Journal of Service Research http://jsr.sagepub.

com/

Job Characteristics and the Creativity of Frontline Service Employees

Filipe Coelho and Mário Augusto
Journal of Service Research 2010 13: 426 originally published online 21 May 2010
DOI: 10.1177/1094670510369379

The online version of this article can be found at:

http://jsr.sagepub.com/content/13/4/426

Published by:

http://www.sagepublications.com

On behalf of:

Center for Excellence in Service, University of Maryland

Additional services and information for Journal of Service Research can be found at:

Email Alerts: http://jsr.sagepub.com/cgi/alerts

Subscriptions: http://jsr.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations: http://jsr.sagepub.com/content/13/4/426.refs.html

>> Version of Record - Oct 24, 2010

OnlineFirst Version of Record - May 21, 2010

What is This?

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

 Journal of Service Research
13(4) 426-438
Job Characteristics and the Creativity of ª The Author(s) 2010
Reprints and permission:
sagepub.com/journalsPermissions.nav
Frontline Service Employees DOI: 10.1177/1094670510369379
http://jsr.sagepub.com

Filipe Coelho1 and Mário Augusto1

Abstract
This study investigates the main and interactive effects of job characteristics on the creativity of frontline service employees. Past
research investigates the link between job complexity, an index measure of job characteristics, and employee creativity. This study
follows a component-wise approach in studying the influence of each job characteristic on the creativity of employees. The
approach overcomes a number of limitations such as the sole reliance on the intrinsic motivation rationale. In addition, we con-
duct this research in services, whereas past research has focused primarily on other settings. The study tests a model with 460
employees from a service setting and shows that consideration of the effects of each job characteristic has an explanatory power
beyond that of job complexity. Furthermore, the results show that job characteristics interact with each other to affect creativity.
Finally, the results also indicate that the component-wise approach is useful for managers for prioritizing efforts. In particular, the
results indicate that to promote creative behaviors, service managers should consider increasing employee job autonomy, variety,
feedback, and identity. However, the interactions between job characteristics also reveal that to build an environment that sti-
mulates creative performance, it is important to find the right balance between job characteristics.

Keywords
creativity, job characteristics, frontline employees, services

Introduction between job complexity, an index measure of job characteristics,

and employee creativity (e.g., Oldham and Cummings 1996).
Building an appropriate context for innovation to happen is
The effect of job characteristics on creativity emerges from the
paramount (Auh and Menguc 2005; Baker and Sinkula 1999; employee motivation that such characteristics can generate.
Day 1994; Slater and Narver 1995). At the heart of innovation,
Basically, jobs that are complex (i.e., rich in variety, identity,
however, lies the creativity of employees. Creative employees
autonomy, feedback, and significance) enhance ‘‘individuals’
develop new ideas and suggest novel products and procedures
excitement about their work activities and their interest in com-
which, in turn, can be subsequently implemented and used to
pleting these activities, and this excitement should foster creativ-
promote the organization’s survival and effectiveness (Oldham
ity’’ (Shalley, Zhou, and Oldham 2004, p. 938). Explicit in this
and Cummings 1996). Not surprisingly, the creativity of
research effort is a composite, aggregate approach to study the
employees is necessary for the competitive advantage of orga-
influences of job characteristics.
nizations (Amabile 1996; Shalley 1995; Shalley, Zhou, and Our study adds to the literature in a number of ways. First,
Oldham 2004; Woodman, Sawyer, and Griffin 1993).
consider that the extant literature relies solely on the intrinsic
Creativity is ‘‘the development of ideas about products,
motivation perspective, together with the composite approach
practices, services, or procedures that are (a) novel and
to job characteristics, to explain the relationship between job
(b) potentially useful to the organization’’ (Shalley, Zhou, and
design and creativity. This is a rather limiting perspective to
Oldham 2004, p. 934). Ideas are considered novel when they
explain the role played by job characteristics, particularly con-
entail a substantial recombination of existing materials or the
sidering their multidimensional nature. Each job characteristic
development of materials that are completely new (Oldham and
has its own locus of interest and involves different cognitive
Cummings 1996, p. 608). They are useful when they can pro-
vide direct or indirect value to an organization in the short or
long term (Shalley, Zhou, and Oldham 2004).
1
The creativity of employees is attracting growing attention Faculdade de Economia da Universidade de Coimbra, Coimbra, Portugal
from scholars, seeking to understand how it can be promoted and
Corresponding Author:
sustained. In this study, we investigate the additive and interac- Filipe Coelho, Faculdade de Economia da Universidade de Coimbra, Av. Dias da
tive effects of job characteristics on the creativity of frontline Silva 165, 3004-512 Coimbra, Portugal
service employees. Past research has already established a link Email: [email protected]

426 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 427

activities. This suggests that each job characteristic influence addressing the unique needs of each customer requires
creativity through different mechanisms. In summary, relying creative behaviors from frontline employees. In particular,
solely on the intrinsic motivation rationale and on the composite creative employees are more likely to interact with the cus-
approach to job characteristics neglects other consequences that tomer in a way that better enables him or her to discover the
can be associated with job characteristics. Accordingly, we aim needs of customers, to communicate with customers in custo-
to expand this by testing the effect of each job characteristic on mized ways, and to craft the service in a way that meets the
creativity. In addition, we advance (but do not test) new mechan- specific needs of each customer. This suggests that organiza-
isms through which different job characteristics may impact on tional performance might be highly dependent on the creative
creativity. efforts exhibited by frontline employees. Not surprisingly,
Evidence from several studies seems to support our Bitner, Booms, and Tetreault (1990) observed that the expe-
approach. Oldham and Cummings (1996), for instance, rience of customers during service encounters was enhanced
obtained a relatively low median alpha reliability for the job when employees were able to meet customers’ special needs
complexity measure. This suggests that the degree of overlap and requests. Therefore, researching the drivers of creativity
or correlation between job characteristics is not that strong, in service settings seems to be of the utmost importance. This
and that possibly each job characteristic, apart from the is further reinforced by Mumford’s (2003) contentions that
intrinsic motivation link, also has some other effects on crea- different task demands may require different backgrounds,
tive performance. Furthermore, Hackman and Oldham (1980, motivations, and cognitive strategies.
p. 79) contend that an employee with a below average signif- In summary, following a component-wise rather than a com-
icance in his or her job may find it meaningful if it is high on posite approach and considering an interactionist model to
skill variety and task identity. This implies that a composite study the influence of job characteristics on the creative perfor-
approach may lead to a cancellation of specific effects. Sev- mance of frontline service employees permits the investigation
eral studies also report a low explanatory power associated of a number of issues that have not yet been dealt with theore-
with job complexity (e.g., Oldham and Cummings 1996). tically and empirically.
Moreover, a number of researchers (e.g., Farmer, Tierney,
and Kung-McIntyre 2003; Tierney and Farmer 2004) observe
no support for a direct link between job complexity and Research Model
employee creativity. These results can probably be attributed,
at least in part, to the composite approach, thus also question-
Research Background
ing its adequacy. Following Amabile’s (1983, 1996) componential framework of
Second, extant literature on job characteristics has been creativity, the production of creative work requires three types
neglecting any interactions between them. However, there are of antecedents, these being domain-relevant skills, creativity-
theoretical and empirical precedents to consider that job char- relevant skills, and task motivation. Domain-relevant skills
acteristics interact with each other to influence employee atti- involve the skills to perform competently in a specific domain,
tudes and behaviors (e.g., Dodd and Ganster 1996), and this including factual knowledge about the domain, special domain-
study addresses this notion. Third, it is also likely that each job relevant talents, and technical skills. Creativity-relevant skills
characteristic has differential importance in explaining creativ- contribute to creativity across domains (Amabile 1996) and
ity, and this will be especially true across business sectors. include an appropriate cognitive style, a conducive work style,
With this knowledge, managers will be in a better position to and implicit or explicit knowledge of heuristics for generating
prioritize efforts in order to maximally influence creativity. novel ideas. Task motivation refers to motivational variables
Consequently, consideration of a component-wise approach determining the way an individual approaches a certain task.
also enables us to identify the relative importance of each job This includes, for example, the individual’s intrinsic motiva-
characteristic. tion toward the task.
Fourth, studies on creativity primarily focus on settings The literature on creativity has followed two major
other than services, yet services dominate the economic streams. One focuses on the personal characteristics that
activity of developed countries, commanding, in many econo- influence creativity, investigating in particular the determin-
mies, over 60% and even 70% of employment and gross ing role of personality and cognitive style. The other focuses
domestic production (GDP). Consequently, it is still not very on contextual factors, defined as ‘‘dimensions of the work
well known how well the previous findings relating to job environment that potentially influence an employee’s creativ-
complexity apply to service settings. Frontline service ity but that are not part of the individual’’ (Shalley, Zhou,
employees in particular, given their boundary spanning posi- and Oldham 2004, p. 935). Some of the context factors inves-
tion and the distinguishing features of services, have jobs that tigated include, for example, rewards (e.g., George and Zhou
tend to be unstructured, nonroutine, and that present many 2002), job complexity (e.g., Tierney and Farmer 2004), rela-
challenges. As these are characteristics that make jobs amen- tionships with supervisor (e.g., Zhou 2003), and relationships
able for creativity, frontline employees can play a pivotal role with coworkers (e.g., Zhou and George 2001). Job character-
in an organization’s innovativeness. Customers’ needs in istics, the focus of this study, concern the latter research
services tend to be heterogeneous, and this suggests that approach. Extant literature justifies a relationship between

