resources did not enhance the
negative relationship between engagement and absentee-
ism. These results come as a surprise given that there are
theoretical arguments to suggest that the relationship
between engagement and absence is strengthened by
resources that help employees achieve work goals, to
cope with work challenges and to develop and grow.
–2
–1.5
–1
–0.5
0
0.5
1
1.5
2
2.5
3
Absence
Low LMX
High LMX
Low Engagement High Engagement
Figure 2. Leadership (LMX) as a moderator on the relationship
between engagement and absence. Variables are plotted at one standard
deviation above and below the mean.
–3
–2
–1
0
1
2
3
4
5
Absence
Low MPS
High MPS
Figure 3. Motivating Potential Scores (MPS) as a moderator on the
relationship between engagement and absence. Variables are plotted at
one standard deviation above and below the mean.
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 537

A potential reason for the nonsignificant finding may
be the positive correlation between work engagement
and the three job resources; highly engaged employees
are more likely to have more resources. A pressing
question is whether this is an empirical or theoretical
issue. Empirically, an interaction may not have surfaced
because of the positive relationship between job
resources and engagement. Hence, there may have been
insufficient power to detect such an interaction, espe-
cially given that this study was conducted in the field
(Morris, Sherman, & Mansfield, 1986). Additionally,
there may be a threshold effect, or restriction of range.
A highly engaged worker may not be intentionally
absent from work at all, or to a very small extent, so
that there is little room for a further decrease in voluntary
absence (Sagie, 1998). McClelland and Judd (1993)
stated that field studies are problematic because the dis-
tributions of the independent and dependent variables are
commonly restricted by their respective ranges, which is
exacerbated by the multiplication of the interaction term,
reducing the power to detect an interaction. Post hoc
tests of our data confirmed that this might explain our
results; the variance of absence was larger at lower levels
compared to higher levels of engagement.
Theoretically, it may be that job resources do not
interact with engagement at high levels to predict rele-
vant outcomes, such as absence. Although broaden and
build theory has been used to position engagement as a
mediator of the relationship between job resources and
outcomes, we used it to support an interaction between
engagement and job resources. More theory building is
needed to understand the condition(s), if any, under
which high levels of engagement can be strengthened.
Although theory and empirical work on the first part of
the JD-R model, that is, between job resources and
engagement, is relatively well developed (e.g., Bakker
& Bal, 2010; Xanthopoulou, Bakker, Demerouti, &
Schaufeli, 2009), in comparison, there is little theoretical
development and/or empirical work that have examined
moderators in the subsequent stages of the JD-R model,
that is, between work engagement and its outcomes.
Although we used existing theories to inform our
hypotheses, and some theory and empirical work sug-
gests that psychological states, such as engagement, may
interact with job resources (e.g., Hackman & Oldham,
1976; Johns, Xie, & Fang, 1992), our nonsignificant
results may point researchers to focus on other resources
(e.g., personal resources) that may amplify the relation-
ship between engagement and absence. Alternatively,
future research may examine whether for some types of
jobs (e.g., professional and managerial), job resources
are multiplicative, whereas for other types of jobs (e.g.,
manual), job resources and engagement act as substitutes
for one another.
Additional noteworthy points of discussion stem from
the additional analyses that we conducted for each mod-
eration hypothesis. Specifically, we examined whether
the points that represent high and low levels of engage-
ment at high and low levels of the moderator were
significantly different from one another (Dawson,
2014). The results for trust and LMX lend strength to
the argument that organizational trust and LMX compen-
sate for lower levels of engagement. The results for
MPS, however, paint a more complex picture.
A substantial number of highly engaged employees
had higher rates of voluntary absence when jobs were
high in motivating potential. Although this may seem
counterintuitive, a plausible rationale is that some highly
engaged employees with demanding and challenging
jobs are more likely to need to take time off from work
to recuperate. Macey and Schneider (2008) warned that
there might be limits to the amount of engagement that is
beneficial for employees and organizations. Halbesleben,
Harvey, and Bolino (2009) confirmed that too much
engagement is detrimental to employees. They found
that engagement is associated with higher levels of
work interference with family because employees ded-
icate too much of their time to helping others at work.
