This article was downloaded by: [University of Guelph]

On: 13 March 2015, At: 07:16

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

European Journal of Work and Organizational

Psychology
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/pewo20

Job crafting and job performance: A longitudinal study

a a a
Maria Tims , Arnold B. Bakker & Daantje Derks
a
Department of Work and Organizational Psychology, Institute of Psychology, Erasmus
University Rotterdam, Rotterdam, The Netherlands
Published online: 03 Nov 2014.

Click for updates

To cite this article: Maria Tims, Arnold B. Bakker & Daantje Derks (2014): Job crafting and job performance: A longitudinal
study, European Journal of Work and Organizational Psychology, DOI: 10.1080/1359432X.2014.969245

To link to this article: http://dx.doi.org/10.1080/1359432X.2014.969245

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 European Journal of Work and Organizational Psychology, 2014
http://dx.doi.org/10.1080/1359432X.2014.969245

Job crafting and job performance: A longitudinal study

Maria Tims, Arnold B. Bakker, and Daantje Derks

Department of Work and Organizational Psychology, Institute of Psychology, Erasmus University Rotterdam,
Rotterdam, The Netherlands

In this three-wave study (N = 288), we examined whether job crafting intentions and work engagement led to actual job
crafting behaviours and, in turn, to higher levels of prospective work engagement and job performance. We used the Job
Demands-Resources model as a theoretical framework and defined job crafting as the self-initiated changes that employees
Downloaded by [University of Guelph] at 07:16 13 March 2015

make in their job demands and resources. One month after reporting their job crafting intentions, respondents rated their actual
job crafting behaviours. Again one month later, they rated their levels of work engagement, in-role performance, and
organizational citizenship behaviour towards individuals (OCBI). Results of structural equation modelling showed that job
crafting intentions and work engagement significantly related to actual job crafting, which, in turn, related to higher levels of
work engagement, while controlling for job characteristics. Results further showed that engaged employees performed better on
their in-role tasks but did not perform more OCBIs. The findings suggest that employees can increase their own work
engagement and job performance through job crafting.

Keywords: Employee engagement; Intentions; Job crafting; Job performance; Organizational citizenship behaviours; Work
engagement.

Organizations benefit from employees who are engaged finding the job so involving that one forgets the time
in their work. Compared to less engaged employees, while working; Schaufeli, Salanova, González-Roma, &
engaged employees report the highest levels of job per- Bakker, 2002).
formance (Bakker & Bal, 2010; Christian, Garza, & Although employers should strive to create jobs with
Slaughter, 2011) and also receive the highest perfor- motivating potential (Hackman & Oldham, 1976),
mance ratings from their supervisors and clients employees also need to take responsibility for their
(Bakker, Demerouti, & Ten Brummelhuis, 2012; own work-related well-being. Wrzesniewski and Dutton
Halbesleben & Wheeler, 2008; Salanova, Agut, & (2001) have argued that employees can and do shape the
Peiró, 2005). In addition, engaged workers deliver boundaries of their jobs and create a work environment
more than what is prescribed by their formal tasks that fits their preferences, skills, and competences. This
(extra-role behaviours; Bakker, Demerouti, & Verbeke, process of proactively influencing one’s job is called job
2004; Christian et al., 2011). These positive outcomes of crafting. Employee involvement in the process of creat-
work engagement for both the organization and the ing a good work environment may be crucial for at least
individual make it meaningful to understand how work two reasons. The first reason is that all employees have
engagement can be advanced. Studies have shown that idiosyncratic characteristics that may result in different
employees working in a resourceful job are most likely preferences for job characteristics or career choices
to be engaged with their work (for meta-analyses, see (Berings, De Fruyt, & Bouwen, 2004). For example, an
Crawford, LePine, & Rich, 2010; Halbesleben, 2010). individual who works best under deadlines will benefit
For example, employees with decision latitude in how more from high time pressure than a person who plans
and when to work, who can use various skills, and can his/her tasks carefully ahead of time. Individuals have
count on colleagues and others are most likely to experi- this knowledge readily accessible, whereas a manager
ence this positive fulfilling state of mind that is charac- may not have the time or resources to find out the
terized by vigour (i.e., feeling fit and energetic while preferences of every employee. A second reason for the
working), dedication (i.e., feeling enthusiastic about the importance of an individual’s involvement in the design
work and involved in the job tasks), and absorption (i.e., of a job is that once employees know how they can

Correspondence should be addressed to Maria Tims, Faculty of Economics and Business Administration, Department of Management and
Organization, VU University Amsterdam, The Netherlands. E-mail: m.tims@vu.nl

© 2014 Taylor & Francis

 2 TIMS, BAKKER, DERKS

create their optimal work environment, they can monitor R) theory (Bakker & Demerouti, 2007; Demerouti,
their job characteristics and intervene when necessary to Bakker, Nachreiner, & Schaufeli, 2001). The advantage
prevent negative outcomes like demotivation and dete- of using the JD-R model to operationalize job crafting is
riorated performance. that it allows us to study job crafting at a more general
Whereas job design scholars such as Kulik, Oldham, level. This general approach is important as job crafting
and Hackman (1987) already suggested that employees behaviours may take many different forms (e.g.,
may redesign their jobs on their own initiative, job Wrzesniewski, 2003), making it difficult to study job
crafting only recently gained attention of work and orga- crafting at the content level. Using JD-R theory that
nizational scholars (Ghitulescu, 2006; Wrzesniewski & focuses on two general classes of job characteristics
Dutton, 2001). Job crafting may be an interesting strat- that will be present in each job and organization
egy used by employees to stay or become engaged in (Bakker & Demerouti, 2007) will improve the knowl-
their work and as a consequence to remain valuable for edge about and understanding of job crafting.
the organization. To further knowledge about job craft- The first class of job characteristics distinguished in
ing, the aim of the present study is to longitudinally JD-R theory is labelled “job demands” and refers to
examine predictors of job crafting (i.e., job crafting those aspects of the job that require sustained physical
intentions and work engagement) as well as the relation- and/or psychological effort or skills (Demerouti et al.,
ship of job crafting with prospective work engagement 2001). Therefore, job demands are associated with cer-
Downloaded by [University of Guelph] at 07:16 13 March 2015

and job performance. We are aware of only a few studies tain physiological or psychological costs. Examples of
that empirically examined the relationship between job job demands are a heavy workload and emotionally
crafting, work engagement, and/or job performance (e.g., demanding interactions with customers. Building on
Bakker, Tims, & Derks, 2012; Leana, Appelbaum, & inconsistent findings regarding relationships between
Shevchuk, 2009; Tims, Bakker, Derks, & Van Rhenen, job demands and work outcomes, Cavanaugh, Boswell,
2013) and these were all cross-sectional studies. For Roehling, and Boudreau (2000) showed that some job
example, job crafting and work engagement are related demands (e.g., role ambiguity, insecurity) were indeed
to each other (Tims, Bakker, & Derks, 2012), but the related to negative work outcomes, such as turnover but
causal order has not been established yet. Some authors that other job demands (e.g., time pressure, number of
argue that job crafting may lead to work engagement projects/assignments) were related to positive work out-
(e.g., Petrou, Demerouti, Peeters, Schaufeli, & Hetland, comes, such as job satisfaction. Thus, depending on the
2012), whereas others argue that engaged workers may demand that was studied, there was a positive or nega-
be more likely to show initiative and be proactive (e.g., tive relationship between job demands and work out-
Hakanen, Perhoniemi, & Toppinen-Tanner, 2008). comes. Job demands that require increased effort of the
Furthermore, Bakker (2011) and Bakker and Demerouti employee and may lead to job satisfaction are called
(2014) suggest that job crafting and work engagement challenging job demands (LePine, Podsakoff, &
are reciprocally related. Moreover, most studies only LePine, 2005), whereas job demands that require
focused on in-role performance whereas extra-role per- increased effort and are related to negative work out-
formance or organizational citizenship behaviour (OCB) comes are called hindering job demands. These job
is also an important aspect of employee performance that demands hinder the employee in achieving valued
relates to helping colleagues and orienting new employ- goals (Cavanaugh et al., 2000).
ees (Williams & Anderson, 1991). The second class of job characteristics is called “job
resources” (Demerouti et al., 2001). Job resources refer
to those aspects of the job that are either/or functional in
JOB CRAFTING
achieving work goals, reducing job demands and the
According to Wrzesniewski and Dutton (2001), job associated physiological and psychological costs, and
crafting can take three different forms: (1) employees stimulating personal growth, learning, and development
can increase or decrease the number of tasks they need (Bakker & Demerouti, 2007). Examples of job resources
to perform or change how they perform their tasks (i.e., are autonomy, performance feedback, skill variety, and
crafting the task boundaries); (2) employees can change social support. High levels of job resources available at
the number and intensity of interactions with others both work lead to work engagement (Bakker & Demerouti,
on the job and outside of the organization (i.e., crafting 2007; Schaufeli & Bakker, 2004).
the relational boundaries); and (3) employees can change Within this theoretical framework, job crafting can be
the meaning of their job by reformulating the impact described as the changes employees make in their job
their work has on themselves or others (i.e., crafting demands and job resources (cf. Tims & Bakker, 2010;
the cognitions about the job). The present research Tims et al., 2012). More specifically, based on a large
takes a more general approach to job crafting, which sample (N = 1,118), Tims and her colleagues validated
focuses on crafting the job characteristics instead of four job crafting dimensions, namely (1) increasing
crafting the job content. More specifically, job crafting structural job resources (i.e., crafting more autonomy,
is conceptualized using the Job Demands-Resources (JD- variety, opportunities for development), (2) increasing
 JOB CRAFTING 3