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 427

428 Journal of Service Research 13(4)

job characteristics and creativity using an intrinsic motivation

perspective. Intrinsic motivation can be defined as the degree Task:
to which an employee is excited by a work activity and is Variety
Identity
motivated to engage in it for the sake of the activity itself Feedback H2,3,4,5
(Oldham and Cummings 1996). From a cognitive evaluation Significance
theory perspective, work environment factors have informa- H6a,6b,6d,6c
tional and controlling aspects (Shalley, Zhou, and Oldham
2004). When the informational aspects prevail, employees Employee
perceive that there is little pressure to undertake job tasks Task autonomy
H1 creativity
in externally determined ways. This is encouraging for
employees, building their intrinsic motivation. However,
when the controlling aspect is more salient, employees feel Figure 1. A Model of the Influence of Job Characteristics on
pressured to undertake job tasks in conformance with the Employee Creativity
external factor, and hence experience a reduction in their
motivation. Creativity requires individuals to have a heigh- generation, dissemination, and responsiveness, thus curtailing
tened interest in a certain problem and in discovering ways the delivery of innovative efforts directed at customers’ needs.
to solve it. In this context, intrinsic motivation has the impor- The services literature also stresses the role of autonomy in
tant role of controlling the attention that individuals attach to these settings. Bowen and Lawler (1992), for example, advance
the heuristic issues of creative tasks (Woodman, Sawyer, a number of benefits accruing from providing autonomy to
and Griffin 1993). In summary, complex jobs increase frontline employees. These include more enthusiastic and
employees’ excitement in connection with their jobs, an out- warm employee interactions with customers during service
come that should result in creative behaviors (Oldham and delivery, innovative service ideas, quicker responses to cus-
Cummings 1996). tomers’ needs and to service recovery situations, word-of-
mouth advertising, and more positive feelings toward the job.
Furthermore, Conger and Kanungo (1988) contend that one
Research Hypotheses of the key outcomes of empowerment is self-efficacy, which
The Job Characteristics Model of Hackman and Oldham (1975) refers to ‘‘beliefs in one’s capabilities to mobilize the motiva-
has been widely used to investigate the effects of job character- tion, cognitive resources, and courses of action needed to meet
istics (see, e.g., de Jong, de Ruyter, and Lemmink 2004; Frese, given situational demands’’ (Wood and Bandura 1989, p. 408).
Garst, and Fay 2007; Jonge and Schaufeli 1998; Singh 1998; Thus, autonomy appears to be pivotal for effective employee
Van Mierlo et al. 2007). The job characteristics in this frame- performance in service settings.
work are associated with several work-related outcomes, such
as employee satisfaction, motivation, and performance (e.g.,
Dubinsky and Skinner 1984; Tyagi 1985). The positive influ- Main Effects
ence of job characteristics, that is of job complexity on creativ- Task autonomy. Autonomy is the degree to which employees
ity also has support in several studies (e.g., Hatcher, Ross, and are free to determine the schedule of their work and the proce-
Collins 1989; Tierney and Farmer 2002). dures and equipment they will use to carry out their assign-
Following the componential model of creativity, job charac- ments. Autonomy contributes to intrinsic motivation, and
teristics impact on creative performance through their effects thus to creativity, because it provides a sense of freedom and
on intrinsic motivation. Notwithstanding, this study proposes also of responsibility and control for work outcomes, making
that job characteristics also impact on domain-task skills, thus the job more exciting (Amabile et al. 1996; Hackman and
further influencing creativity. In order to discuss the unique Oldham 1980; Tyagi 1985). Bowen and Lawler (1992, p. 33)
influences of each job characteristic on creative performance, pinpoint that autonomy gives employees the confidence that
this study follows a component-wise rather than a composite they have the resources to respond appropriately to customers’
approach to job design. In this model (see Figure 1), autonomy requests and that it ‘‘can lead to the sort of spontaneous, crea-
has a pivotal role in fostering creativity, as it provides employ- tive rule-breaking that can turn a potentially frustrated or angry
ees with the resources to experiment and, thus, to be creative. customer into a satisfied one.’’
More specifically, the model predicts that autonomy interacts However, autonomy should also impact on the domain-
with the remaining job characteristics. Of the five job charac- relevant skills of service employees. The intangibility, hetero-
teristics, autonomy has probably been the most widely studied. geneity, inseparability, and perishability of services create
At the organizational level, empowerment can be considered an many problems in the service delivery process (Zeithaml,
element of the organizational structure that is considered para- Bitner, and Gremler 2006), and this requires frontline employ-
mount for the implementation of a customer orientation ees to be flexible and innovative in performing their jobs
organization-wide (Kennedy, Goolsby, and Arnould 2003). In (cf. Chebat and Kollias 2000; Dubinsky et al. 1986). Generally,
this regard, Jaworski and Kohli (1993) obtained evidence that autonomy allows employees to approach customers and their
centralization, the reverse of empowerment, limits information jobs ‘‘in ways that make the most of their expertise and their