The results also showed that at low levels of engage-
ment, absenteeism did not vary with MPS. Although it is
plausible that this is a consequence of the high correla-
tion between engagement and MPS, we encourage future
research to take a closer look at the relationship between
lower levels of engagement and perceptions of job
design. The classic Hackman and Oldham model of job
design posits that a condition under which enriched jobs
produce their intended motivational effect is when job-
holders have a desire to grow and develop. The present
results may lend weight to the argument that at lower
levels of engagement, the effect of MPS may be con-
tingent on a third resource, such as growth needs
strength or a similar personal resource.
This finding also questions whether high levels of
motivational job characteristics have uniformly positive
outcomes. There have been some research findings, like
ours, which suggest that very high levels of MPS may be
detrimental to employees. For instance, research has
revealed that high responsibility (French & Caplan,
1973) and high mental and social demands
(Schaubroeck & Ganster, 1993) are associated with
stress. Research has also found a curvilinear U-shaped
relationship between job scope and emotional exhaus-
tion, such that at very high levels of job scope, employ-
ees become increasingly emotionally exhausted (Xie &
Johns, 1995).
Our findings therefore call for a reevaluation of
whether job characteristics are best conceived of as a
job resource and/or a job demand. In the case of job
design, there may be an optimal level in which job
characteristics act as a resource, but when the MPS
reaches a certain point, job characteristics may become
a job demand, causing the jobholder to experience stress
and ill health. This has implications for the study of
employee absence, which has tended to focus on the
538 SHANTZ AND ALFES

cause of absence as due to either ill health (i.e., can’t
attend work) or motivation (i.e., won’t attend work) and
suggests that different factors independently predict each
(e.g., Schaufeli et al., 2009). Task characteristics may in
fact predict both.
Future research should employ longitudinal, diary-
study and/or experimental designs to further investigate
the possibility that very high levels of both engagement
and MPS are detrimental for employees and organiza-
tions. For instance, researchers could follow employees
over time to examine how changes in engagement and
perceptions of job design influence important organiza-
tional outcomes such as absenteeism and performance.
Field experiments may involve a 2 (engaging work-
related task versus nonengaging work-related task) × 2
(provided with a role with high motivating potential
versus not provided with a role with high motivating
potential) design. For instance, if a sample of engineer-
ing technicians was attained, half could be provided with
an engaging task (e.g., constructing buildings) and the
other half with a less engaging task (e.g., tallying petrol
mileage to and from work sites). Each condition would
then be provided with either very high levels of feedback
from the job or not. Dependent measures may include
quality and quantity of performance, along with mea-
sures of well-being or emotional exhaustion, to tease out
whether there are negative consequences for employees
who have high levels of both engagement and motiva-
tional job design.
In summary, the present study contributes to the exist-
ing literature on both work engagement and absence.
Although the work engagement literature has primarily
conceptualized job resources as antecedents of work
engagement (e.g., Bakker & Demerouti, 2007), we
examined the interaction between job resources and
work engagement on absence. This is because Johns
(1997) suggested that voluntary absence might best be
predicted by interactions among psychological variables.
Whether an interaction would take place between
engagement and job resources on other outcomes, such
as task performance, citizenship behaviours or counter-
productive work behaviours, is a worthy avenue for
future research. So too is research on the effect of other
job resources to determine the extent to which other job
resources are interchangeable with the ones tested in the
present study. Although COR theory suggests that job
resources tend to travel together in “resource caravans”
(Hobfoll, 2011), future research is yet to confirm whether
some job resources do or do not moderate the effect of
engagement on relevant outcomes. The results of the
present study hint that resources may not always be
interchangeable with one another; MPS did not buffer
the effect of lower levels of engagement on absence like
organizational trust and LMX.
Our study also contributes to the absence literature by
using a relatively under-utilized measure of voluntary
absence, that is, the Bradford Factor. Future research
should consider using the Bradford Factor as a measure
of voluntary absence because it emphasizes the number
of absence spells, while not neglecting absence duration.
Practical implications
The primary practical implication of the present study is
to ensure that employees trust the organization and have
a high-quality relationship with their leader. Regardless
of whether these job resources lead to engagement, or act
as its substitute, ensuring that these job resources are in
place for employees is advantageous for organizations.
