International Journal of Hospitality Management 32 (2013) 132–140

Contents lists available at SciVerse ScienceDirect

International Journal of Hospitality Management

journal homepage: www.elsevier.com/locate/ijhosman

High-performance work practices and hotel employee performance:

The mediation of work engagement
Osman M. Karatepe ∗
Faculty of Tourism, Eastern Mediterranean University, Gazimagusa, Turkish Republic of Northern Cyprus, Via Mersin 10, Turkey

a r t i c l e i n f o a b s t r a c t

Keywords: This study proposes and tests a research model that investigates whether work engagement functions
High-performance work practices as a mediator of the effects of high-performance work practices (HPWPs) on job performance and extra-
Hotel employees role customer service. These relationships were assessed through LISREL 8.30 using structural equation
Performance outcomes
modeling (SEM). Data were obtained from full-time frontline hotel employees and their managers in the
Romania
Poiana Brasov region in Romania. The results suggest that work engagement acts as a full mediator of the
Work engagement
effects of HPWPs on job performance and extra-role customer service. Specifically, HPWPs, as manifested
by frontline employees’ appraisal of training, empowerment, and rewards, enhance work engagement.
Such HPWPs in turn trigger job performance and extra-role customer service. Implications of the results
are discussed and avenues for future research are offered.
© 2013 Elsevier Ltd. All rights reserved.

1. Introduction vigor, dedication, and absorption” (Schaufeli et al., 2002, p. 74).

Vigor, dedication, and absorption are the three dimensions of work
The recognition that frontline employees play a pivotal role in engagement. In Schaufeli et al.’s (2002) study, vigor is defined as
delivery of service quality, successful service recovery, and reten- “high levels of energy and mental resilience while working, the
tion of satisfied and loyal customers in the hospitality industry willingness to invest efforts in one’s work, and persistence even in
is prevalent among researchers and practitioners (e.g., Chi and the face of difficulties”, while dedication refers to “a sense of signifi-
Gursoy, 2009; Chiang and Birtch, 2008; Guchait et al., 2012). There- cance, enthusiasm, inspiration, pride, and challenge” (p. 74). Finally,
fore, retention of frontline employees who can have high quality absorption refers to “being fully concentrated and deeply engrossed
performance in the workplace is as important to business suc- in one’s work, whereby time passes quickly and one has difficul-
cess as customer loyalty and profitability. The presence of HPWPs ties with detaching oneself from work” (p. 75). As stated by Bakker
sends powerful signals to employees that the organization pays (2011), engaged employees are more likely to work harder through
utmost attention to human capital for improving service capacity elevated levels of discretionary efforts when compared to those
and productivity (Tang and Tang, 2012). Internal career/promotion who are disengaged. However, it should be noted that employees
opportunities, teamwork, employment security, staffing selectiv- cannot always be engaged in their work; they need some time and
ity, work-family balance, training, empowerment, and rewards are opportunities for recovery (Bakker, 2011). Otherwise, employees
among the indicators of HPWPs (e.g., Cho et al., 2006; Kusluvan who are always engaged in their work cannot devote their scarce
et al., 2010; Murphy and Murrmann, 2009; Tang and Tang, 2012). resources (e.g., time) to family domain and experience work-family
Frontline employees working in environments where there are a conflict (Halbesleben et al., 2009).
number of HPWPs may have high levels of energy and feel ded- Overall, it is important to understand how HPWPs influence
icated and may be happily immersed in their work. Retention of employees’ motivation to display elevated levels of performance in
engaged employees is critical, because they are likely to display the workplace. By doing so, it would be possible to analyze the black
high quality performance in the workplace and fulfill their formal box stage between HPWPs and performance outcomes in strate-
role requirements for business success (cf. Bakker and Demerouti, gic human resource management research (Boselie et al., 2005;
2008; DiPietro and Pizam, 2008; Slåtten and Mehmetoglu, 2011). Takeuchi et al., 2007; Tang and Tang, 2012). Social exchange theory
As a motivational construct, work engagement refers to “a pos- (SET) provides useful guidelines for explaining the aforemen-
itive, fulfilling, work-related state of mind that is characterized by tioned relationships. Specifically, employees receiving economic
and socioemotional resources from the organization tend to feel
obliged to repay the organization through work engagement
∗ Tel.: +90 392 630 1116; fax: +90 392 365 1584. and better performance (Cropanzano and Mitchell, 2005; Saks,
E-mail address: osman.karatepe@emu.edu.tr 2006).