social job resources (i.e., crafting more social support, Intentions to engage in job crafting were assessed in
feedback, coaching), (3) increasing challenging job the first survey, and based on the proactive nature of job
demands (i.e., crafting involvement in new projects), crafting, we expect that job crafting intentions predict
and (4) decreasing hindering job demands (i.e., crafting actual job crafting behaviour in the next month.
fewer emotional and cognitive demands). Research
showed that employees naturally engage in all four Hypothesis 1a: Job crafting intentions predict
types of job crafting (Tims et al., 2012), indicating that actual job crafting (all four dimensions).
it is relevant to study these types of job crafting in more
detail. But when are employees most likely to engage in
job crafting?
Work engagement
Next to behavioural intentions to engage in job crafting,
motivational aspects may also play an important role in
PREDICTORS OF JOB CRAFTING predicting job crafting. Based on the job design litera-
Job crafting intentions ture, it is known that the work environment influences
how employees feel and behave at work (Hackman &
It is often argued that the most immediate and important
Oldham, 1976). As job crafting is about proactively
predictor of behaviour is the intention to perform that
Downloaded by [University of Guelph] at 07:16 13 March 2015

changing the job characteristics, it has been argued that

behaviour (Ajzen, 1991; Sheeran, 2002). Intentions are
this may lead employees to feel better at their work.
described as self-instructions to perform a specific beha- Namely, when employees successfully create a work
viour or to obtain specified outcomes (Triandis, 1980).
environment characterized by job resources and challen-
People who hold strong intentions towards specific beha-
ging job demands, their work engagement and job satis-
viours are more likely to perform those behaviours
faction may increase (Tims, Bakker, & Derks, 2013). At
because intentions comprise the motivational factors
the same time, it is likely that employees who feel
that influence a behaviour (e.g., effort, planning, 2012;
motivated and enthusiastic will be more likely to be
Ajzen, 1991). The strongest link between intention and
proactive. JD-R theory suggests that job resources and
behaviour will be found when the behaviour is voli-
job challenges are related to work engagement and that
tional, and opportunities and resources to perform the engaged employees mobilize their own job resources and
behaviour are available (Webb & Sheeran, 2006). As job
job challenges which may, in turn, foster engagement
crafting is proactive and discretionary behaviour that
over time (Bakker, 2011). As such, a positive gain spiral
employees spontaneously engage in, it is reasonable to
may emerge over time (Bakker, 2011). Research has
expect that employees with job crafting intentions will
shown that positive emotions that accompany work
also convert these intentions into actions. More specifi-
engagement (i.e., joy and enthusiasm) may make
cally, proactive behaviour reflects self-initiation of
employees more open to their environment
change, anticipatory action, and taking control over
(Fredrickson, 2000) and more likely to take initiative at
one’s work environment (Parker & Collins, 2010). work (Salanova & Schaufeli, 2008; Sonnentag, 2003).
These characteristics indicate that especially these
Engaged workers who care about their work, have
employees who have the intention to craft their job
resources to spend, and are concentrated on their work
demands and/or job resources are likely to take action
likely see value in proactively improving their job char-
because job crafting refers to autonomous, self-con-
acteristics (Sonnentag, 2003). Interestingly, Warr and
trolled behaviour.
Inceoglu (2012) argued that engaged employees are con-
In the proactivity literature, the intention to make
stantly on the lookout for more job characteristics
changes at work may reside in the “reason to” motiva-
because their greater motivation means that they want
tional state (Parker, Bindl, & Strauss, 2010). The “reason more from their job, for example, in terms of opportu-
to” motivation reflects that people must see value in
nities for personal influence (i.e., autonomy) and chal-
being proactive and changing job characteristics. For
lenging workload. Crafting the job may thus be a means
example, Parker and colleagues state that individuals
to realize their needs.
may be more likely to strive for proactive goals when
Hakanen, Perhoniemi, et al. (2008) found that work
they want to increase their challenges at work as chal-
engagement and personal initiative were positively and
lenges may fulfil the needs for competence and auton-
reciprocally related with each other over time, supporting
omy (Deci & Ryan, 2000). Wrzesniewski and Dutton
a gain spiral. Recently, Lu, Wang, Lu, Du, and Bakker
(2001) also refer to job crafting as a means to take back (2014) found that engaged employees crafted their phy-
control over work. When employees have the intention
sical and relational boundaries more often than their less
to change specific job characteristics, we expect they are
engaged counterparts in a period of three months. The
more likely to actually change them because they believe
present study therefore incorporates a relationship
it will make their experience of the job better and they
between work engagement assessed at the first measure-
perceive to have control over their job.
ment occasion and T2 job crafting. However, we do not
 4 TIMS, BAKKER, DERKS

expect that T1 work engagement predicts T2 crafting of challenging job demands, reported higher levels of
hindering job demands because engaged workers are work engagement than employees who did not craft
more likely to see hindering demands as job challenges these job characteristics. In addition, a quantitative
or to have sufficient levels of job resources available to diary study showed that employees who crafted their
deal with these demands. In support of this reasoning, challenging job demands during the day were more
Petrou et al. (2012) found that on days in which employ- likely to feel vigorous, dedicated, and absorbed on
ees experienced high work pressure and high job auton- that day (Petrou et al., 2012). These findings indicate
omy (“active jobs; Karasek, 1979), they were least likely that employees who are able to mobilize their own job
to decrease their hindering job demands. Furthermore, as resources and challenging job demands are more likely
research has shown that employees with active jobs are to experience their work as a psychologically fulfilling
more likely to be engaged (Bakker, Hakanen, Demerouti, activity.
& Xanthopoulou, 2007), they will be least likely to In contrast to the positive relationships reported
decrease their hindering job demands. between increasing job resources and increasing challen-
ging job demands and work engagement, research has
Hypothesis 1b: T1 work engagement predicts T2 shown that the job crafting dimension of decreasing hin-
crafting of job resources and challenging job dering job demands showed another pattern of relation-
demands. ships with work engagement. Namely, Petrou et al. (2012)
Downloaded by [University of Guelph] at 07:16 13 March 2015

found that reducing hindering job demands related nega-

tively to work engagement at the day level. Similarly,
Tims, Bakker, Derks, and Van Rhenen (2013) reported a
THE RELATIONSHIP BETWEEN JOB
negative relationship between decreasing hindering job
CRAFTING, WORK ENGAGEMENT, AND
demands and vigour, which reflects the energy aspect of
JOB PERFORMANCE
work engagement. In other words, the expected positive
Employees who craft their jobs and create a work relationship of this type of job crafting with motivational
environment with sufficient job resources and challen- outcomes is not present. In contrast, it seems more likely
ging job demands but without hindering job demands that decreasing hindering job demands is negatively
are expected to feel engaged with their job tasks (Tims related to work engagement. The meta-analyses of
& Bakker, 2010). This expectation is based on the idea Crawford et al. (2010) and Nahrgang, Morgeson, and
that job resources are instrumental in achieving work Hofmann (2011) showed that hindering job demands
goals (Bakker & Demerouti, 2007) and enable employ- may be more likely to affect work engagement negatively.
ees to learn new things, to personally grow, and to However, as job crafting is about changing job demands,
develop themselves (Bakker, Demerouti, & Sanz- this relationship needs further exploration in order to
Vergel, 2014). In this way, job resources are related to understand the process that underlies this negative rela-
autonomous forms of motivation, such as intrinsic moti- tionship. Based on the JD-R model (Demerouti et al.,
vation and identified motivation (Deci & Ryan, 2000). 2001), in which job demands are stronger related to
Baard, Deci, and Ryan (2004) demonstrated that aspects energy depletion and negative health outcomes such as
of the environment that satisfy peoples’ need for auton- burnout than to positive outcomes (Hakanen, Schaufeli,
omy, relatedness, and competence, such as decisional et al., 2008), it may be more likely that decreasing hinder-
control, social support, and developmental opportu- ing job demands is related to a decrease in exhaustion or
nities, likely yield positive outcomes, such as motiva- stress because people crafted a work environment without
tion. Similarly, Deci et al. (2001) showed that obstacles such that it may free up energy. Taking these
satisfaction of the psychological needs of autonomy, findings into account, we formulate the following
competence, and relatedness was positively related to hypothesis:
work engagement.
Preliminary support for the relationship between job Hypothesis 2a: Increasing structural and social job
crafting, in which employees themselves change their resources and increasing challenging job demands at
job characteristics, and work engagement comes from T2 is positively related to work engagement at T3.
the study of Tims and colleagues (2012). These authors More specifically, increasing structural and social
found that employees who reported that they crafted job resources and increasing challenging job
their job resources and challenging job demands were demands at T2 mediate the relationship between
rated by their colleagues as more engaged than employ- T1 job crafting intentions and T3 work engagement.
ees who did not craft their job resources and challen- Hypothesis 2b: Decreasing hindering job demands
ging job demands. Bakker, Tims, et al. (2012) also at T2 is negatively related to work engagement at
found a positive relationship between job crafting and T3. More specifically, decreasing hindering job
work engagement. More specifically, their results demands at T2 mediates the relationship between
showed that employees who proactively increased T1 job crafting intentions and T3 work
their structural and social job resources, and their engagement.
 JOB CRAFTING 5