In order to engender organizational trust, employers
should implement clearly defined structures, roles and
guidelines with regard to decision-making and employee
conduct to provide direction about acceptable behaviour
at work (Gillespie & Dietz, 2009). Trust can also be
heightened when the organization emphasizes inclusive-
ness, open communication and individuality and
encourages its managers to provide ongoing feedback
to employees (Whitener, Brodt, Korsgaard, & Werner,
1998). Employers may also wish to consider investing in
corporate social responsibility initiatives. Such activities
signal to employees that the organization acts with moral
concern for the well-being of its stakeholders (Aguilera,
Rupp, Williams, & Ganapathi, 2007) and is negatively
related to absenteeism (Shapira-Lishchinsky &
Rosenblatt, 2009).
Organizations should also focus on ensuring that lea-
ders develop high-quality LMX relationships with their
employees. Mayfield and Mayfield (1998) suggested that
the formation of a high-quality LMX relationship begins
at recruitment. At this stage, the leader should provide a
realistic preview of the benefits and responsibilities of
the relationship. Once hired, a leader should begin the
LMX process through communicating mutual work
expectations with employees and informing them of
behaviours that will be rewarded. Research has estab-
lished a number of leader behaviours that can improve
LMX relationships, including leading by example,
recognizing subordinates for their successes, consulting
with employees on a variety of work-related matters and
delegating important tasks to employees (Yukl,
O'Donnell, & Taber, 2009). Organizations can also facil-
itate formal training programmes that invite both leaders
and followers and focus on an explanation of the LMX
process and outcomes and LMX communication training
(Mayfield & Mayfield, 1998). Research shows that LMX
relationships improve when leaders and members are
trained accordingly (e.g., Scandura & Graen, 1984).
This study also has practical implications for how HR
managers use engagement scores from annual employee
surveys. Many organizations today compute an engage-
ment index to evaluate their people management prac-
tices. Practitioners should be cautioned against relying
solely on such indices, as employee engagement scores
alone may not tell the full story; even if engagement
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 539

scores are relatively low, if employees are provided with
sufficient job resources, their level of absenteeism may
be similar to those who are engaged with their work.
Afinal implication of this study is a reminder to
consider the manner in which absence is measured in
organizations. There is some evidence from the practi-
tioner literature that the use of Bradford Factor acts as a
deterrent to employees who take leaves of absence for
nongenuine reasons. A number of organizations have
reported a reduction in absence when Bradford Factor
is introduced, which may be due to the use of the system
as a visible warning signal that voluntary absence is not
tolerated by the organization (IDS, 2007).
Although it is recommended that practitioners use the
Bradford Factor as a measure of absence, it should not
be used in isolation. Instead, practitioners should com-
bine it with employee consultation. This is because not
all employees with high Bradford scores are absent
because of nongenuine reasons. For instance, an
employee who has a child with a chronic illness may
need to take a series of 1-day absences during the year.
This will lead to a relatively high Bradford score, which
may trigger the organization’s absence review proce-
dures. If the review is conducted with the initial assump-
tion that the absences are illegitimate, then the employee
may feel unfairly judged by the organization and subse-
quently may become less engaged. It is therefore impor-
tant to suspend judgement until a formal meeting is held
with the employee.
Limitations
Aside from the measure of absenteeism used in the
current study, the independent and moderating variables
were measured at the same time. Although the study
hypotheses are based on a strong theoretical foundation,
alternative causal ordering is a possibility. Testing the
proposed model using a longitudinal research design
would help address this limitation. In addition, the sam-
ple used in the present study was drawn from one orga-
nization in the United Kingdom, which may limit the
generalizability of the findings. For instance, Lam,
Schaubroeck, and Aryee (2002) found that the negative
relationship between perceptions of the work environ-
ment and absenteeism was stronger at lower levels of
power distance. Likewise, it may be that our results are
more pronounced in lower power distance cultures.
Additionally, all variables aside from absenteeism
were derived from self-report measures, raising concerns
of common method variance. However, established
recommendations for controlling for the influence of
common method bias were followed, such as the use of
established scales, guaranteed anonymity and a clear
explanation of procedures (Podsakoff, MacKenzie,
Jeong-Yeon Lee, & Podsakoff, 2003). In addition, the
statistical analyses revealed that common method var-
iance did not wholly explain the associations in the data,
and the variables in the analyses were distinct from one
another.