428 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 429

creative-thinking skills’’ (Amabile 1998, p. 82). Consequently, summary, identity enables employees to be more creative in
autonomy motivates and enables employees to try new ideas serving customers and to develop more interesting service
and learn from the consequences, and this expands their ideas. Therefore, we hypothesize the following.
domain-relevant skills, regardless of the effect on intrinsic
motivation. We thus predict the following. Hypothesis 3: There is a positive association between task
identity and employee creativity.
Hypothesis 1: There is a positive association between task
autonomy and employee creativity. Task feedback. Feedback is the degree to which employees
obtain, while in work, clear and direct information regarding
Task variety. Variety is the extent to which an employee has their job performance (Hackman and Oldham, 1980). In the
to perform a wide range of activities and/or operate with a vari- absence of performance feedback, employees have no
ety of equipment and procedures, involving the utilization of grounds for feeling good or bad about their performance, and
diverse skills (Hackman and Oldham 1980; Sims, Szilagyi, and this reduces their internal motivation. This, in turn, and
Keller 1976). Jobs high in variety increase intrinsic motivation according to the componential model of creativity, adversely
because they provide employees with the feeling that their affects creativity. Additionally, as feedback provides infor-
assignments are meaningful and worthwhile. Thus, and accord- mation that can be used to evaluate current efforts, it stimu-
ing to the componential model of creativity, variety contributes lates the employee to explore different courses of action in
to creative performance. Furthermore, employees in jobs with the pursuit of a better result (Earley et al. 1990), and this
these characteristics have more opportunities ‘‘to explore and should affect creative performance. As employees obtain
manipulate their environments and to gain a sense of efficacy information about the results of their actions and are stimu-
by testing and using their skills’’ (Hackman and Oldham lated to pursue different work routes, they learn and obtain
1980, p. 78). Consequently, varied jobs stretch the skills and a better understanding of their jobs, and this fosters their
abilities of employees. Therefore, apart from impacting on domain-relevant skills. We thus propose the following:
employee task motivation, job variety also affects the
domain-relevant skills of employees. This further induces an Hypothesis 4: There is a positive association between task
increase in creativity, as there is a strong relationship between feedback and employee creativity.
domain proficiency and creativity (Amabile 1996). We thus
propose the following. Task significance. Significance is the extent to which the
employee perceives the job to make a substantial contribution
Hypothesis 2: There is a positive association between task to the organization or other people (Hackman and Oldham,
variety and employee creativity. 1980). When the work that is undertaken affects the health or
happiness of other people, employees will care more about it.
Task identity. Identity concerns the extent to which employ- Significance thus increases intrinsic motivation and, therefore,
ees complete a whole and identifiable piece of work and can according to the componential model of creativity, drives crea-
identify the results of their efforts (Hackman and Oldham, tive performance. We thus propose the following:
1980). Similar to job variety, identity also promotes the feeling
that the job is meaningful and worthwhile, increasing intrinsic Hypothesis 5: There is a positive association between task
motivation and, thus, employee creativity. However, identity significance and employee creativity.
should also promote the domain-relevant skills of frontline ser-
vice employees. In a service context, frontline employees with
jobs high on identity are responsible for tasks from ‘‘prospect-
Interactive Effects
ing through post sale-service’’ (Becherer, Morgan, and Richard Interactive effects between job characteristics have rarely been
1982, p. 127). Consequently, employees deal with customers considered in the literature, one of the rare examples being
across the service delivery process and are able to develop a Dodd and Ganster (1996) who, based on a review, identify both
better understanding of customers’ needs. By improving theoretical and empirical support for considering such effects.
knowledge about customers and about the service delivery pro- Theoretical support is provided by the Motivating Potential
cess, identity increases the domain-relevant skills of employ- Score (MPS) index proposed by Hackman and Oldham
ees, thus fostering their creativity. (1980), which has frequently been used to assess job complex-
In fact, creative skills involve the ability to be flexible and ity (e.g., Oldham and Cummings 1996). The score is computed
imaginative in approaching problems (Amabile 1998). By as follows: MPS ¼ Autonomy  Feedback  (Variety þ Iden-
developing a more complete understanding of customers’ tity þ Significance)/3. This formulation clearly postulates that
needs and of the stages of the service delivery process, employ- job characteristics interact with each other to affect attitudinal
ees will have an increased likelihood to explore and transform and behavioral job outcomes. Dodd and Ganster (1996) also
conceptual spaces, resulting in the emergence of new associa- derive theoretical support for the interactive effects using the
tions between knowledge structures. They will thus be in a control theory from Carver and Scheier (1981). Accordingly,
better position to take advantage of their creative skills. In Dodd and Ganster reason that feedback, which motivates

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 429

430 Journal of Service Research 13(4)

employees to think about their performance levels and, subse- and by the restricted command over, and comprehension
quently, further strategy development (Campbell 1988), should of, the service delivery process. This adversely affects
not produce such effects in the absence of autonomy, since employee motivation, task-related learning, and, therefore,
employees will be unable to change their behavior. Further sup- creative performance. Likewise, when employees have a job
port comes from Karasek’s (1979) job demands-decision lati- high on identity but low on autonomy, they will not be able
tude model, in which the job decision latitude comprises a to use their knowledge of the service stages and customers’
combination of skill variety and autonomy (Dodd and Ganster, requirements in order to creatively address those needs.
1996). Empirical support for the existence of interactive effects Therefore, creativity should be maximized when both auton-
is provided by Campbell and Gingrich (1986) and Dodd and omy and identity are high. We, therefore, expect the
Ganster (1996). Therefore, sufficient support seems to exist for following:
considering the interactive effects of job characteristics on job
creativity. Hypothesis 6b: The autonomy-employee creativity relation-
In summary, we argue that the potential benefits ship will be stronger when identity is high than when it is
associated with autonomy should be maximized only when low.
the remaining job characteristics are at certain levels. In
addition, we also consider that the effects of the other char- Autonomy and feedback. Feedback allows employees to
acteristics depend on the level of autonomy that employees compare their performance with goals, and therefore deter-
are allowed to exercise. Empirical support for this is pro- mine the need for changes in behavior. In this context, feed-
vided by Dodd and Ganster (1996), who tested whether job back contributes to the subsequent information search and
characteristics interacted with each other to influence per- development of task strategies (Earley et al. 1990). However,
formance and job satisfaction and only obtained significant the lack of autonomy, which curtails the possibility that
interactions between job characteristics when autonomy was employees might adjust their behavior, should reduce
involved. employees’ motivation to think about, and change, their work
conduct (Dodd and Ganster 1996). This, in turn, also reduces
Autonomy and variety. In narrowly focused jobs involving a the extent to which employees learn about their jobs and
small number of skills, employees have control over a recombine knowledge structures. This reduced intrinsic moti-
restricted number of activities. Furthermore, they will have vation and learning should detrimentally affect creativity.
fewer opportunities to manipulate their environments and to Likewise, empowered employees with no or low amounts of
stretch their skills. For employees with autonomy but placed task feedback, will not be able to evaluate the consequences
in jobs with a small variety of tasks, the flexibility and sense of their decisions, and this reduces their learning and the
of freedom arising from autonomy become constrained by the potential for expanding their domain-relevant skills. There-
range of activities commanded by employees, therefore cur- fore, we expect that feedback coupled with autonomy will
tailing the positive effects of autonomy on creativity. Like- motivate employees to think about alternative courses of
wise, employees without autonomy and placed in jobs with action, and to learn with the consequences of their decisions,
a wide variety of tasks, will not be able to accumulate much thereby increasing creativity. In this context, we propose the
knowledge and experience, given that they have to carry out following:
their tasks in externally determined ways. The low autonomy
constrains the extent to which employees in varied jobs can Hypothesis 6c: The autonomy-employee creativity relation-
explore and manipulate their environments and test their ship will be stronger when feedback is high than when it
skills. For higher levels of variety and autonomy, however, is low.
employees have room to maneuver through the different
tasks, enjoying greater flexibility in deciding how to address Autonomy and significance. As significance increases the
the unique needs of customers. This observation accords with intrinsic motivation of employees, an empowered employee
Campbell and Gingrich (1986), who determined that for sim- in a highly significant job will explore more exhaustively new
ple tasks, participating in defining the way to accomplish ideas and solutions to problems, and will feel more excited and
them had no meaningful effect on performance. We thus pro- personally challenged to explore and pursue risk-taking ave-
pose the following: nues. We thus offer the following:

Hypothesis 6a: The autonomy-employee creativity relation- Hypothesis 6d: The autonomy-employee creativity rela-
ship will be stronger when variety is high than when it is tionship will be stronger when significance is high
low. than when it is low.