The relationship between job crafting and job perfor- Konovsky, 1989). In line with our focus on volitional
mance has been examined in several studies. From behaviours (and not compliance behaviours), we
qualitative interview studies (Berg et al., 2010; included OCBI in this study.
Lyons, 2008) as well as quantitative survey studies Indeed, Bakker, Tims, et al. (2012) reported that job
(Bakker, Tims, et al., 2012; Leana et al., 2009; Tims crafting facilitates work engagement and indirectly in-
et al., 2012), it appears that job crafting mainly has a role performance. However, these authors only focused
positive relationship with job performance. For exam- on increasing job resources (both structural and social
ple, Leana et al. found that teachers who crafted their job resources) and challenging job demands. In a study
work received higher observer ratings of quality of that incorporated all four job crafting dimensions, the
care. Moreover, Berg, Dutton, and Wrzesniewski same three job crafting dimensions related positively to
(2008) summarized four lines of literature about job job performance via work engagement (Tims, Bakker,
crafting and concluded that job crafting leads to posi- Derks, & Van Rhenen, 2013). However, it was again
tive experiences such as achievement, enjoyment, and found that decreasing hindering job demands showed a
meaning; to resilience, as exemplified by increased negative relationship with job performance via work
competence, personal growth, and ability to cope engagement. In other words, employees who decreased
with future adversity; and to a changed work meaning their hindering job demands reported significantly lower
and work identity. Job crafting may relate positively to levels of work engagement and job performance com-
Downloaded by [University of Guelph] at 07:16 13 March 2015

job performance because employees make changes to pared to employees who did not decrease their hindering
their jobs to enable better performance or be more job demands. We predict that:
efficient but also to be able to do tasks they find
interesting or rewarding. Job crafters may thus direct Hypothesis 3: Work engagement (T3) mediates the
their energy to change job characteristics to achieve relationship between job crafting (T2) and in-role
goals they value or goals they believe to lead to performance and OCBI (T3).
rewarding outcomes (Warr & Inceoglu, 2012).
Based on the arguments that crafting one’s job
demands and job resources may lead to work engage-
ment and job performance, we expect that work engage- METHOD
ment may mediate the relationship between job crafting
Participants and procedure
and job performance. Namely, work engagement seems
to influence job performance too. The relationship This study addresses a unique topic by making limited
between work engagement and job performance has use of data that have been reported in Tims, Bakker, and
been theorized to exist because engaged individuals Derks (2013). Nevertheless, the focus of the two papers
approach their tasks with a sense of self-investment is different. Tims et al. investigated whether job crafting
(i.e., dedication) and energy which should lead to is related to a change in the job demands and job
higher levels of in-role and extra-role performance resources and whether the change in these job character-
(Christian et al., 2011; Rich, LePine, & Crawford, istics mediated the relationship between job crafting and
2010). In-role performance likely improves because a change in well-being. In contrast, the current article
employees are invested in their tasks and will persist adds to the literature by focusing on the predictive value
to perform their tasks effectively. In addition, based on of job crafting intentions and work engagement on job
social exchange theory, it is expected that employees crafting behaviour and of job crafting behaviour on job
with resources are willing to invest these resources in performance over time. Thus, the first study explains the
performing their tasks (Saks, 2006). Extra-role perfor- mechanism through which job crafting may influence
mance or OCB will occur more often among engaged employee well-being (i.e., via a change in job character-
employees because they are concerned about facilitating istics), whereas the present study focuses on how job
the organization and the people within the organization crafting intentions relate to job crafting, how work
(Rich et al., 2010). Regarding OCB, this study focused engagement and job crafting strengthen each other over
on OCB behaviours that benefit individuals and thus time, and we include job performance as an outcome
indirectly contribute to the organization, such as helping measure.
others who have been absent. This type of OCB is Participants were recruited at a chemical plant in The
referred to as OCB at the individual level (i.e., OCBI), Netherlands, which produces plastics for several indus-
whereas behaviours that benefit the organization in gen- tries, such as the automotive, health care, and electro-
eral (e.g., giving advance notice when unable to come nics industry (see also Tims, Bakker, & Derks, 2013).
to work) are referred to as OCB at the organizational The organization was interested in the job crafting and
level (OCBO; Williams & Anderson, 1991). work engagement of their employees. Therefore, all
Furthermore, OCB towards individuals (OCBI) has employees (N ≈ 1,250) were invited to participate in
been labelled “altruism”, whereas OCBO has been the study via a message that was posted on their intra-
labelled “generalized compliance” (cf. Organ & net and via information posters throughout the
 6 TIMS, BAKKER, DERKS

organization. Every employee received an e-mail that Time 2 measure

described the aims of the study, emphasized confidenti-
Actual job crafting was measured with an adapted ver-
ality, and contained a personal login code. The e-mail
sion of the 21-item scale described earlier (Tims et al.,
also provided the link to the survey. The study had a
2012). We adapted the items such that they referred to
three-wave design, with one month in between the
job crafting in the preceding month. The item stem was
waves. Because job crafting is found to occur on a
“Last month, …” and then the item was presented. For
daily basis (Petrou et al., 2012), we expect that employ-
example, the dimension increasing structural job
ees should have had at least some opportunities to craft
resources was assessed with items such as “Last month,
their jobs during the time period of one month. In
I tried to learn new things at work” (α = .78). Cronbach’s
addition, intentions correlate more closely with beha-
alphas for the dimensions increasing social job
viour when the time interval between measurement of
resources, increasing challenging job demands, and
the intention and behaviour is shorter (Davidson &
decreasing hindering job demands, were .82, .81, and
Jaccard, 1979).
.81, respectively. Responses were given on a 5-point
At Time 1, the survey was completed by 564 (45.1%)
frequency scale that ranged from 1 (never) to (very
employees. At Time 2, the number of participants was
often).
468 (37.4%), and at Time 3, the survey was completed
by 477 (38.1%) employees. Due to study drop-out, not
Downloaded by [University of Guelph] at 07:16 13 March 2015