The use of the Bradford Factor as a measure of
voluntary absence is not without its limitations. For
instance, the weighting of frequency versus duration in
the equation may not necessarily be accurate. Moreover,
the Bradford Factor (like frequency and duration mea-
sures of absence) does not indicate why an employee was
absent from work. Furthermore, an employee may be
genuinely ill periodically for short periods of time due
to chronic medical condition, such as migraines. Such an
employee will have a high Bradford Factor, but it is not
an indication of a lack of motivation.
An additional limitation of the present study is that
we did not assess a mediator of the moderated relation-
ships in order to conduct an empirical test of the theore-
tical explanation for our hypotheses. We encourage
future research to measure positive emotions as a poten-
tial mediator in order to tease out the underlying pro-
cesses of the moderating relationship between
engagement and job resources on outcomes.
Afinal limitation is that little variance in absence was
explained by the hypothesized models. However, the R
2
statistics in the present study are commensurate with
some prior research, notably Soane et al.’s(2013) study
of the relationship between meaningfulness, well-being
and engagement on absence duration. Given the compli-
cated nature of identifying moderating relationships in
field studies, even 1% of incremental variance explained
by an interaction is a significant finding (Evans, 1985;
McClelland & Judd, 1993). In our study, the interaction
term explained from 2% to 8% of incremental variance
demonstrating the strength of the interactive effect of job
resources and engagement on absence. Nonetheless, an
expanded research model should be developed to con-
tribute to a better understanding of the role of engage-
ment in explaining absence.
CONCLUSION
The present study examined the moderating role of job
resources on the relationship between engagement and
voluntary absenteeism, as measured by the Bradford
Factor. The results showed that organizational trust and
a high-quality LMX relationship moderated the relation-
ship between engagement and voluntary absence, such
that they buffered the negative relationship between
engagement and absence.
REFERENCES
Aguilera, R. V., Rupp, D. E., Williams, C. A., & Ganapathi, J. (2007).
Putting the S back in corporate social responsibility: A multilevel
theory of social change in organizations. Academy of Management
Review,32, 836–863. doi:10.5465/AMR.2007.25275678
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Newbury Park, CA: Sage.
540 SHANTZ AND ALFES

Arbuckle, J. L. (2006). AMOS (Version 7.0) [Computer program].
Chicago, IL: SPSS.
Bakker, A. B., & Bal, P. M. (2010). Weekly work engagement and
performance: A study among starting teachers. Journal of
Occupational and Organizational Psychology,83, 189–206.
doi:10.1348/096317909X402596
Bakker, A. B., & Demerouti, E. (2007). The job demands-resources
model: State of the art. Journal of Managerial Psychology,22,
309–328. doi:10.1108/02683940710733115
Bakker, A. B., Demerouti, E., De Boer, E., & Schaufeli, W. B. (2003).
Job demands and job resources as predictors of absence duration
and frequency. Journal of Vocational Behavior,62, 341–356.
doi:10.1016/S0001-8791(02)00030-1
Bakker, A. B., Hakanen, J. J., Demerouti, E., & Xanthopoulou, D.
(2007). Job resources boost work engagement, particularly when
job demands are high. Journal of Educational Psychology,99, 274–
284. doi:10.1037/0022-0663.99.2.274
Becker, T. E. (2005). Potential problems in the statistical control of
variables in organizational research: A qualitative analysis with
recommendations. Organizational Research Methods,8, 274–289.
doi:10.1177/1094428105278021
Bentler, P. M. (1990). Comparative fit indexes in structural models.
Psychological Bulletin,107, 238–246. doi:10.1037/0033-
2909.107.2.238
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and good-
ness-of-fit in the analysis of covariance structures. Psychological
Bulletin,88, 588–606. doi:10.1037/0033-2909.88.3.588
Blumberg, M., & Pringle, C. D. (1982). The missing opportunity in
organizational research: Some implications for a theory of work
performance. The Academy of Management Review,7, 560–569.
doi:10.5465/AMR.1982.4285240
Brooke Jr., P. P. (1986). Beyond the Steers and Rhodes model of
employee attendance. The Academy of Management Review,11,
345–361. doi:10.5465/AMR.1986.4283124
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing
model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural
equation models (pp. 136–162). Newbury Park, CA: Sage.