Autonomy and identity. In a job high on autonomy and low Finally, as the above discussion suggests, enriching jobs is
on identity, the sense of freedom, flexibility, and the poten- likely to fuel creative employee behaviors. However, such jobs
tial for experimentation associated with autonomy is may not be appropriate for every employee. Hackman and
severely hampered by the limited contact with customers Oldham (1980) indicate that employees with low personal

430 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 431

growth needs and with a low knowledge and skill base may feel Measures
overstretched in enriched jobs. This is likely to reduce creative
The items for measuring job characteristics are from the Job
behaviors and produce detrimental results both for employees
Diagnostic Survey (Hackman and Oldham 1980). The five
and for the employer. This is also likely to yield negative out-
items concerning job creativity are from Ganesan and Weitz
comes for customers. (1996), who adapted the scale from Scott and Bruce (1994).
Similar measures have also been used, for example, by Rice
(2006). The questionnaire was built with multiple-item 7-
Method point scales ranging from strongly disagree (1) to strongly
agree (7). Pretests were subsequently conducted with 16 front-
Sample Selection and Data Collection line employees from other health organizations. Based on the
This work investigates the impact of job characteristics interviews, minor adjustments were made to the items, to
on the creativity of frontline service employees. As employ- improve clarity.
ees react to their perceptions rather than to the objective A preliminary data analysis to detect ill-fitting items
characteristics of the work environment, the frontline included item-to-total correlations and exploratory factor
employee is the appropriate unit of analysis, and this is con- analysis, searching for items that were poorly correlated
sistent with previous research (e.g., Dubinsky and Skinner with the remaining items in each scale, and that had
1984; Farmer, Tierney, and Kung-McIntyre 2003; Oldham cross-loadings. Subsequently, the remaining items were sub-
and Cummings 1996). Data were collected with the colla- mitted to a 6-factor confirmatory analysis to further exam-
boration of three hospitals with a total of 2,279 frontline ine the psychometric properties of the scales. The initial
employees. The health industry is an important economic analysis indicated that some items had correlated errors.
sector that is under growing pressure to become more produc- This led to a step-by-step process for refining the measure-
tive, given cost escalation. It also faces mounting budgetary ment model, under which a number of items were elimi-
constraints. Resource-scarce environments create pressure nated. The chi-square of the final model is statistically
to perform effectively (Licata et al. 2003), and consequently significant (w2 ¼ 460.18; df ¼ 155) but the remaining mea-
sures indicate good fit (Goodness of Fit Index [GFI] ¼ .91;
employee creativity becomes more important in such
Tucker-Lewis Index [TLI] ¼ .91; Comparative Fit Index
contexts.
[CFI] ¼ .93; and Root Mean Square Error of Approximation
The frontline employees of the three hospitals received
[RMSEA] ¼ .07). The reliability of each scale equals or
a pack containing a cover letter, a self-administered ques-
exceeds the .8 threshold, except for task identity, which
tionnaire, and a stamped self-addressed envelope. To max-
has a coefficient alpha of .68. The composite reliability of
imize the response rate, these employees could either post
each scale also exceeds the .8 level, except for task identity
the stamped self-addressed envelope or deposit it in a bal-
(.68). As to the average variance extracted, the scales meet
lot box that was left in a central location in each of the
the .5 target, the exceptions being task identity (.42) and
three hospitals. Of the 2,279 frontline employees that were
creativity (.48). Overall, these results provide evidence
surveyed, 525 responded to the questionnaire. Due to miss-
that the scales are internally consistent. The results also
ing data, 65 surveys were eliminated, thus obtaining 460
support the unidimensionality and convergent validity of the
usable questionnaires, a 20.2% net response rate. The sam-
scales, as all items load on their specified constructs, with
ple is 64% female and 58% are between 31 and 50 years
each standardized loading exceeding 0.5 (but one) and being
old. The sample includes nurses, doctors, health techni-
highly significant, with all t statistics above 10. To
cians, administrative staff, and other job categories of
test for discriminant validity, we compared the square of the
frontline employees. The most frequent respondents are
correlation coefficient with the average variance extracted
nurses (55%) and doctors (15%), and this is broadly in line
for each pair of variables, and observed that in all
with the composition of the staff in the three hospitals.
cases, the average variance extracted was larger than the
This diversity of jobs contributes to the generalization of
respective squared correlation. This provides evidence of
the findings. In addition, focusing on a narrow spectrum
discriminant validity (Fornell and Larcker 1981). Therefore,
of job positions might diminish the variability of
the results support the psychometric properties of the scales.
responses, and this could reduce the capacity to detect
The Appendix presents univariate statistics, correlation
associations between variables and, thus, test the study’s
coefficients, Cronbach’s as, as well as details of the scales’
hypotheses. Not surprisingly, other studies have followed
items.
a similar strategy in terms of sample composition (e.g.,
de Jong, de Ruyter, and Lemmink 2004; Peccei and
Rosenthal 2001). In addition to collecting this quantitative Results
data, we also interviewed five frontline service employees,
three nurses and two employees, each working for a differ- Estimations
ent large retail chain, seeking qualitative feedback to fur- This study uses multiple regression analysis to investigate the
ther enlighten the research findings. main and moderating effects of job characteristics on employee

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 431

432 Journal of Service Research 13(4)

Table 1. Results: The Influence of Job Characteristics on Employee one five-way interaction. Therefore, and as a further test to the
Creativity predicted model, we estimated new regression equations with
Model 1 Model 2 the additional interactions (for parsimony, the results are not
shown). To ensure that the new models are hierarchically cor-
Coefficient Coefficient rect (see Aiken and West 1991), we started by adding to the
Unstand. Stand. Unstand. Stand. Hyp. predicted model, all the three-way interactions. The resulting
increase in the R2 was not significant. Subsequently, we added
Constant 5.36 ** 5.33 ** to the latter model all four-way interactions. The resulting
Autonomy .07 .10* .07 .10* H1 (þ) change in the R2 was also not significant. Finally, we added the
Variety .25 .32** .20 .25** H2 (þ) five-way interaction, which was also not significant. Of all the
Identity .13 .16** .17 .21** H3 (þ)
three-way, four-way, and five-way interactions, only one
Feedback .13 .16** .15 .18** H4 (þ)
Significance .02 .03 .02 .04 H5 (þ) obtained significance. Overall, it is evident that there are no
Interactions systematic interactions at higher order levels. These results
Autonomy  .07 .13** H6a (þ) suggest that the predicted model adequately captures the
Variety inter-relationships between job characteristics.
Autonomy  .06 .09* H6b (þ) Before proceeding with hypotheses testing, we investigated
Identity the potential for multicolinearity and heteroskedasticity. The
Autonomy  .05 .08* H6c (þ)
Feedback
largest variance inflation factor (1.5) and condition index
Autonomy  .00 .00 H6d (þ) (2.6) are both below the threshold of 10 and 30, respectively
Significance (Hair et al. 1998). We, hence, conclude that there are no multi-
R2 .295 .322 colinearity problems. In addition, heteroskedasticity was
F 37.96** 23.72** rejected by the White, and the Breusch-Pagan/Cook-
Note. *p < .05.**p < .01 (one-tail tests).
Weisberg tests (p > .05).

Hypotheses Testing
creativity. Because higher order terms (the interactions, in this The results obtained indicate that the coefficients are broadly in
case) should only be introduced in equations when they accordance with the majority of the predictions put forward in this
improve significantly the explanation provided by their first- work. Autonomy, variety, identity, and feedback have a positive
order terms (Aiken and West 1991), we first estimated an equa- association with job creativity. This provides support for hypoth-
tion that omitted the interaction terms. The results reveal that eses Hypothesis 1, Hypothesis 2, Hypothesis 3, and Hypothesis 4.
the R2 of this model is 29.5% (Model 1, Table 1). Against predictions, task significance is not related to creative
We subsequently estimated the predicted model. Given the performance, thus not supporting Hypothesis 5. Of the four inter-
presence of multiplicative terms in the equation, the variables actions, three are statistically significant. The coefficient for the
are centered to reduce the resulting nonessential ill- interaction between autonomy and variety is significant and neg-
conditioning multicolinearity (Aiken and West 1991). Table ative, and this contradicts the positive sign predicted in Hypoth-
1 presents the estimation of the predicted model (Model 2). esis 6a. The interactions associated with identity and feedback
Overall, the results indicate that the predicted model explains are significant and positive. Thus, and according to the predic-
a significant amount of variance (R2 ¼ 32.2%). Therefore, the tions of Hypothesis 6b and Hypothesis 6c, autonomy interacts
addition of the interaction terms resulted in an improvement in with both identity and feedback in such a way that creativity is
the explanation of creativity that amounts to 2.7%, an improve- highest for higher levels of the variables involved in each interac-
ment which is statistically significant (p < .01). tion. Task significance does not interact with autonomy. There-
Subsequently, a new regression equation was estimated solely fore, no support is obtained for Hypothesis 6d.
with the MPS, which was formed according to the procedure sug- To facilitate the interpretation of the interactions, these are
gested by Hackman and Oldham (1980). This is an aggregate depicted graphically using the procedure suggested by Aiken
measure for job complexity that has been used in several studies and West (1991). Figure 2A presents the complex relationship
(e.g., Oldham and Cummings 1996). In this separate estimation, of how autonomy and variety interact to affect creative perfor-
sole consideration of the composite measure of job complexity mance. When variety is small, the more employees are auton-
explains 22% of the variation in employee creativity. Thus, by omous, the more they are creative. However, as indicated in
considering the separate effects of each job characteristic and Figure 2A, when variety is high, an increase in autonomy pro-
including the moderating effects, the explanatory power duces no effect on creativity. Figure 2B and C also indicate that
increased from 22% to 32.2%, an improvement of about 46%. there is a positive relationship between autonomy and creative
The model in this study only predicts two-way interactions performance when employees have a job high in identity and
between job characteristics. However, consideration of a disag- feedback. However, in the presence of jobs low in identity and
gregated model of job characteristics enables the testing of feedback, increasing the level of autonomy does not produce
three-way interactions, along with four-way interactions, and much effect on creativity.