all participants could be matched to earlier measurement Time 3 measure

occasions, which resulted in a final sample of 288 parti-
cipants who completed the survey at all three occasions Work engagement was measured with the 9-item version
(response rate 23%). In line with the general distribution of the Utrecht Work Engagement Scale (UWES;
of gender within the organization, participants were Schaufeli, Bakker, & Salanova, 2006). The UWES
mainly male (82.6%), with a mean age of 45.19 years items reflect three underlying dimensions, measured
(SD = 8.71), and a mean tenure of 18.31 years with three items each: vigour (α = .93; e.g., “At my
(SD = 9.95). On average, participants worked 39.15 hr work, I feel bursting with energy”), dedication
a week (SD = 6.81). Employees had primary/secondary (α = .95; e.g., “I am enthusiastic about my job”), and
education (15%), vocational education (40%), or higher absorption (α = .85; e.g., “I get carried away when I am
professional/scientific education (45%). Comparisons of working”). A 7-point scale was used with response cate-
those who dropped out and those who completed all gories ranging from 0 (never) to 6 (always).
three surveys revealed no significant differences on the In-role performance and OCBI were each measured
demographic variables reported earlier and on the study with seven items developed by Williams and Anderson
variables. We therefore concluded that the dropouts were (1991). The scale was translated to Dutch and back-
comparable with the sample group and that no selective translated to English to check the validity of the transla-
dropout had occurred. tion. An example item that measures in-role performance
is “I adequately complete assigned duties” and of OCBI
is “I help others who have been absent”. A 5-point scale
was used with response categories ranging from 1
Time 1 measure (totally disagree) to 5 (totally agree). Cronbach’s α of
Job crafting intentions were measured with a recently the in-role performance measure was .86 and of the
validated job crafting scale (Tims et al., 2012) that was OCBI measure it was .84. Since the scale was translated
adapted such that the items referred to the intention to from English to Dutch, we examined the factor structure
engage in job crafting. The item stem was “For the using confirmatory factor analysis. After correlating the
upcoming period, I intend to …” and then the item was error terms of two in-role performance items that were
presented. An example item that reflects “Intention to reverse-scored (Floyd & Widaman, 1995), the fit of the
increase structural job resources” (five items, α = .81) is two-factor model, in which in-role performance and
“… make sure to use my capacities to the fullest”. An OCBI were distinguished, was adequate (χ2 = 203.35,
example item for “Intention to increase social job df = 75, comparative fit index (CFI) = .93, Tucker–Lewis
resources” (five items, α = .83) is “…ask colleagues for index (TLI) = .91, root mean square error of approxima-
advice”. “Intention to increase challenging job demands” tion (RMSEA) = .08) and better than a one-factor model,
was measured with items such as “… start new projects in which the items for in-role performance and OCBI
when there is not much to do at work” (five items, loaded on one factor (Δχ2 = 338.89, Δdf = 1, p < .01).
α = .80). Finally, “Intention to decrease hindering job
demands” (six items, α = .87) was measured with items
Control variables
such as “…ensure that my work is emotionally less
intense”. Employees indicated to what extent they agreed In the analyses, we control for T1 work engagement, in-
with each statement with a response scale ranging from 1 role performance, and OCBI. These variables were
(totally disagree) to 5 (totally agree). assessed with the scales described earlier. The reliability
 JOB CRAFTING 7

of the subscales measuring work engagement was .92, addition to model fit, we used bootstrapping to test the
.95, and .83 for vigour, dedication, and absorption, significance of the hypothesized indirect effects. The
respectively, at T1. Cronbach’s α of in-role performance bootstrap procedure estimates more accurate confidence
at T1 was .88 and of OCBI it was .83. intervals of indirect effects because it resamples from the
In addition, we control for the effect of T1 job data set and estimates the indirect effect in every new
resources and job demands on T3 work engagement as sample. Also, it does not impose the assumption of
they are important predictors of work engagement normality upon the sampling distribution (Preacher &
(Halbesleben, 2010). The following job resources mea- Hayes, 2008), which is important because the sampling
sured at T1 were included: autonomy (three items, distribution of the indirect effect is skewed relative to a
α = .84; Bakker, Demerouti, Taris, Schaufeli, & normal distribution.
Schreurs, 2003), social support (three items, α = .79;
Bakker et al., 2003), supervisory coaching (five items,
α = .92; Le Blanc, 1994), and feedback (three items, RESULTS
α = .85; Bakker et al., 2003). With regard to job Descriptive statistics
demands, the following demands measured at T1 were
included: emotional demands (three items, α = 63; Van The correlations among the study variables are displayed
Veldhoven & Meijman, 1994), cognitive demands (four in Table 1. Before testing the hypothesized model, we
items, α = .84; Van Veldhoven & Meijman, 1994), and examined the longitudinal invariance of our measures
Downloaded by [University of Guelph] at 07:16 13 March 2015

workload (three items, α = .94; Furda, 1995). across measurement times (Horn & McArdle, 1992)
such that solid conclusions can be drawn based on our
longitudinal data.
Analysis strategy
The hypothesized model was examined using structural
Longitudinal invariance
equation modelling in AMOS (Arbuckle, 2005). We
modelled the variables as latent variables with the scale Three scales were used at T1 and T3, namely work
means or parcels as the indicators given the relatively engagement, in-role performance, and OCBI. The invar-
small sample size compared to the number of items iance test requires analysing data by fitting the two
(Marsh, Hau, Balla, & Grayson, 1998). For example, waves of data with two separate models simultaneously.
for work engagement, we used the mean scale scores Therefore, two separate factor models were constructed
(i.e., vigour, dedication, absorption) as the indicators of for each of the T1 and T3 factors, resulting in six latent
the latent construct, and for the latent constructs job variables. In addition, correlations were estimated among
performance and OCBI, parcels were created. A parcel all possible pairs of uniqueness between T1 and T3
can be defined as an aggregate-level indicator compris- because the same items were used across two time points
ing the average of two or more items. The psychometric (Pitts, West, & Tein, 1996). The invariance routine
advantage of parcelling is that parcels result in more involves testing and comparing models that impose suc-
reliable measurement models (Little, Cunningham, cessive restrictions on model parameters. A significant
Shahar, & Widaman, 2002). Parcels were created based chi-square difference between models indicates that the
on random assignment of items to one of the two parcels equality constraints are not consistent with the data
(see Little et al., 2002). In addition, the reliability of the (hence not invariant across time). Model A freely esti-
parcels was checked and all reliability estimates were mated all the parameters across the two measurement
greater than .71. times, which is the unrestricted model. Next, factor load-
Based on previous research, job crafting was mod- ings were constrained to be equal across measurement
elled with two latent factors. The first latent factor com- times in Model B. In Model C, factor loadings and factor
prised the mean scores on the dimensions Intention to variances are constrained to be equal in both measure-
increase structural job resources, Intention to increase ment times, and in Model D, factor loadings, factor
social job resources, and Intention to increase challen- variances, and factor uniqueness were constrained to be
ging job demands. The second latent factor contained equal across the two measurement times.
two parcels of Intention to decrease hindering job As indicated in Table 2, the results largely support
demands that were formed on the basis of similarity longitudinal metric invariance of our measures. The first
(Little et al., 2002). With regard to actual job crafting, three models do not differ significantly from each other,
the same procedure was followed. indicating that the factor loadings and factor variances
Model fit was assessed with the chi-square statistic are invariant across the two measurement times. Model
(χ2), the RMSEA, the standardized root mean square D, however, shows that the factor uniquenesses are not
residual (SRMR), the CFI, and the TLI. The conven- invariant, as this model showed a significant chi-square
tional cut-off values of these fit indices were used to change from the previous model. However, this is a very
assess model fit (i.e., CFI, TLI, IFI > .90, RMSEA < .08 stringent test (Byrne, 2004), and invariance of the factor
to indicate good fit; Marsh, Hau, & Wen, 2004). In loadings is usually the only form of invariance tested in
 Downloaded by [University of Guelph] at 07:16 13 March 2015

8 TIMS, BAKKER, DERKS

TABLE 1
Correlations between study variables (N = 288)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

1 Int. to incr. Struc JR T1 –

2 Int. to incr. Soc JR T1 .69** –
3 Int. to incr. Chall JD T1 .74** .62** –
4 Int. to decr. Hind JD T1 .30** .42** .21** –
5 Work engagement T1 .32** .25** .34** −.04 –
6 Autonomy T1 .28** .27** .31** .16** .56** –
7 Social support T1 .20** .12* .29** −.07 .36** .42** –
8 Coaching T1 .21** .29** .32** −.09 .43** .47** .57** –
9 Feedback T1 .24** .17** .33** .12* .51** .50** .50** .64** –
10 Emotional demands T1 .00 09 −.04 .02 −.01 −.03 −.24** −.16** −.10 –
11 Cognitive demands T1 .11 .12* .10 .07 .31** .22** .02 .12* .14* .31** –
12 Workload T1 .00 .07 −.06 .05 .23** .09 −.09 −.01 .03 .44** .58** –
13 In-role performance T1 .11 .03 .19** −.11 .31** .28** .30** .24** .29** −.09 .25** .05 –
14 OCBI T1 .29** .19** .35** −.01 .25** .24** .34** .29** .29** −.04 .12* .06 .45** –
15 Incr. Struc JR T2 .45** .30** .41** .04 .49** .43** .23** .20** .26** .04 .16** .09 .33** .26** –
16 Incr. Soc JR T2 .41** .49** .38** .19** .31** .23** .11 .21** .14* .15** .08 .10 .07 .15** .51** –
17 Incr. Chall JD T2 .33** .21** .37** −.01 .39** .30** .10 .08 .13* .10 .13* .12* .24** .27** .67** .56** –
18 Decr. Hind JD T2 .09 .20** .02 .40** −.03 −.15* −.09 −.07 −.16** .13* −.07 .03 −.15** −.06 .12 .33** .14* –
19 Work engagement T3 .28** .23** .32** −.05 .85** .54** .31** .39** .41** −.02 .27** .23** .31** .23** .49** .27** .39** −.05 –
20 In-role performance T3 .14* .15* .23** −.10 .30** .27** .21** .23** .24** −.09 .23** .08 .55** .36** .27** .08 .23** .25** .34** –
21 OCBI T3 .29** .20** .36** .01 .25** .27** .29** .25** .24** −.01 .05 .05 .31** .62** .32** .18** .27** −.05 .29** .40** –