Caplan, G. (1974). Support systems and community mental health. New
York, NY: Behavioral Publications.
Chadwick-Jones, J. K., Nicholson, N., & Brown, C. (1982). Social
psychology of absenteeism. New York, NY: Praeger.
CIPD. (2010). Absence management. London: Chartered Institute of
Personnel Management.
Cohen, S., & Mckay, G. (1984). Social support, stress and the buffering
hypothesis: A theoretical analysis. In A. Baum, S. E. Taylor, & J. E.
Singer (Eds.), Handbook of psychology and health (pp. 253–267).
Hillsdale, NJ: Erlbaum.
Colquitt, J. A., Scott, B. A., & Lepine, J. A. (2007). Trust, trust-
worthiness, and trust propensity: A meta-analytic test of their
unique relationships with risk taking and job performance.
Journal of Applied Psychology,92, 909–927. doi:10.1037/0021-
9010.92.4.909
Conlin, M. (2007). Shirking hooky: The war on hooky. Business Week.
Retrieved from http://www.businessweek.com/stories/2007-11-11/
shirking-working-the-war-on-hooky
Cortina, J. M., Chen, G., & Dunlap, W. P. (2001). Testing interaction
effects in LISREL: Examination and illustration of available proce-
dures. Organizational Research Methods,4, 324–360. doi:10.1177/
109442810144002
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experi-
ence. New York, NY: Harper.
Darr, W., & Johns, G. (2008). Work strain, health, and absenteeism: A
meta-analysis. Journal of Occupational Health Psychology,13,
293–318. doi:10.1037/a0012639
Dawson, J. F. (2014). Moderation in management research: What, why,
when, and how. Journal of Business Psychology,29,1–19.
doi:10.1007/s10869-013-9308-7
Dekkers-Sánchez, P. M., Hoving, J. L., Sluiter, J. K., & Frings-Dresen,
M. H. W. (2008). Factors associated with long-term sick leave in
sick-listed employees: A systematic review. Occupational and
Environmental Medicine,65, 153–157. doi:10.1136/oem.2007.
034983
Demerouti, E., Bouwman, K., & Sanz-Vergel, A. I. (2011). Job
resources buffer the impact of work-family conflict on absenteeism
in female employees. Journal of Personnel Psychology,10, 166–
176. doi:10.1027/1866-5888/a000044
Demerouti, E., Nachreiner, F., Baker, A. B., & Schaufeli, W. B. (2001).
The job demands-resources model of burnout. Journal of Applied
Psychology,86, 499–512. doi:10.1037/0021-9010.86.3.499
Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange
model of leadership: A critique and further development. Academy
of Management Review,11, 618–634. doi:10.5465/AMR.1986.
4306242
Dirks, K. T., & Ferrin, D. L. (2001). The role of trust in organizational
settings. Organization Science,12, 450–467. doi:10.1287/orsc.12.4.
450.10640
Edwards, J. R., & Cable, D. M. (2009). The value of value congruence.
Journal of Applied Psychology,94, 654–677. doi:10.1037/
a0014891
Evans, M. G. (1985). A Monte-Carlo study of the effects of correlated
method variance in moderated multiple regression analysis.
Organizational Behavior and Human Decision Processes,36,
305–323. doi:10.1016/0749-5978(85)90002-0
Farrell, D., & Stamm, C. (1988). Meta-analysis of the correlates of
employee absence. Human Relations,41,211–227. doi:10.1177/
001872678804100302
Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th
ed.). London: Sage.
Fredrickson, B. L. (1998). What good are positive emotions? Review
of General Psychology,2,300–319. doi:10.1037/1089-
2680.2.3.300
Fredrickson, B. L. (2001). The role of positive emotions in positive
psychology: The broaden-and-build theory of positive emotions.
American Psychologist,56, 218–226. doi:10.1037/0003-
066X.56.3.218
Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden
the scope of attention and thought‐action repertoires. Cognition and
Emotion,19, 313–332. doi:10.1080/02699930441000238
French Jr., J. R. P., & Caplan, R. D. (1973). Organizational stress and
individual strain. In A. J. Murrow (Ed.), The failure of success (pp.