432 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 433

their study, that autonomy plays a less prominent role in fuelling

(A)
creativity. However, the positive link between autonomy and crea-
High variety
tivity obtained in this study reinforces the idea that to be creative,
employees need some degree of discretion over the execution of
their tasks in order to produce creative outcomes (Shalley and
Low variety
Creativity

Gilson 2004). The employees we interviewed stressed the impor-

tance of having a voice in how things were achieved, otherwise, no
one would venture into supplying new ideas.
Furthermore, the results indicate that the influence of auton-
omy on creative performance is further exercised through the
interactive effects with other characteristics. When variety is
Low autonomy High autonomy low, autonomy is positively related with creativity. However,
for high levels of variety, the effect of autonomy on creativity
becomes irrelevant. It is possible that a varied job might be
(B) rather stimulating and enable the development of competen-
cies, given the diversity of skills that are exercised in such jobs,
High identity regardless of autonomy.
Autonomy, when accompanied by high identity, exerts a
positive effect on creativity. However, the link between auton-
Creativity

omy and creativity becomes irrelevant in the presence of low

task identity. Low identity implies limited contact with custom-
Low identity
ers and partial knowledge of their needs and of the service
delivery process. Consequently, the extent to which autonomy
can be exercised and result in the expansion of domain-relevant
Low autonomy High autonomy skills becomes severely restricted, thus eroding its effects on
creativity. Similarly, when autonomy is low, a high identity
should not have much effect on creativity, given employees’
(C) limited capacity to use their customer knowledge to address the
unique needs of customers.
The study indicates that autonomy only positively influ-
High feedback
ences creativity in jobs rich in feedback. With autonomy,
employees view their work outcomes as the result of their own
Creativity

efforts. The lack of feedback, however, implies that employees

have difficulties in identifying the results of their work. With-
Low feedback out this knowledge, employees have no stimuli or directions to
guide their behavior and cannot learn through the process of
trial and error. This curtails their intrinsic motivation and learn-
ing, impedes the expansion of domain-relevant skills, and
Low autonomy High autonomy builds up their frustration.
Regarding identity and feedback, the discussion with the
Figure 2. Moderating effects. A, Interaction between autonomy and employee working for the retail clothing chain demonstrated
variety. B, Interaction between autonomy and identity. C, Interaction the following situation to occur:
between autonomy and feedback.
Accompanying the customer to the fitting room provides fur-
Discussion and Implications ther feedback about customers’ needs and feelings, improving
employees’ knowledge about the customer. This additional
Discussion of Hypotheses Testing
feedback frequently led the employee to seek new pieces of
Consistent with our hypotheses, autonomy, variety, identity, and clothing or combinations. Furthermore, failing to accompany
feedback have positive main effects on creativity. Task signifi- the customer to the fitting room frequently leads to losing sales,
cance, however, has no significant main effect on creativity. This as unaccompanied customers often simply leave the store when
indicates that each job characteristic tends to have its own effects the clothing they have tried on does not fit.
on creativity. This is not surprising, as each characteristic involves
different cognitive processes, affecting creativity in unique ways One of the nurses working for a local health service explained that:
through intrinsic motivation and domain-relevant skills.
Of the five job characteristics, autonomy has been the individ- Upon the reorganization of the local health service, the nurse
ual characteristic most studied. Amabile et al. (1996) conclude, in started following a set of patients across different health issues,

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 433

434 Journal of Service Research 13(4)

including, for example, diabetes, pregnancy, and vaccination. least when it comes to explaining the creativity of frontline
As a result, this nurse accumulated greater knowledge of each service employees, with autonomy interacting with variety,
patient’s needs and more feedback, namely through the larger identity, and feedback. Consequently, it seems appropriate for
number of interactions with each patient. Consequently, she academics and managers to consider how changes in one job
now feels much more equipped to serve the specific needs of characteristic might interplay with other job characteristics to
each patient. influence the psychological and behavioral responses of
employees. The approach of this study also enables the identi-
This study further shows that task significance is not related to fication of each job characteristic’s relative importance. Anal-
creativity. Significance has an intrinsic motivation link with ysis of the standardized effects indicates that the five job
creativity, which is shared with the other characteristics of jobs. characteristics are not equally important drivers of creativity,
This redundancy possibly explains the lack of significant and this has clear implications for managers.
results. Another possible explanation relates to the hospital Consequently, promoting the creativity of frontline employ-
context in which this study is conducted. Most of the surveyed ees should have important payoffs (cf. Wang and Netemeyer
jobs have implications in the survival or well-being of patients. 2004). In services, employee creativity can be exhibited in the
This suggests that in samples with a greater diversity in terms discovery of the latent needs of customers, in crafting and deli-
of job significance, this characteristic might have a more sub- vering the service that addresses those needs, in communicat-
stantive role. ing with customers, in building customer rapport, and in
Finally, the results indicate that autonomy has the lowest dealing with complaints. These creative efforts should be par-
main effect compared to the remaining significant main effects ticularly relevant in personal and customizable services, due to
of variety, identity, and feedback. There are two possible expla- the closer and more intensive interactions between customers
nations for this. One is that this small effect is caused by the and service providers.
presence of autonomy in several interactions. The other is that, Notwithstanding, employee creativity should also have
following Bowen and Lawler (1992), different people may some importance in services directed at possessions, or that
react differently to autonomy, thus reducing the explanatory are more standard. One reason is that the experience of con-
power of the variable. Bowen and Lawler (1992, p. 39) advance sumers in such services also depends, to some extent, on the
that a positive response to empowerment will take place to the interaction between the customer and the service provider. In
extent to which employees ‘‘have strong needs to grow and to this respect, Brown and Lam (2008, p. 243), state that ‘‘high
deepen and test their abilities—at work.’’ quality interactions with service providers often result in cus-
In summary, the results of this study indicate that, in a hos- tomer satisfaction despite problems with other aspects of ser-
pital context, in order to build a social environment that stimu- vice delivery, whereas dissatisfying experiences with service
lates creative performance, it is important to find the right providers have the potential to ruin otherwise pleasant service
balance between job characteristics. encounters.’’ The employee working in the electronics
department (selling hardware and software) of a large retail
chain dealing with household appliances and electronics
General Discussion commented:
The influence of job characteristics on creativity is estab-
lished. Notwithstanding, by following a component-wise Each working day is an adventure. We do the same things every
approach to investigate the effects of job characteristics on day, but serving customers is always a surprise. Each customer
creativity, this work intends to overcome some of the limita- is a different situation, even when they are looking for exactly
tions associated with previous studies. Overall, the results the same product, with the same colours and characteristics . . .
seem to support the major concerns of this research. The no two customers are the same, they are all different, some are
composite approach provides an incomplete account of the nicer than others, some chat more than others, some even joke
effects associated with job characteristics. It obscures the . . . each customer is a surprise.
possibility of each characteristic exerting a specific effect
Frontline employees can also deploy creative efforts in per-
on creative performance. Furthermore, it omits the possibility
forming their ‘‘internal work,’’ from paperwork to coordinated
of effects canceling each other out due to the averaging
actions and mechanisms with other departments and to sugges-
approach. The component-wise perspective yielded an expla-
tions for new services and procedures. Creative efforts in this
natory power considerably above that associated with the
area are likely to take place regardless of the type of service
MPS. This result demonstrates the relevance of considering
and/or job. From the discussion with a frontline employee
job design in a disaggregated fashion.
Another important outcome relates to the interactionist working for an international clothing store chain, the following
emerged:
model. The composite approach of the creativity literature
impedes the consideration of any interaction between job char- Employees at the store frequently complained about colleagues.
acteristics. Furthermore, most empirical studies have assumed Then, one experienced employee came up with the idea of cre-
job characteristics to be independent of one another. This ating a box where each employee could anonymously put in the
research, however, shows that this is not always the case, at reasons for their own dissatisfaction, along with suggestions for