Int. = Intention; Incr. = increase; Struc = structural; Soc = social; Chall = challenging; Hind = hindering; JR = job resources, JD = job demands; T1 = Time 1; T2 = Time 2; T3 = Time 3.
*p < .05, **p < .01.
 JOB CRAFTING 9

TABLE 2
Invariance test of measures of work engagement, in-role performance, and OCBI

χ2 df CFI TLI RMSEA Δχ2/Δdf p

Model A 922.87 494 .93 .92 .06

Model B 936.50 508 .93 .92 .05 13.63/14 .48
Model C 940.44 511 .93 .92 .05 3.94/3 .27
Model D 977.11 528 .93 .92 .05 36.67/17 <.01

χ2 = Chi-square; df = degrees of freedom; CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root-mean-square error of
approximation; Δχ2/Δdf = change in chi-square and degrees of freedom.

the field of social sciences (Horn, McArdle, & Mason, challenging job demands was positively related to
1983; Kim & Ji, 2009). We therefore conclude that the Crafting these job resources and challenging job
measures show satisfactory longitudinal invariance. demands (λ = .38, p < .001). In addition, the direct effect
from T1 Intention to craft hindering job demands to T2
Crafting hindering job demands was also significant
Downloaded by [University of Guelph] at 07:16 13 March 2015

Measurement models (λ = .46, p < .001). Thus, Hypothesis 1a was supported.

Measurement models were examined to check the validity Hypothesis 1b suggested that T1 work engagement
of the constructs. Important to check first is whether job predicts T2 job crafting. Work engagement T1 signifi-
crafting intentions and actual job crafting are different con- cantly related to Crafting job resources and challenging
structs. The four-factor model with two latent factors for job job demands at T2 (λ = .21, p < .001).
crafting intentions and two latent factors for actual job
crafting behaviours showed a significantly better fit than a
two-factor model, in which corresponding job crafting Job crafting mediates the relationship
intentions and actual job crafting behaviours loaded on between job crafting intentions and work
the same factor (Δχ2 = 333.43 Δdf = 5, p < .01). Model fit engagement
indices of the hypothesized model were CFI = .94, In Hypothesis 2a, it was expected that Crafting job
TLI = .91, RMSEA = .10, and SRMR = .07. resources and challenging job demands at T2 mediate
Next, we examined the full measurement model, the relationship between T1 Intention to craft structural
which includes 12 latent variables: two latent factors job resources and challenging job demands and T3 work
for the job crafting intentions construct, labelled engagement. Hypothesis 1a already showed that
“Intention to craft job resources and challenging job Intention to craft job resources and challenging job
demands” and “Intention to craft hindering job demands and crafting these job resources and challen-
demands”, and two latent factors for the job crafting ging job demands were positively related. Crafting job
construct, named “Crafting job resources and challen- resources and challenging job demands, in turn, related
ging job demands” and “Crafting hindering job significantly to work engagement (β = .20, p < .05),
demands”. In addition, the measurement model included while controlling for T1 work engagement and T1 job
two latent factors for job demands and job resources, two resources and job demands (see Figure 1). The direct
latent factors for work engagement at T1 and T3, two relationship between Intention to craft job resources and
latent factors for in-role performance at T1 and T3, and challenging job demands and work engagement was not
two latent factors for OCBI at T1 and T3. Residual significant (λ = −.03, p = .77), indicating full mediation.
correlations between T1 and T3 indicators of the same The bootstrap estimate (.07) for the indirect effect fell
construct were allowed to account for their dependence within the 95% bias-corrected confidence interval
(Brown, 2006). The fit of this measurement model was (B-CCI: .019–.137), supporting the significance of
adequate: χ2 = 371.20, df = 195, TLI = .95, CFI = .96, Crafting job resources and challenging job demands as
RMSEA = .06, SRMR = .05. All indicators had signifi- a mediator in the relationship between Intention to craft
cant loadings on the intended factors (range λ = .67–.95; job resources and challenging job demands an work
p < .001). engagement.
The results regarding Hypothesis 2b showed that
Intention to craft hindering job demands and Crafting
Job crafting intentions predict actual job
hindering job demands were not related to work engage-
crafting
ment (β = −.03, p = .74 and β = −.06, p = .38, respec-
In Hypothesis 1a, we expected that T1 job crafting tively). Thus, Crafting hindering job demands at T2 did
intentions predict actual job crafting at T2. The results not mediate the relationship between T1 Intention to
showed that Intention to craft job resources and craft hindering job demands and T3 work engagement.
 10 TIMS, BAKKER, DERKS
Downloaded by [University of Guelph] at 07:16 13 March 2015

Figure 1. Longitudinal model showing that job crafting positively relates to work engagement and performance. JR = job resources; JD = job
demands; OCBI = organizational citizenship behaviour towards individuals. Dotted lines represent controlled relationship. Independent and dependent
variables were free to correlate with each other. **p < .01, *p < .05.

Work engagement as a mediator in the hypothesized model was acceptable, χ2 = 794.00,

relationship between job crafting and job df = 382, TLI = .91, CFI = .93, RMSEA = .06,
performance SRMR = .07.
In Hypothesis 3, it was proposed that T3 work engage-
ment mediates the relationship between T2 job crafting DISCUSSION
and T3 in-role performance and OCBI, while controlling
for all T1 variables. The results showed that T3 work The present study examined whether job crafting inten-
engagement was significantly associated with in-role tions and work engagement predicted actual job crafting
performance (β = .10, p < .05) but not with OCBI activities in the next month and whether actual job
(β = .08, p = .22). The relation between Crafting job crafting, in turn, predicted work engagement and indir-
resources and challenging job demands and in-role per- ectly in-role performance and OCBI over time. In what
formance was not significant (β = .11, p = .14), indicat- follows, we discuss the most important contributions of
ing full mediation. The bootstrap results showed that the the present study.
indirect effect of Crafting job resources and challenging
job demands to in-role performance via work engage-
ment was significant (.01, B-CCI = .001–.022, p < .05). Theoretical contributions
The relation between Crafting job resources and challen- The first contribution of the present study is that it
ging job demands and OCBI was not significant provides more insight into the predictors of job crafting
(β = .09, p = .11), indicating that T3 OCBI was not behaviours. It was found that the relationships between
influenced by T2 Crafting job resources and challenging job crafting intentions and actual job crafting behaviours
job demands and T3 work engagement. were strongly significant. The standardized estimates
Crafting hindering job demands related negatively to from job crafting intentions to actual job crafting were
in-role performance (β = −.23, p < .001) but was unre- close to the magnitude of the sample-weighted average
lated to work engagement and OCBI (β = −.06, p = .38, correlation of .53 between intentions and behaviour
β = −.02, p = .70, respectively). These findings show that found in the meta-analysis by Sheeran (2002). This
T3 work engagement does not mediate the relationship effect is considered to be large (Sheeran, 2002) and
between T2 Crafting hindering job demands and T3 job suggests that employees with job crafting intentions
performance and OCBI. In sum, Hypothesis 3 was sup- were most likely to perform job crafting activities in
ported with regard to the mediation of work engagement the future. In addition, work engagement reported at T1
in the relationship between Crafting job resources and was also predictive of crafting job resources and challen-
challenging job demands and job performance. In refer- ging job demands (but not crafting hindering job
ence to Crafting hindering job demands, the hypothesis demands) in the next month. This finding underscores
was not supported, but an interesting negative effect on the importance of work engagement for employee proac-
in-role performance was found. The fit of the tive behaviours (Sonnentag, 2003) and may also indicate
 JOB CRAFTING 11

that engaged employees want more from their work conceptually different from coping as coping refers to
(Warr & Inceoglu, 2012). strategies to manage stress (Folkman & Moskowitz,
A second theoretical contribution of this study is that 2004), whereas job crafting is a strategy to manage
crafting job resources and challenging job demands posi- work in such a way that it better fits the preferences,
tively predicted work engagement one month later, skills, and values of the employee (Wrzesniewski &
implying that employees who increased these job char- Dutton, 2001). Clearly, further research is required to
acteristics (but not those who decreased their hindering examine the process that underlies this relationship.
job demands) increased their own work engagement Finally, building on extant research, our study con-
through job crafting. Because we were able to control tributed to the literature on work engagement and job
for job characteristics, this finding suggests that only performance. Engaged employees are focused on per-
focusing on the effect of job characteristics on employ- forming their core job tasks well (Bakker & Bal, 2010;
ees’ work experience does not fully explain the existence Christian et al., 2011; Demerouti & Bakker, 2011;
or emergence of work engagement. Importantly, job Halbesleben & Wheeler, 2008), but engaged employees
crafting predicted work engagement over and above job did not report more OCBI than nonengaged employees.
characteristics. By autonomously increasing the avail- It may be that engaged employees are more focused on
ability of job resources, employees allow themselves to their in-role performance than their extra-role perfor-
do their job in a way that is consistent with their pre- mance. However, this relationship needs further explora-
Downloaded by [University of Guelph] at 07:16 13 March 2015