30–66). New York, NY: AMACOM.
Fried, Y., & Ferris, G. R. (1987). The validity of the job characteristics
model: A review and meta-analysis. Personnel Psychology,40,
287–322. doi:10.1111/j.1744-6570.1987.tb00605.x
Gillespie, N., & Dietz, G. (2009). Trust repair after an organization-
level failure. Academy of Management Review,34, 127–145.
doi:10.5465/AMR.2009.35713319
Gorgievski-Duijvesteijn, M. J., & Hobfoll, S. E. (2008). Work can burn
us out or fire us up: Conservation of resources in burnout and
engagement. In J. R. B. Halbesleben (Ed.), Handbook of stress
and burnout in healthcare (pp. 7–22). New York, NY: Nova
Science Publishers.
Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of
dyadic organizing. Research in Organizational Behavior,9,
175–208.
Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to
leadership: Development of leader-member exchange (LMX) the-
ory of leadership over 25 years: Applying a multi-level multi-
domain perspective. The Leadership Quarterly,6, 219–247.
doi:10.1016/1048-9843(95)90036-5
Hackett, R. D., & Guion, R. M. (1985). A reevaluation of the absentee-
ism–job satisfaction relationship. Organizational Behavior and
Human Decision Processes,35, 340–381. doi:10.1016/0749-5978
(85)90028-7
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 541

Hackman, J. R., & Oldham, G. R. (1975). Development of the job
diagnostic survey. Journal of Applied Psychology,60, 159–170.
doi:10.1037/h0076546
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the
design of work: Test of a theory. Organizational Behavior and
Human Performance,16, 250–279. doi:10.1016/0030-5073(76)
90016-7
Hackman J. R., & Oldham G. R. (1980). Work redesign. Reading, MA:
Addison-Wesley.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2009).
Multivariate data analysis (7th ed.). Upper Saddle River, NJ:
Pearson Prentice Hall.
Halbesleben, J. R. B., Harvey, J., & Bolino, M. C. (2009). Too
engaged? A conservation of resources view of the relationship
between work engagement and work interference with family.
Journal of Applied Psychology,94, 1452–1465. doi:10.1037/
a0017595
Harvey, J., & Nicholson, N. (1999). Minor illness as a legitimate reason
for absence. Journal of Organizational Behavior,20, 979–993.
doi:10.1002/(SICI)1099-1379(199911)20:6<979::AID-JOB914>
3.0.CO;2-0
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at
conceptualizing stress. American Psychologist,44, 513–524.
doi:10.1037/0003-066X.44.3.513
Hobfoll, S. E. (2011). Conservation of resource caravans and engaged
settings. Journal of Occupational and Organizational Psychology,
84,116–122. doi:10.1111/j.2044-8325.2010.02016.x
Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure
modelling: Sensitivity to underparameterized model misspecifica-
tion. Psychological Methods,3, 424–453. doi:10.1037/1082-
989X.3.4.424
IDS. (2007). Absence management (HR Study 842). London: Author.
Johns, G. (1997). Contemporary research on absence from work:
Correlates, causes and consequences. International Review of
Industrial and Organizational Psychology,12,115–173.
Johns, G., Xie, J. L., & Fang, Y. (1992). Mediating and moderating
effects in job design. Journal of Management,18, 657–676.
doi:10.1177/014920639201800404
Kahn, W. A. (1990). Psychological conditions of personal engagement
and disengagement at work. Academy of Management Journal,33,
692–724. doi:10.2307/256287
Lam, S. S. K., Schaubroeck, J., & Aryee, S. (2002). Relationship
between organizational justice and employee work outcomes: A
cross-national study. Journal of Organizational Behavior,23,
1–18. doi:10.1002/job.131
Levine, H. (1960). Robust tests for equality of variances. In I. Olkin, S.
G. Ghurye, W. Hoeffding, G. Madow, & H. B. Mann (Eds.),
Contributions to probability and statistics (pp. 278–292).
Stanford, CA: Stanford University Press.