434 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 435

improvements. One of the problems reported was related with 1984). Finally, although our results do not support a relation-
the inexperience of newly hired employees. This store chain ship between task significance and creativity, considering past
hires many part-time employees, frequently university students, evidence regarding the role of task significance, it does not
which originates a high turnover ratio. Employees new to the seem reasonable to neglect this job facet. Developing jobs in
company are usually accompanied by an experienced employee which tasks are grouped into natural units contribute to task
for about 2 weeks. However, this was not enough for them to identity but also to task significance, as employees are more
learn how to manage their work time. At the end of the day the likely to perceive the impact they have on others (Hackman and
sections where newly hired employees worked were not tidied Oldham 1980).
up and so, the others, who were ready to go home, had to help,
Finally, managers should consider that these indications are
and this caused disputes and comments like ‘‘he/she is new, but
not universal assuming, in particular, employees with strong
I am not guilty of that’’ and ‘‘I earn my wage, not his/hers.’’ The
growth needs and with the competencies to deal with job
idea that was implemented in the store involved an experienced
demands (cf. Hackman and Oldham 1980). In addition, firms
employee accompanying each new recruit for a longer period,
and this helped in improving the work environment.
dealing with interpersonal services probably have more to win
from developing complex jobs. Following Bowen and Lawler
From an interview with a nurse working in the operating (1992), who advanced a number of conditions under which the
room for maxillar surgery, and who had a routine job with provision of autonomy (which enriches jobs) to frontline
limited customer interaction, we were able to establish the employees might work best, we further add that the design of
following: complex jobs in order to improve creativity should provide
more benefits for firms pursuing a customized strategy, serving
Most of the working time is spent in the operating room assist- customers who seek a personalized solution, looking for long-
ing the surgery. The day before each surgery, however, it is nec- term relationships with customers, and working in uncertain
essary to talk with the patient for about 10 minutes to obtain and environments.
provide information. Dealing creatively with patients, particu-
larly children, cancer patients, and patients with mental disabil- Limitations and Directions for Future
ities is extremely important. With children, for example, if the
nurse manages to gain their empathy, this helps in inserting the
Research
catheter and the electrodes the following day just before the sur- This study addresses a number of shortcomings in past research
gery, speeding the process and increasing productivity. by considering, using an interactionist model, a component-wise
approach to investigate the effects of job characteristics on the
creative performance of frontline service employees. Neverthe-
less, this contribution must be considered with an appropriate
Implications for Managers understanding of its limitations, which should be addressed in
The results obtained in this study indicate that managers future research. The study considered the highly specific health
working in interpersonal services seeking to promote the sector, and this implies that some of the results may not apply in
satisfaction of customers should design the tasks of their other settings. Consequently, it would be interesting to conduct
frontline employees in a way that increases their creativity. similar research, for example, in other service activities and also
In particular, managers should consider increasing the auton- in R&D-related samples. Nevertheless, the closeness of the find-
omy of their employees, namely by reducing the need to con- ings to general theoretical predictions is noteworthy.
sult superiors, and giving them latitude in deciding about Moreover, the sample considered in this study consists of
work organization and schedules. Variety should also be frontline employees occupying different positions in a hospital
incorporated into job tasks by enlarging the set of activities setting. However, it is possible that employees with different
performed by employees. jobs may respond differently to job characteristics. Therefore,
Managers can also spur creativity by providing task identity, it would be interesting to assess the extent to which employees
that is by designing jobs requiring the employee to complete an with different tasks respond to the same job characteristics.
entire piece of work. This can be accomplished by enabling That said, the advantage of pooling the subjects is that it facil-
employees to follow their customers from the beginning to the itates the generalizability of the results. Not surprisingly, many
end of the service delivery process. This enables employees to studies in the past also considered samples of employees occu-
build a much better knowledge of customers’ needs, which pying a variety of positions (e.g., de Jong, de Ruyter, and Lem-
should influence the development of more creative solutions mink 2004; Hartline and Ferrel 1996; Hartline, Maxham, and
to customer problems. Managers can also fuel creativity by McKee 2000; Peccei and Rosenthal 2001). Moreover, as stated
providing higher task feedback. This enables employees to before, concentrating on a single position would possibly
learn about their jobs and stimulates them to try alternative lower sample variability, therefore reducing the variance of job
courses of action to obtain better results. Feedback can be pro- characteristics, causing hypotheses testing to be virtually
vided by having customers evaluate employees and providing meaningless.
the latter with such information and also through frequent per- The measure of creativity in this study is based on employee
formance appraisals by supervisors (Dubinsky and Skinner perceptions. Some past studies also use this type of measure

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 435

436 Journal of Service Research 13(4)

(e.g., Ganesan and Weitz 1996; Rice 2006). Nonetheless, it is Appendix

worthwhile to ascertain the extent to which the results replicate
with objective measures of creativity. This study also assumes Table A1. Scale Items
a causation mechanism from job design to creativity. This
implies the suitableness of a longitudinal design, but this work Standardized Critical
relies instead on cross-sectional data. Construct Items Loadings Ratio
Another limitation of this research involves the exclusive Employee I try to be as creative as I .81 19.75
utilization of self-reported perceptual measures. Despite being creativity can in my job
commonly used, this approach heightens the likelihood of com- I experiment with new .75 17.91
mon method variance affecting the results. To mitigate this approaches in performing
possibility, and in line with Podsakoff et al. (2003), respon- my job
dents were not told of the specific purpose of the research, and When new trends .55 11.89
develop, I am usually the
the construct items were mixed so that respondents should not
first to get on board
have been able to associate particular items with specific fac- My boss feels that I am .47 10.03
tors. In addition, we ran a single factor confirmatory analysis, creative in performing my
with all items loading on a single common method variance job
factor. The fit indices of the resulting model were unaccepta- On the job I am inventive .82 20.24
ble, indicating that respondents were able to differentiate the in overcoming barriers
constructs, implying that the results should not be much Task autonomy I have many opportunities .89 23.07
for independent thought
affected by common method variance.
and action in my job
Current research on creativity postulates that job character- I have many opportunities .90 23.27
istics, and contextual factors in general, impact on creativity to take the initiative in this
through intrinsic motivation. However, research on creativity job
has neglected this mediating link, testing instead the direct rela- I am encouraged to find .57 12.72
tionship between contextual factors and creative behaviors. solutions to problems
Therefore, future research should investigate the extent to I have a great deal of .59 13.17
control over the pace of
which intrinsic motivation does indeed mediate the effects of
my work
job characteristics on creativity (Shalley, Zhou, and Oldham Task variety I do many different things .80 19.29
2004). In this study, we advanced the argument that job char- in this job
acteristics also impact on creativity by building employee com- I perform different tasks .84 20.56
petencies, but we did not test this mediating effect, and this during a typical work day
should be addressed in the future. This job requires me to .79 19.07
Finally, future research should also further explore the use a number of skills and
talents
potential for interactions between job characteristics. In
Task identity I have many opportunities .52 10.47
this study, such moderating effects are considered in an to complete the work I
employee creativity context, but the rationale underlying started
these interactions can easily be extended to other research In this job I can see the .73 15.30
domains, such as employee role stress and turnover inten- entire piece of work
tions. Furthermore, it would be interesting to investigate I have many opportunities .67 13.87
the extent to which the effects of different job characteris- to do a job from beginning
to end (i.e., the chance to
tics are affected by personal characteristics. It is quite
do a whole job)
likely, for example, that some people will be more tolerant Task feedback I easily identify how well I .74 17.17
concerning the lack of feedback than others. Thus, investi- am doing in the job I am
gation concerning how different cognitive styles and other working on
person-related characteristics moderate the effects of job I can easily ascertain .90 22.17
characteristics on creativity and other employee behaviors whether I am performing
can result in the significant addition of knowledge. well or poorly in this job
I have many opportunities .70 16.01
to find out how well I am
Declaration of Conflicting Interests doing in my job
Task significance My work significantly .78 11.80
The author(s) declared no conflicts of interest with respect to the
affects the lives and
authorship and/or publication of this article.
well-being of other
people
Funding A lot of other people can .94 12.79
be affected by how well
The author(s) received no financial support for the research and/or the work gets done
authorship of this article.