0278-4319/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijhm.2012.05.003
 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140 133

1.1. Purpose and employees) abide by certain ‘rules’ of exchange (Cropanzano

and Mitchell, 2005). Economic and social exchange relationships
Using SET as the theoretical framework, this study proposes and are the two types of these relationships in the workplace. Eco-
tests a research model that investigates whether work engagement nomic relationships are tangible and appear to be more short term,
mediates the effects of HPWPs on job performance and extra-role while social exchange relationships are related to “close personal
customer service. Training, empowerment, and rewards are the attachment and open-ended obligations” (Cropanzano et al., 2003,
indicators of HPWPs examined in the present study. Data gath- p. 161). When the organization takes care of employees, there are
ered from the hotel industry in Romania were used to assess these social exchange relationships that give rise to positive job or orga-
relationships. nizational outcomes (Cropanzano and Mitchell, 2005). Employees
This study makes several contributions to the existing knowl- who obtain economic and socioemotional resources in the work-
edge base in the following ways. First, although there are various place tend to feel obligated to repay the organization via various
indicators of HPWPs, training, empowerment, and rewards are con- ways.
sidered as the indicators of HPWPs predicting work engagement. Work engagement and positive job outcomes are among the
In other words, HPWPs are manifested through management’s ways for employees to repay the benefits given by the organiza-
simultaneous emphasis on training, empowerment, and rewards. tion (Karatepe, 2011; Saks, 2006). That is, employees devote their
As discussed by Takeuchi et al. (2007), training and empowerment cognitive, emotional, and physical resources to work roles by hav-
signal that employees are regarded as strategic partners for the sur- ing high levels of energy, being enthusiastic about their work, and
vival and success of the organization and are recognized of their being fully engrossed in their work. When employees have HPWPs
value to the organization. The presence of appropriate rewards that are manifested by training, empowerment, and rewards, they
motivates employees to deal with customer requests and prob- repay the organization via their level of work engagement. It seems
lems effectively (Babakus et al., 2003). A careful examination of that if employees believe that the availability of HPWPs emerges
the current literature suggests that training, empowerment, and from management’s commitment to service excellence that is con-
rewards are in Pfeffer’s (1994) well-known list of effective prac- sistent with the organizational mission, they are more engaged in
tices for managing people and have been reported in the top six their work and contribute more to service delivery process. In addi-
human resource practices (Boselie et al., 2005). Such HPWPs have tion, HPWPs improve the quality of the social exchanges between
been shown among the most important indicators of management the employer and employees (Takeuchi et al., 2007; Tang and Tang,
commitment to service quality (Ashill et al., 2008; Babakus et al., 2012). In short, the presence of HPWPs creates a work environ-
2003; Kim et al., 2009). Training, empowerment and/or rewards ment where engaged employees seem to have more trusting and
have also been considered to be the useful managerial implica- high quality relationships with their employer and report better
tions for business practice in the hospitality industry (Chiang and performance outcomes.
Birtch, 2011; Henry et al., 2004; Karatepe and Olugbade, 2009;
Karatepe and Uludag, 2007; Kusluvan et al., 2010). Therefore, this 2.2. HPWPs and work engagement
study investigates the effects of HPWPs, as manifested by training,
empowerment, and rewards, on work engagement. Training, empowerment, and rewards are the indicators of
Second, in a recent meta-analytic study, Christian et al. (2011) HPWPs used in this study. Such indicators are important in frontline
discuss that little is known about work engagement as a deter- service jobs and are crucial to organizational success in the hospital-
minant of performance outcomes. They also discuss that there ity industry. As stated before, training, empowerment, and rewards
is a need for examining work engagement that may simultane- are among the most important indicators of HPWPs (Boselie et al.,
ously lead to in-role and extra-role performances. Consistent with 2005; Pfeffer, 1994). They have also been shown to be among the
Christian et al.’s (2011) suggestion, this study tests job performance most critical indicators of management commitment to service
and extra-role customer service as performance outcomes of work quality (e.g., Kim et al., 2009).
engagement. Frontline employees should acquire technical and interpersonal
Third, empirical research pertaining to the antecedents and con- skills to have solid knowledge about service delivery process and
sequences of work engagement in frontline service jobs is still deal with customer requests and complaints. The lack of ongo-
scarce (Karatepe, 2011; Slåtten and Mehmetoglu, 2011). With this ing and effective training programs in hospitality firms leads to
realization, work engagement is treated as a full mediator of the a pool of employees who do not have the requisite skills or are
impacts of HPWPs on job performance and extra-role customer not willing to respond to customer requests and problems. As
service for addressing the aforementioned voids in the existing another indicator of HPWPs, empowerment refers to “the free-
knowledge base. dom and ability to make decisions and commitments” (Forrester,
By analyzing the black box stage between HPWPs and per- 2000, p. 67). Empowered frontline employees can fulfill their duties,
formance outcomes, the findings of this study may prove useful use their creative personal judgment, and provide quick and fair
insights pertaining to the management of HPWPs and retention of responses to the complaining customers (Guchait et al., 2012; Yavas
engaged employees with heightened performance in frontline ser- et al., 2010). However, empowering employees without training
vice jobs. The next section of the article consists of the description programs would not produce the intended outcomes or training
of SET as the theoretical framework. This is followed by research employees without empowerment for effective complaint man-
hypotheses developed based on SET and empirical evidence, and agement would be useless. Training and empowerment should
the research model. Then, discussions of the method and results be accompanied by rewards, because trained and empowered
are given. The article concludes with implications for managers and employees should obtain adequate rewards for serving customers
future research. and dealing with disgruntled customers successfully (Kim et al.,
2009; Yavas et al., 2010).
2. Theoretical framework, hypotheses, and research model Combs et al. (2006) argue that HPWPs such as training, empow-
erment, and rewards enhance employees’ knowledge, skills, and
2.1. SET abilities and motivate them to reach high levels of productivity.
When these HPWPs are used in a coordinated way, they reinforce
SET states that relationships evolve over time into trusting, loyal, and support each other (Combs et al., 2006). Such an assertion
and mutual commitments if the parties (e.g., managers/supervisors is also supported by internal fit. According to internal fit, “their
134 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140