Macey, W., & Schneider, B. (2008). The meaning of employee engage-
ment. Industrial and Organizational Psychology,1,3–30.
doi:10.1111/j.1754-9434.2007.0002.x
Martocchio, J. J. (1989). Age-related differences in employee absentee-
ism: A meta-analysis. Psychology and Aging,4, 409–414.
doi:10.1037/0882-7974.4.4.409
Mathieu, J. E., Tannenbaum, S. I., & Salas, E. (1992). Influences of
individual and situational characteristics on measures of training
effectiveness. Academy of Management Journal,35, 828–847.
doi:10.2307/256317
Mayfield, J., & Mayfield, M. (1998). Increasing worker outcomes by
improving leader follower relations. Journal of Leadership &
Organizational Studies,5,72–81. doi:10.1177/107179199800500108
Mcclelland, G. H., & Judd, C. M. (1993). Statistical difficulties of
detecting interactions and moderator effects. Psychological
Bulletin,114 , 376–390. doi:10.1037/0033-2909.114.2.376
McEvily, B., Perrone, V., & Zaheer, A. (2003). Trust as an organizing
principle. Organization Science,14,91–103. doi:10.1287/
orsc.14.1.91.12814
Miner, K. N., Settles, I. H., Pratt-Hyatt, J. S., & Brady, C. C. (2012).
Experiencing incivility in organizations: The buffering effects of
emotional and organizational support. Journal of Applied Social
Psychology,42, 340–372. doi:10.1111/j.1559-1816.2011.00891.x
Morgeson, F. P., & Humphrey, S. E. (2006). The work design ques-
tionnaire (WDQ): Developing and validating a comprehensive mea-
sure for assessing job design and the nature of work. Journal of
Applied Psychology,91, 1321–1339. doi:10.1037/0021-
9010.91.6.1321
Morris, J. H., Sherman, J. D., & Mansfield, E. R. (1986). Failures to
detect moderating effects with ordinary least squares-moderated
multiple regression: some reasons and a remedy. Psychological
Bulletin,99, 282–288. doi:10.1037/0033-2909.99.2.282
Nicholson, N., & Johns, G. (1985). The absence culture and the
psychological contract: Who’s in control of absence? Academy
of Management Review,10, 397–407. doi:10.5465/AMR.1985.
4278945
Podsakoff, P. M., Mackenzie, S. B., Jeong-Yeon, L., & Podsakoff, N. P.
(2003). Common method biases in behavioural research: A critical
review of the literature and recommended remedies. Journal of
Applied Psychology,88, 879. doi:10.1037/0021-9010.88.5.879
Robinson, S. L., & Rousseau, D. M. (1994). Violating the psychological
contract: Not the exception but the norm. Journal of Organizational
Behavior,15, 245–259. doi:10.1002/job.4030150306
Sagie, A. (1998). Employee absenteeism, organizational commitment,
and job satisfaction: Another look. Journal of Vocational Behavior,
52, 156–171. doi:10.1006/jvbe.1997.1581
Sagie, A., Birati, A., & Tziner, A. (2002). Assessing the costs of
behavioural and psychological withdrawal: A new model and an
empirical illustration. Applied Psychology: An International
Review,51,67–89. doi:10.1111/1464-0597.00079
Scandura, T. A., & Graen, G. B. (1984). Moderating effects of initial
leader-member exchange status on the effects of a leadership inter-
vention. Journal of Applied Psychology,69, 428–436. doi:10.1037/
0021-9010.69.3.428
Schaubroeck, J., & Ganster, D. C. (1993). Chronic demands and
responsivity to challenge. Journal of Applied Psychology,78,73–
85. doi:10.1037/0021-9010.78.1.73
Schaufeli, W. B., Bakker, A. B., & Van Rhenen, W. (2009). How
changes in job demands and resources predict burnout, work
engagement, and sickness absenteeism. Journal of Organizational
Behavior,30, 893–917. doi:10.1002/job.595
Schaufeli, W. B., Salanova, M., Gonzalez-Roma, V., & Bakker, A. B.