ferences, values, and/or skills (Berg et al., 2008; Tims tion as there are also indications that work engagement is
et al., 2012). As a result, employees become more associated with both in-role performance and OCB (Rich
engaged. In a similar vein, initiating more challenging et al., 2010). In addition, in the present study, the corre-
job tasks or demands predicted higher levels of vigour, lations between increasing job resources and increasing
dedication, and absorption. Complex and intellectually challenging job demands and OCBI are all significant,
demanding tasks are found to give rise to feelings of whereas these relationships do not reach significance in
interest and involvement (Amabile, 1996) and therefore the research model. While relating job crafting beha-
likely foster work engagement. These results provide viours to the specific OCBI behaviours (e.g., helping
additional support for a gain spiral in which engaged others who have been absent or have a high workload,
employees are more likely to proactively change their assisting supervisor, taking personal interest in other
job resources and challenging job demands which in turn employees), these positive relationships are as expected.
leads to higher levels of work engagement (Bakker, For example, employees who enjoy having variety at
2011; Bakker & Demerouti, 2014; Hakanen, work and learning new things may be more likely to
Perhoniemi, et al., 2008). help others and assist their supervisor without being
In contrast, however, crafting decreasing hindering asked as this OCB may create learning opportunities.
job demands was not related to work engagement. We therefore expect that the nonsignificant finding is
Although earlier studies theoretically argued why likely due to the stringent test we performed by control-
decreasing hindering job demands may be beneficial ling for T1 OCBI.
for employees (e.g., to reduce negative work and worker
outcomes; Podsakoff, LePine, & LePine, 2007; to avoid
Limitations
or cope with negative job aspects; Berg et al., 2008),
there was no statistical evidence for this assertion. In line There are some limitations that must be acknowledged.
with earlier studies (Tims et al., 2012; Tims, Bakker, First, work engagement, in-role performance, and OCBI
Derks, & Van Rhenen, 2013), we can now conclude were measured at the same measurement occasions (T1
that decreasing hindering job demands does not lead to and T3), which precludes causal inferences regarding the
enhanced motivation and performance. This job crafting mediating role of work engagement. Additional analyses,
dimension is different from the other three job crafting in which all three variables were modelled as dependent
dimensions. variables, showed that the fit of the hypothesized model
Some studies have found that decreasing hindering was significantly better than the alternative model, pro-
job demands is related to burnout (Tims et al., 2012), viding more evidence that work engagement is not only
which may indicate that employees who attempt to an outcome of job crafting but also a mediator between
decrease their hindering demands may perceive to have antecedents and outcomes (cf. Saks, 2006). In addition,
no other options to prevent worse negative health out- these findings are in line with previous studies that
comes. Seen this way, decreasing hindering job demands examined the causal effect of work engagement on job
may share similarities with avoidance-oriented coping, in performance (Bakker & Bal, 2010; Salanova et al.,
which people try to evade a problem. The other job 2005).
crafting dimensions may be more similar to approach- Another limitation is that we were not able to measure
oriented coping, in which people attempt to confront all variables at all measurement occasions because we
problems and deal directly with them (Roth & Cohen, had to reduce the time participants needed to fill out the
1986). Important to note is that job crafting is questionnaires. However, we did make sure that we had
 12 TIMS, BAKKER, DERKS

premeasures and postmeasures of the dependent vari- translated into behaviours (Webb & Sheeran, 2006).
ables so that we could control for previous levels of Formulating implementation intentions about when,
the dependent variables. where (e.g., in the next staff meeting), and how (e.g.,
A third limitation of the study is that we were not able asking the team) to craft a specific job characteristic may
to explain the finding with respect to the job crafting be an important first step towards optimizing the work
dimension “Decreasing hindering job demands”. This environment and improving engagement and job
type of job crafting consistently shows opposite relation- performance.
ships in comparison to the job crafting dimensions Using the theory of planned behaviour (Ajzen, 1991,
related to increasing job resources and challenging job 2012), another way to encourage employees to formulate
demands. Although employees are least likely to engage job crafting intentions, and thereby the likelihood that
in this type of job crafting (in terms of mean level), it is they show job crafting behaviours, may be by changing
important to learn what their motivations are for decreas- employees’ attitudes about job crafting, creating social
ing these demands. Is it indeed a strategy used by norms for job crafting, and increasing employees’ per-
slightly burned-out employees who try to prevent drop- ceived behavioural control. Individuals will be most
ping out from a burnout? Or can burnout function as a likely to formulate job crafting intentions when they
mediator between decreasing hindering job demands and hold a positive attitude towards job crafting, when they
in-role performance instead of work engagement? feel social pressure to engage in job crafting, and when
Downloaded by [University of Guelph] at 07:16 13 March 2015

According to self-determination theory (Deci & Ryan, they perceive job crafting to be easy. Organizational
2000), the frustration of basic psychological needs, for interventions could use these principles to stimulate job
example, through high levels of hindering job demands, crafting among employees.
may explain the emergence of burnout (Van Den Broeck, Finally, the study showed that employees who
Vansteenkiste, De Witte, & Lens, 2008). In addition, decreased their hindering job demands were not more
what harm may decreasing hindering job demands do engaged. Moreover, crafting hindering job demands was
in relation to the employees’ reputation or performance related to lower levels of in-role performance. It seems
assessment? What do employees think to win from important that hindering demands that by definition hin-
decreasing hindering job demands? Furthermore, it der the attainment of goals (Crawford et al., 2010) are
could be that increasing social job resources, such as reduced to a minimum. Although employees may
support from colleagues or the supervisor, may be a decrease their hindering job demands, they seem to
better strategy to deal with these job demands. This impair work motivation and subsequently job
could be an interesting research question for future performance.
studies.
Conclusion
Practical implications
Several recent studies demonstrated that job crafting is
The findings of the present study have at least two an important work-related phenomenon that is related to
practical implications. First, intentions are found to be positive outcomes, such as work engagement and job
of importance for future job crafting behaviours. As not performance. The present study contributed to this lit-
every individual may be inclined to engage in job craft- erature by showing that job crafting intentions and work
ing, a means to increase employees’ awareness about engagement are important predictors of actual job craft-
their work environment and how they can influence ing which in turn related to work engagement and job
their job characteristics may be a feedback intervention. performance. We can conclude that employees experi-
By providing employees with feedback on their job ence engagement when they craft their job resources and
demands and job resources combined with suggestions challenging job demands. These employees are valuable
how levels of job demands and resources may be influ- for organizations because they are more likely to focus
enced, employees may be encouraged to think about their energies on their in-role tasks.
how they can change their work environment with rela-
tively simple adaptations. Also, the suggestions may REFERENCES
signal to employees that these proactive behaviours are
Ajzen, I. (1991). The theory of planned behaviour. Organizational
welcomed by the organization. As a result, employees
Behavior and Human Decision Processes, 50, 179–211.
may formulate intentions to enact their job characteristics doi:10.1016/0749-5978(91)90020-T
and take more responsibility for their work engagement. Ajzen, I. (2012). The theory of planned behaviour. In P. A. M. Lange,
Such an intervention may also be helpful in teaching A. W. Kruglanski, and E. T. Higgins (Eds.), Handbook of theories
employees to form implementation intentions of social psychology (Vol. 1, pp. 438–459). London: Sage.
Amabile, T. M. (1996). Creativity in context. Boulder, CO: Westview
(Gollwitzer, 1993, 1999) that specify when, where, and
Press.
how one will perform specific behaviours. Arbuckle, J. L. (2005). Amos 6.0 User’s Guide. Chicago, IL: SPSS, Inc.
Implementation intentions may increase the stability of Baard, P. P., Deci, E. L., & Ryan, R. M. (2004). Intrinsic need satisfac-
the intention and make it more likely that intentions are tion: A motivational basis of performance and well-being in two
 JOB CRAFTING 13