436 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012

Coelho and Augusto 437

Table A2. Descriptive Statistics, Correlation Matrix, Reliability, and Campbell, Donald J. (1988), ‘‘Task Complexity: A Review and Anal-
Variance Extracted Estimates ysis,’’ Academy of Management Review, 13 (1), 40-52.
Campbell, Donald J. and Karl F. Gingrich (1986), ‘‘The Interactive
Stand.
Dev. X1 X2 X3 X4 X5 X6 CR AVE Effects of Task Complexity and Participation on Task Perfor-
mance,’’ Organizational Behavior and Human Decision Pro-
Identity 0.76 .68 .68 .42 cesses, 38 (2), 40-52.
(X1) Carver, Charles S. and Michael F. Scheier (1981), Attention and Self-
Variety 0.95 .47 .85 .85 .66
Regulation: A Control-Theory Approach to Human Behavior. New
(X2)
Feedback 0.91 .62 .42 .81 .82 .61 York: Springer-Verlag.
(X3) Chebat, Jean-Charles and Paul Kollias (2000), ‘‘The Impact of
Autonomy 1.20 .56 .36 .51 .82 .83 .57 Empowerment on Customer Contact Employees’ Roles in Service
(X4) Organizations,’’ Journal of Service Research, 3 (1), 66-81.
Significance 1.48 .14 .26 .01 .04 .84 .85 .75 Conger, Jay A. and Rabindra N. Kanungo (1988), ‘‘The Empower-
(X5) ment Process: Integrating Theory and Practice,’’ Academy of Man-
Creativity 0.77 .53 .53 .45 .40 .08 .80 .82 .48
agement Review, 13 (3), 471-82.
(X6)
Day, George S. (1994), ‘‘The Capabilities of Market-Driven Organiza-
Note. CR ¼ composite reliability; AVE ¼ average variance extracted. Diagonal tions,’’ Journal of Marketing, 58 (4), 37-52.
entries in boldface are Cronbach’s a coefficients. de Jong, Ad, Ko de Ruyter, and Jos Lemmink (2004), ‘‘Antecedents
and Consequences of the Service Climate in Boundary-Spanning
Self-Managing Teams,’’ Journal of Marketing, 68 (2), 18-35.
References Dodd, Nancy G. and Daniel C. Gangster (1996), ‘‘The Interactive
Aiken, Leona S. and Stephen G. West (1991), Multiple Regression: Effects of Variety, Autonomy, and Feedback on Attitudes and Per-
Testing and Interpreting Interactions. London: SAGE. formance,’’ Journal of Organizational Behavior, 17 (4), 329-347.
Amabile, Teresa M. (1998), ‘‘How to Kill Creativity: Keep Doing Dubinsky, Alan J., Roy D. Howell, Thomas N. Ingram, and Danny N.
what you’re Doing or, if you want to Spark Innovation, Rethink Bellenger, (1986), ‘‘Sales Force Socialization,’’ Journal of Market-
how you Motivate, Reward, and Assign Work to People,’’ Harvard ing, 50 (4), 192-207.
Business Review, 76 (5), 77-87. Dubinsky, Alan J. and Steven J. Skinner (1984), ‘‘Impact of Job
Amabile, Teresa M. (1996), Creativity in Context. Boulder, CO: West Characteristics on Retail Salespeople’s Reactions to their Jobs,’’
View Press Inc. Journal of Retailing, 60 (2), 35-62.
Amabile, Teresa M., Regina Conti, Heather Coon, Jeffrey Lazenby, Earley, P. Cristopher, Gregory B Northeraft, Cynthia Lee, and Terri R.
and Michael Herron (1996), ‘‘Assessing the Work Environment for Lituchy (1990), ‘‘Impact of Process and Outcome Feedback on the
Creativity,’’ Academy of Management Journal, 39 (5), 1154-1184. Relation of Goal Setting to Task Performance,’’ Academy of Man-
Amabile, Teresa M. (1983), ‘‘Social Psychology of Creativity: A agement Journal, 33 (1), 87-105.
Componential Conceptualization,’’ Journal of Personality and Farmer, Steven M., Palmela Tierney, and Kate Kung-McIntyre (2003),
Social Psychology, 45 (2), 357-376. ‘‘Employee Creativity in Taiwan: An Application of Role Identity
Auh, Seigyoung and Bulent Menguc (2005), ‘‘Top Management Team Theory,’’ Academy of Management Journal, 46 (5), 618-630.
Diversity and Innovativeness: The Moderating Role of Interfunc- Fornell, Claes and David F. Larcker (1981), ‘‘Evaluating Structural
tional Coordination,’’ Industrial Marketing Management, 34 (3), Equation Models with Unobservable Variables and Measurement
249-261. Error,’’ Journal of Marketing Research, 18 (1), 39–50.
Baker, William E. and James M. Sinkula (1999), ‘‘The Synergistic Frese, Michael, Harry Garst, and Dories Fay (2007), ‘‘Making Things
Effect of Market Orientation and Learning Orientation on Organi- Happen: Reciprocal Relationships Between Work Characteristics
zational Performance’’ Journal of Academy of Marketing Science, and Personal Initiative in a Four-Wave Longitudinal Structural Equa-
27 (4), 411-427. tion Model,’’ Journal of Applied Psychology, 92 (4), 1084-1102.
Becherer, Richard C., Fred W. Morgan, and Lawrence M. Richard Ganesan, Shankar and Barton Weitz (1996), ‘‘The Impact of Staffing
(1982), ‘‘The Job Characteristics of Industrial Salespersons: Rela- Policies on Retail Buyer Job Attitudes and Behaviors,’’ Journal of
tionships to Motivation and Satisfaction,’’ Journal of Marketing, Retailing, 72 (1), 31-56.
46 (4), 125-135. George, Jennifer M. and Jing Zhou (2002), ‘‘Understanding When Bad
Bitner, Mary J., Booms H. Booms, and Mary S. Tetreault (1990), Moods Foster Creativity and Good Ones Don’t: The Role of Context and
‘‘The Service Encounter: Diagnosing Favorable and Unfavorable Clarity of Feelings,’’ Journal of Applied Psychology, 87 (4), 687-697.
Incidents,’’ Journal of Marketing, 54 (1), 71-84. Hackman, J. R. and Greg R. Oldham (1980), Work Redesign. London:
Bowen, David E. and Edward E. Lawler (1992), ‘‘The Empowerment Addison-Wesley.
of Service Workers: What, Why, and When,’’ Sloan Management Hackman, J. R. and Greg R. Oldham (1975), ‘‘Development of the Job
Review, 33 (3), 31-39. Diagnostic Survey,’’ Journal of Applied Psychology, 60 (2), 159-170.
Brown, Steven P. and Son K. Lam (2008), ‘‘A Meta-Analysis of Rela- Hair, Joseph F. Jr., Rolph E. Anderson, Ronald L. Tatham, and
tionships Linking Employee Satisfaction to Customer Responses,’’ William C. Black (1998), Multivariate Data Analysis. Englewood
Journal of Retailing, 84 (3), 243-255. Cliffs, NJ: Prentice Hall.