(2002). The measurement of engagement and burnout: A two
sample confirmatory factor analytic approach. Journal of
Happiness Studies,3,71–92. doi:10.1023/A:1015630930326
Scott, K. D., & Taylor, G. S. (1985). An examination of conflicting
findings on the relationship between job satisfaction and absentee-
ism: A meta-analysis. The Academy of Management Journal,28,
599–612. doi:10.2307/256116
Shapira-Lishchinsky, O., & Rosenblatt, Z. (2009). Perceptions of orga-
nizational ethics as predictors of work absence: A test of alternative
absence measures. Journal of Business Ethics,88, 717–734.
doi:10.1007/s10551-008-9977-8
Soane, E., Shantz, A., Alfes, A., Truss, C., Rees, C., & Gatenby, M.
(2013). The association of meaningfulness, wellbeing and engage-
ment with absenteeism: A moderated mediation model. Human
Resource Management,52, 441–456. doi:10.1002/hrm.21534
Stevens, J. P. (2002). Applied multivariate statistics for social sciences
(4th ed.). Hillsdale, NJ: Erlbaum.
Taylor, S. (2008). People resourcing (4th ed.). London: CIPD.
tenBrummelhuis,L.L.,Bakker,A.B.,&Euwema,M.C.(2010).Isfamily-
to-work interference related to co-workers’work outcomes? Journal of
Vocational Behavior,77, 461–469. doi:10.1016/j.jvb.2010.06.001
ten Brummelhuis, L. L., Ter Hoeven, C. L., Bakker, A. B., & Peper, B.
(2011). Breaking through the loss cycle of burnout: The role of
motivation. Journal of Occupational and Organizational
Psychology,84, 268–287. doi:10.1111/j.2044-8325.2011.02019.x
542 SHANTZ AND ALFES

ten Brummelhuis, L. L., Ter Hoeven, C. L., De Jong, M. D. T., &
Peper, B. (2013). Exploring the linkage between the home domain
and absence from work: health, motivation, or both? Journal of
Organizational Behavior,34, 273–290. doi:10.1002/job.1789
Towers Watson. (2012). Retrieved December 28, 2012, from http://
www.towerswatson.com/canada-english/research/7731
van Dierendonck, D., Le Blanc, P. M., & van Breukelen, W. (2002).
Supervisory behaviour, reciprocity and subordinate absenteeism.
Leadership & Organization Development Journal,23,84–92.
doi:10.1108/01437730210419215
Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organiza-
tional support and leader-member exchange: A social exchange
perspective. Academy of Management Journal,40,82–111.
doi:10.2307/257021
Whitener, E. M., Brodt, S. E., Korsgaard, M. A., & Werner, J. M.
(1998). Managers as initiators of trust: An exchange relationship
framework for understanding managerial trustworthy behaviour.
Academy of Management Review,23, 513–530. doi:10.5465/
AMR.1998.926624
Wollard, K. K. (2011). Quiet desperation: Another perspective on
employee engagement. Advances in Developing Human Resources,
13,526–537. doi:10.1177/1523422311430942
Wright, T. A. (2005). The role of happiness in organizational research:
Past, present and future directions. In P. L. Perrewe & D. C. Ganster
(Eds.), Exploring interpersonal dynamics: Research in occupa-
tional stress and wellbeing (Vol. 4, pp. 221–264). New York, NY:
Elsevier Science/JAI Press.
Xanthopoulou, D., Bakker, A. B., Demerouti, E., & Schaufeli, W. B.
(2009). Reciprocal relationships between job resources, personal
resources, and work engagement. Journal of Vocational Behavior,
74(3), 235–244. doi:10.1016/j.jvb.2008.11.003
Xanthopoulou, D., Bakker, A. B., Dollard, M. F., Demerouti, E.,
Schaufeli, W. B., Taris, T. W., & Schreurs, P. J. G. (2007). When
do job demands particularly predict burnout? The moderating role
of job resources. Journal of Managerial Psychology,22, 766–786.
doi:10.1108/02683940710837714
Xie, J. L., & Johns, G. (1995). Job scope and stress: Can job scope be
too high? Academy of Management Journal,38, 1288–1309.
doi:10.2307/256858
Yukl, G., O’Donnell, M., & Taber, T. (2009). Influence of leader
behaviours on the leader-member exchange relationship.
Journal of Managerial Psychology,24,289–299. doi:10.1108/
02683940910952697
Original manuscript received January 2013
Revised manuscript received May 2014
Revised manuscript accepted June 2014
First published online July 2014
ENGAGEMENT VOLUNTARY ABSENCE JOB RESOURCES 543