work settings. Journal of Applied Social Psychology, 34, Davidson, A. R., & Jaccard, J. J. (1979). Variables that moderate the
2045–2068. doi:10.1111/j.1559-1816.2004.tb02690.x attitude-behaviour relation: Results of a longitudinal survey.
Bakker, A. B. (2011). An evidence-based model of work engagement. Journal of Personality and Social Psychology, 37, 1364–1376.
Current Directions in Psychological Science, 20, 265–269. doi:10.1037/0022-3514.37.8.1364
doi:10.1177/0963721411414534 Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits:
Bakker, A. B., & Bal, P. M. (2010). Weekly work engagement and Human needs and the self-determination of behaviour. Psychological
performance: A study among starting teachers. Journal of Inquiry, 11, 227–268. doi:10.1207/S15327965PLI1104_01
Occupational and Organizational Psychology, 83, 189–206. Deci, E. L., Ryan, R. M., Gagné, M., Leone, D., Usunov, J., &
doi:10.1348/096317909X402596 Kornazheva, B. P. (2001). Need satisfaction, motivation, and well-
Bakker, A. B., & Demerouti, E. (2007). The job demands-resources bing in the work organizations of a former eastern Bloc country: A
model: State of the art. Journal of Managerial Psychology, 22, cross-cultural study of self-determination. Personality and Social
309–328. doi:10.1108/02683940710733115 Psychology Bulletin, 27, 930–942. doi:10.1177/0146167201278002
Bakker, A. B., & Demerouti, E. (2014). Job demands-resources theory. Demerouti, E., & Bakker, A. B. (2011). The job demands-resources
In P. Y. Chen and C. L. Cooper (Eds.), Work and wellbeing: model: Challenges for future research. South African Journal of
Wellbeing: A complete reference guide (Vol. III, pp. 37–64). Industrial Psychology, 37, 974–983.
Chichester: Wiley-Blackwell. Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B.
Bakker, A. B., Demerouti, E., & Sanz-Vergel, A. I. (2014). Burnout and (2001). The job demands-resources model of burnout. Journal of
work engagement: The JD-R approach. Annual Review of Applied Psychology, 86, 499–512. doi:10.1037/0021-9010.86.3.499
Organizational Psychology and Organizational Behavior, 1, Floyd, F. J., & Widaman, K. F. (1995). Factor analysis in the develop-
389–411. doi:10.1146/annurev-orgpsych-031413-091235 ment and refinement of clinical assessment instruments.
Bakker, A. B., Demerouti, E., Taris, T. W., Schaufeli, W. B., & Psychological Assessment, 7, 286–299. doi:10.1037/1040-
Downloaded by [University of Guelph] at 07:16 13 March 2015

Schreurs, P. J. G. (2003). A multigroup analysis of the job 3590.7.3.286

demands-resources model in four home care organizations. Folkman, S., & Moskowitz, J. T. (2004). Coping: Pitfalls and promise.
International Journal of Stress Management, 10, 16–38. Annual Review of Psychology, 55, 745–774. doi:10.1146/annurev.
doi:10.1037/1072-5245.10.1.16 psych.55.090902.141456
Bakker, A. B., Demerouti, E., & Ten Brummelhuis, L. L. (2012). Work Fredrickson, B. (2000). Why positive emotions matter in organizations:
engagement, performance, and active learning: The role of con- Lessons from the broaden-and-build model. The Psychologist-
scientiousness. Journal of Vocational Behavior, 80, 555–564. Manager Journal, 4, 131–142. doi:10.1037/h0095887
doi:10.1016/j.jvb.2011.08.008 Furda, J. (1995). Werk, persoon en welzijn: Een toets van het JD-C
Bakker, A. B., Demerouti, E., & Verbeke, W. (2004). Using the job model [Work, personality, and wellbeing: A test of the JD-C model]
demands-resources model to predict burnout and performance. (Unpublished doctoral dissertation). Utrecht University, Utrecht.
Human Resource Management, 43, 83–104. doi:10.1002/hrm.20004 Ghitulescu, B. E. (2006). Shaping tasks and relationships at work:
Bakker, A. B., Hakanen, J. J., Demerouti, E., & Xanthopoulou, D. Examining the antecedents and consequences of employee job
(2007). Job resources boost work engagement particularly when job crafting (Unpublished dissertation). University of Pittsburgh,
demands are high. Journal of Educational Psychology, 99, Pittsburgh, PA.
274–284. doi:10.1037/0022-0663.99.2.274 Gollwitzer, P. M. (1993). Goal achievement: The role of intentions. In
Bakker, A. B., Tims, M., & Derks, D. (2012). Proactive personality and W. Stroebe and M. Hewstone (Eds.), Handbook of motivation and
job performance: The role of job crafting and work engagement. cognition (Vol. 2, pp. 53–92). Chichester: Wiley.
Human Relations, 65, 1359–1378. doi:10.1177/0018726712453471 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of
Berg, J. M., Dutton, J. E., & Wrzesniewski, A. (2008). What is job simple plans. American Psychologist, 54, 493–503. doi:10.1037/
crafting and why does it matter? Retrieved May 30, 2013, from 0003-066X.54.7.493
http://www.centerforpos.org/ Hackman, J. R., & Oldham, G. R. (1976). Motivation through the
Berg, J. M., Wrzesniewski, A., Dutton, J. E., Grant, A., Fried, Y., design of work: Test of a theory. Organizational Behavior and
Parker, S., & Frese, M. (2010). Perceiving and responding to Human Performance, 16, 250–279. doi:10.1016/0030-5073(76)
challenges in job crafting at different ranks: When proactivity 90016-7
requires adaptivity. Journal of Organizational Behavior, 31, 158– Hakanen, J. J., Perhoniemi, R., & Toppinen-Tanner, S. (2008). Positive
186. doi:10.1002/job.645 gain spirals at work: From job resources to work engagement,
Berings, D., De Fruyt, F., & Bouwen, R. (2004). Work values and personal initiative and work-unit innovativeness. Journal of
personality traits as predictors of enterprising and social vocational Vocational Behavior, 73, 78–91. doi:10.1016/j.jvb.2008.01.003
interests. Personality and Individual Differences, 36, 349–364. Hakanen, J. J., Schaufeli, W. B., & Ahola, K. (2008). The Job
doi:10.1016/S0191-8869(03)00101-6 Demands-Resources model: A three-year cross-lagged study of
Brown, T. A. (2006). Confirmatory factor analysis for applied burnout, depression, commitment, and work engagement. Work &
research. New York, NY: The Guilford Press. Stress, 22, 224–241. doi:10.1080/02678370802379432
Byrne, B. M. (2004). Testing for multigroup invariance using AMOS Halbesleben, J. R. B. (2010). A meta-analysis of work engagement:
graphics: A road less traveled. Structural Equation Modeling, 11, Relationships with burnout, demands, resources, and consequences.
272–300. In A. B. Bakker and M. P. Leiter (Eds.), Work engagement: A
Cavanaugh, M. A., Boswell, W. R., Roehling, M. V., & Boudreau, J. handbook of essential theory and research (pp. 102–117). New
W. (2000). An empirical examination of self-reported work stress York, NY: Psychology Press.
among US managers. Journal of Applied Psychology, 85, 65–74. Halbesleben, J. R. B., & Wheeler, A. R. (2008). The relative roles of
doi:10.1037/0021-9010.85.1.65 engagement and embeddedness in predicting job performance and
Christian, M. S., Garza, A. S., & Slaughter, J. E. (2011). Work engage- intention to leave. Work & Stress, 22, 242–256. doi:10.1080/
ment: A quantitative review and test of its relations with task and 02678370802383962
contextual performance. Personnel Psychology, 64, 89–136. Horn, J. L., & McArdle, J. J. (1992). A practical and theoretical guide
doi:10.1111/j.1744-6570.2010.01203.x to measurement invariance in aging research. Experimental Aging
Crawford, E. R., LePine, J. A., & Rich, B. L. (2010). Linking job Research, 18, 117–144.
demands and resources to employee engagement and burnout: A Horn, J. L., McArdle, J. J., & Mason, R. (1983). When is invariance
theoretical extension and meta-analytic test. Journal of Applied not invarient: A practical scientist's look at the ethereal concept of
Psychology, 95, 834–848. doi:10.1037/a0019364 factor invariance. Southern Psychologist, 1, 179–188.
 14 TIMS, BAKKER, DERKS