Downloaded from jsr.sagepub.com by joao correa on August 25, 2012 437

438 Journal of Service Research 13(4)

Hartline, Michael D., James G. Maxham, and Daryl O. McKee (2000), Shalley, Christina E. (1995), ‘‘Effects of Coaction, Expected Evalua-
‘‘Corridors of Influence in the Dissemination of Customer- tion, and Goal Setting on Creativity and Productivity,’’ Academy of
Oriented Strategy to Customer Contact Service Employees,’’ Jour- Management Journal, 38 (2), 483-503.
nal of Marketing, 64 (4), 35-50. Sims, Henry P., Andrew D. Szilagyi, and Robert T. Keller (1976),
Hartline, Michael D. and O. C. Ferrell (1996), ‘‘The Management of ‘‘The Measurement of Job Characteristics,’’ Academy of Manage-
Customer-Contact Service Employees: An Empirical investiga- ment Journal, 19 (2), 195-212.
tion,’’ Journal of Marketing, 60 (4), 52-70. Singh, Jagdip (1998), ‘‘Striking a Balance in Boundary-Spanning
Hatcher, Larry, Timothy L. Ross, and Denis Collins (1989), ‘‘Proso- Positions: An Investigation of Some Unconventional Influences
cial Behavior, Job Complexity, and Suggestion Contribution of Role Stressors and Job Characteristics on Job Outcomes of
Under Gainsharing Plans,’’ Journal of Applied Behavioral Science, Salespeople,’’ Journal of Marketing, 62 (3), 69-86.
25 (3), 231-248. Slater, Stanley F. and John C. Narver (1995), ‘‘Market Orientation and
Jaworski, Bernard and Ajay Kohli (1993), ‘‘Market Orientation: Ante- the Learning Organization,’’ Journal of Marketing, 59 (3), 63-74.
cedents and Consequences,’’ Journal of Marketing 57 (3), 53-70. Tierney, Pamela and Steven M. Farmer (2004), ‘‘The Pygmalion Pro-
Jonge, Jan De and Wilmar B. Schaufeli (1998), ‘‘Job Characteristics cess and Employee Creativity,’’ Journal of Management, 30 (3),
and Employee Well-Being: A Test of Warr’s Vitamin Model in 413-432.
Health Care Workers Using Structural Equation Modelling,’’ Jour- Tierney, Pamela and Steven M. Farmer (2002), ‘‘Creative Self-
nal of Organizational Behavior, 19 (4), 387-407. Efficacy: Potential Antecedents and Relationship to Creative Per-
Karasek, Robert A. Jr. (1979), ‘‘Job Demands, Job Decision Latitude, formance,’’ Academy of Management Journal, 45 (6), 1137-1148.
and Mental Strain: Implications for Job Redesign,’’ Administrative Tyagi, Pradeep K. (1985), ‘‘Relative Importance of Key Job Dimen-
Science Quarterly, 24 (2), 285-308. sions and Leadership Behaviors in Motivating Salesperson Work
Kennedy, Karen N., Jerry R. Goolsby, and Eric J. Arnould (2003), Performance,’’ Journal of Marketing, 49 (3), 76-86.
‘‘Implementing a Customer orientation: Extension of theory and Van Mierlo, Heleen, Christee G. Rute, J. K. Vermunt, M. A. J.
application,’’ Journal of Marketing, 67 (4), 67-81. Kompier, and J. A. C. M. Doorewaard (2007), ‘‘A Multi-Level
Licata, Jane W., John C. Mowen, Eric G. Harris, and Tom J. Brown Mediation Model of the Relationships Between Team Autonomy
(2003), ‘‘On the Trait Antecedents and Outcomes of Service Individual Task Design, and Psychological Well Being,’’ Journal
Worker Job Resourcefulness: A Hierarchical Model Approach,’’ of Occupational and Organizational Psychology, 80 (4), 647-664.
Journal of the Academy of Marketing Science, 31 (3), 256-271. Wang, Guangping and Richard G. Netemeyer (2004), ‘‘Salesperson
Mumford, Michael D. (2003), ‘‘Where Have We Been, Where Are We Creative Performance: Conceptualization, Measurement, and
Going? Taking Stock in Creativity Research,’’ Creativity Research Nomological Validity,’’ Journal of Business Research, 57 (8),
Journal, 15 (2/3), 107-20. 805-812.
Oldham, Greg R. and Anne Cummings (1996), ‘‘Employee Creativity: Wood, Robert and Albert Bandura (1989), ‘‘Impact of Conceptions of
Personal and Contextual Factors at Work,’’ Academy of Manage- Ability on Self-regulatory Mechanisms and Complex Decision
ment Journal, 39 (3), 607-634. Making,’’ Journal of Personality and Social Psychology, 56 (3),
Peccei, Riccardo and Patrice Rosenthal (2001), ‘‘Delivering Cus- 407-415.
tomer-Oriented Behaviour Through Empowerment: An Empirical Woodman, Richard W., John E. Sawyer, and Ricky W. Griffin (1993),
Test of HRM Assumptions,’’ Journal of Management Studies, 38 ‘‘Toward a Theory of Organizational Creativity,’’ Academy of
(6), 831-857. Management Review, 18 (2), 293-321.
Podsakoff, Philip M., Scott B. MacKenzie, Jeong-Yeon Lee, and Zeithaml, Valarie, Mary J. Bitner, and Dwayne D. Gremler (2006),
Nathan P. Podsakoff (2003), ‘‘Common Method Biases in Beha- Services Marketing. London: McGraw-Hill.
vioral Research: A Critical Review of the Literature and Recom- Zhou, Jing (2003), ‘‘When the Presence of Creative Workers is
mended Remedies,’’ Journal of Applied Psychology, 88 (5), Related to Creativity: Role of Supervisor Close Monitoring,’’ Jour-
879-903. nal of Applied Psychology, 88 (3), 413-422.
Rice, Gillian (2006), ‘‘Individual Values, Organizational Context, and Zhou, Jing and Jennifer M. George (2001), ‘‘When Job Dissatisfaction
Self-Perceptions of Employee Creativity: Evidence from Egyptian Leads to Creativity: Encouraging the Expression of Voice,’’ Acad-
Organizations,’’ Journal of Business Research, 59 (2), 233-241. emy of Management Journal, 44 (4), 682-696.
Scott, Susanne G. and Reginald A. Bruce (1994), ‘‘Determinants of
Innovative Behaviour: A Path Model of Individual Innovation in
the Workplace,’’ Academy of Management Journal, 37 (3), Bios
580-607. Filipe Coelho is assistant professor and his current research
Shalley, Christina E. and Lucy L. Gilson (2004), ‘‘What Leaders Need interests include the behavior of frontline employees, namely
to Know: A Review of Social and Contextual Factors that Can Fos- their creativity, customer orientation, and psychological con-
ter or Hinder Creativity,’’ The Leadership Quarterly, 15, 33-53. tracts.
Shalley, Christina E., Jing Zhou, and Greg R. Oldham (2004), ‘‘The
Effects of Personal and Contextual Characteristics on Creativity: Mário Augusto is assistant professor and his research interests
Where Should we go from Here?’’ Journal of Management, 30 are in the field of firm performance and in the interface
(6), 933-958. between marketing and finance.

438 Downloaded from jsr.sagepub.com by joao correa on August 25, 2012