Karasek, R. A. (1979). Job demands, job decision latitude, and mental Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
strain: Implications for job redesign. Administrative Science strategies for assessing and comparing indirect effects in multiple
Quarterly, 24, 285–307. mediator models. Behavior Research Methods, 40, 879–891.
Kim, H., & Ji, J. (2009). Factor structure and longitudinal invariance of doi:10.3758/BRM.40.3.879
the Maslach Burnout Inventory. Research on Social Work Practice, Rich, B. L., LePine, J. A., & Crawford, E. R. (2010). Job engagement:
19, 325–339. Antecedents and effects on job performance. Academy of Management
Kulik, C. T., Oldham, G. R., & Hackman, J. R. (1987). Work design as Journal, 53, 617–635. doi:10.5465/AMJ.2010.51468988
an approach to person-environment fit. Journal of Vocational Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with
Behavior, 31, 278–296. doi:10.1016/0001-8791(87)90044-3 stress. American Psychologist, 41, 813–819. doi:10.1037/0003-
Le Blanc, P. M. (1994). De steun van de leiding: Een onderzoek naar 066X.41.7.813
het Leader Member Exchange model in de verpleging [Leader’s Saks, A. M. (2006). Antecedents and consequences of employee
support: A study of the leader member exchange model among engagement. Journal of Managerial Psychology, 21, 600–619.
nurses]. Amsterdam: Thesis Publishers. doi:10.1108/02683940610690169
Leana, C., Appelbaum, E., & Shevchuk, I. (2009). Work process and Salanova, M., Agut, S., & Peiró, J. M. (2005). Linking organizational
quality of care in early childhood education: The role of job craft- resources and work engagement to employee performance and
ing. Academy of Management Journal, 52, 1169–1192. customer loyalty: The mediation of service climate. Journal of
doi:10.5465/AMJ.2009.47084651 Applied Psychology, 90, 1217–1227. doi:10.1037/0021-
LePine, J. A., Podsakoff, N. P., & LePine, M. A. (2005). A meta- 9010.90.6.1217
analytic test of the challenge stressor-hindrance stressor framework: Salanova, M., & Schaufeli, W. B. (2008). A cross-national study of
An explanation for inconsistent relationships among stressors and work engagement as a mediator between job resources and
performance. Academy of Management Journal, 48, 764–775. proactive behaviour. The International Journal of Human
Downloaded by [University of Guelph] at 07:16 13 March 2015

doi:10.5465/AMJ.2005.18803921 Resource Management, 19, 116–131. doi:10.1080/

Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. 09585190701763982
(2002). To parcel or not to parcel: Exploring the question, weighing Schaufeli, W. B., & Bakker, A. B. (2004). Job demands, job resources,
the merits. Structural Equation Modeling: A Multidisciplinary and their relationship with burnout and engagement: A multi-sam-
Journal, 9, 151−173. doi:10.1207/S15328007SEM0902_1 ple study. Journal of Organizational Behavior, 25, 293–315.
Lu, C., Wang, H., Lu, J., Du, D., & Bakker, A. B. (2014). Does work doi:10.1002/job.248
engagement increase person-job fit? The role of job crafting and job Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The measure-
insecurity. Journal of Vocational Behavior, 84, 142–152. ment of work engagement with a short questionnaire: A cross-
doi:10.1016/j.jvb.2013.12.004 national study. Educational and Psychological Measurement, 66,
Lyons, P. (2008). The crafting of jobs and individual differences. 701–716. doi:10.1177/0013164405282471
Journal of Business Psychology, 23, 25–36. doi:10.1007/s10869- Schaufeli, W. B., Salanova, M., González-Roma, V., & Bakker, A. B.
008-9080-2 (2002). The measurement of engagement and burnout: A two
Marsh, H. W., Hau, K., & Wen, Z. (2004). In search of golden rules: sample confirmatory factor analytic approach. Journal of
Comment on hypothesis-testing approaches to setting cutoff values Happiness Studies, 3, 71–92. doi:10.1023/A:1015630930326
for fit indexes and dangers in overgeneralizing Hu and Bentler’s Sheeran, P. (2002). Intention-behaviour relations: A conceptual and
(1999) findings. Structural Equation Modeling: A Multidisciplinary empirical review. European Review of Social Psychology, 12, 1–
Journal, 11, 320–341. doi:10.1207/s15328007sem1103_2 36. doi:10.1080/14792772143000003
Marsh, H. W., Hau, K. T., Balla, J. R., & Grayson, D. (1998). Is more Sonnentag, S. (2003). Recovery, work engagement, and proactive
ever too much? The number of indicators per factor in confirmatory behaviour: A new look at the interface between nonwork and
factor analysis. Multivariate Behavioral Research, 33, 181–220. work. Journal of Applied Psychology, 88, 518–528. doi:10.1037/
doi:10.1207/s15327906mbr3302_1 0021-9010.88.3.518
Nahrgang, J. D., Morgeson, F. P., & Hofmann, D. A. (2011). Safety at Tims, M., & Bakker, A. B. (2010). Job crafting: Towards a new model
work: A meta-analytic investigation of the link between job of individual job redesign. South African Journal of Industrial
demands, job resources, burnout, engagement, and safety outcomes. Psychology, 36, 1–9.
Journal of Applied Psychology, 96, 71–94. doi:10.1037/a0021484 Tims, M., Bakker, A. B., & Derks, D. (2012). The development and
Organ, D. W., & Konovsky, M. (1989). Cognitive versus affective validation of the job crafting scale. Journal of Vocational Behavior,
determinants of organizational citizenship behaviour. Journal of 80, 173–186. doi:10.1016/j.jvb.2011.05.009
Applied Psychology, 74, 157–164. doi:10.1037/0021-9010.74.1.157 Tims, M., Bakker, A. B., & Derks, D. (2013). The impact of job crafting on
Parker, S. K., Bindl, U. K., & Strauss, K. (2010). Making things job demands, job resources, and well-being. Journal of Occupational
happen: A model of proactive motivation. Journal of Health Psychology, 18, 230–240. doi:10.1037/a0032141
Management, 36, 827–856. doi:10.1177/0149206310363732 Tims, M., Bakker, A. B., Derks, D., & Van Rhenen, W. (2013). Job
Parker, S. K., & Collins, C. G. (2010). Taking stock: Integrating and crafting at the team and individual level: Implications for work
differentiating multiple proactive behaviours. Journal of engagement and performance. Group and Organization
Management, 36, 633–662. doi:10.1177/0149206308321554 Management, 38, 427–454. doi:10.1177/1059601113492421
Petrou, P., Demerouti, E., Peeters, M. C. W., Schaufeli, W., & Hetland, Triandis, H. C. (1980). Values, attitudes, and interpersonal behaviour. In
J. (2012). Crafting a job on a daily basis: Contextual correlates and H. E. Howe Jr. and M. Page (Eds.), Nebraska symposium of motivation
the link to work engagement. Journal of Organizational Behavior, (Vol. 27, pp. 195–259). Lincoln: University of Nebraska Press.
33, 1120–1141. doi:10.1002/job.1783 Van Den Broeck, A., Vansteenkiste, M., De Witte, H., & Lens, W.
Pitts, S. C., West, S. G., & Tein, J. Y. (1996). Longitudinal measure- (2008). Explaining the relationships between job characteristics,
ment models in evaluation research: Examining stability and burnout, and engagement: The role of basic psychological need
change. Evaluation and Program Planning, 19, 333–350. satisfaction. Work & Stress, 22, 277–294. doi:10.1080/
Podsakoff, N. P., LePine, J. A., & LePine, M. A. (2007). Differential 02678370802393672
challenge stressor-hindrance stressor relationships with job atti- Van Veldhoven, M. J. P. M., & Meijman, T. (1994). Het meten van
tudes, turnover intentions, turnover, and withdrawal behaviour: A psychosocialearbeidsbelasting met een vragenlijst: De vragenlijst-
meta-analysis. Journal of Applied Psychology, 92, 438–454. beleving en beoordeling van de arbeid. Amsterdam: Nederlands
doi:10.1037/0021-9010.92.2.438 Instituut voor Arbeidsomstandigheden.
 JOB CRAFTING 15

Warr, P., & Inceoglu, I. (2012). Job engagement, job satisfaction, and Wrzesniewski, A. (2003). Finding positive meaning in work. In K. S.
contrasting associations with person-job fit. Journal of Cameron, J. E. Dutton, and R. E. Quinn (Eds.), Positive organiza-
Occupational Health Psychology, 17, 129–138. doi:10.1037/ tional scholarship: Foundations of a new discipline (pp. 298–308).
a0026859 San Francisco, CA: Berrett-Koehler.
Webb, T. L., & Sheeran, P. (2006). Does changing behavioural inten- Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning
tions engender behaviour change? A meta-analysis of the experi- employees as active crafters of their work. Academy of
mental evidence. Psychological Bulletin, 132, 249–268. Management Review, 26, 179–201.
doi:10.1037/0033-2909.132.2.249
Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organi- Original manuscript received January 2014
zational commitment as predictors of organizational citizenship and
Revised manuscript received September 2014
in-role behaviours. Journal of Management, 17, 601–617. Revised manuscript accepted September 2014
doi:10.1177/014920639101700305 First published online October 2014
Downloaded by [University of Guelph] at 07:16 13 March 2015