collective effect will be greater than the sum of their individual 2.4. Work engagement as a full mediator
parts” (Wall and Wood, 2005, p. 431). As a result, the joint presence
of training, empowerment, and rewards would create synergy As a motivational construct, work engagement plays a media-
in the organization (Wall and Wood, 2005), trigger employee tor role between HPWPs and performance outcomes. SET presents
engagement (Hughes and Rog, 2008), and lead to retention of a viable theoretical framework for these relationships (Takeuchi
employees in the hospitality industry (Henry et al., 2004). As SET et al., 2007). Specifically, Cropanzano and Mitchell (2005) state,
contends, employees who find that the organization really invests “Social exchange relationships evolve when employers take care
in human resources through training, empowerment, and rewards of employees, which thereby engenders beneficial consequences”
repay the organization via their work engagement. Therefore, the (p. 882). Accordingly, it appears that when the organization
following hypotheses are proposed: takes care of employees through the simultaneous implementa-
tion of training, empowerment, and rewards, employees are more
H1. Frontline employees’ perceptions of HPWPs (as manifested engaged in their work, leading to better performance outcomes
by training, empowerment, and rewards) are related to their work in the workplace. In other words, employees receiving various
engagement. resources/benefits from the organization would be more likely to
feel obliged to repay the organization through greater work engage-
ment, and in turn, display job performance and extra-role customer
service (Karatepe, 2011; Saks, 2006).
2.3. Work engagement and performance outcomes According to the strategic human resource management
perspective, employee performance and organizational or firm per-
Job performance and extra-role customer service are the two formance are influenced by a set of HPWPs (e.g., Huselid et al.,
critical performance outcomes of work engagement examined in 1997). However, HPWPs should motivate employees to perform
this study. There are at least two reasons for investigating these their jobs effectively (Huselid, 1995). Employees who participate in
relationships. First, job performance is defined as “the level of pro- training programs containing the use of empowerment and obtain
ductivity of an individual employee, relative to his or her peers, appropriate levels of pay and related rewards are motivated to
on several job-related behaviors and outcomes” (Babin and Boles, accomplish their work goals. Such employees are engaged in their
1998, p. 82). Employees in frontline service jobs are expected to deal work, and in turn, deliver quality services, deal with customer
with customer requests and problems in a responsive and cour- problems successfully, and willingly go out of their way for satisfy-
teous manner, because service quality perceptions and customer ing customers. Consequently, HPWPs that are manifested through
satisfaction largely depend on frontline employees’ service deliv- management’s simultaneous emphasis on training, empowerment,
ery behaviors (Bettencourt et al., 2005). When engaged frontline and rewards would motivate frontline employees to provide qual-
employees are more vigilant and centered on their tasks, they deal ity services, offer successful complaint resolution, and demonstrate
with customer requests and problems successfully and have high extra-role performance (cf. Babakus et al., 2003; Yavas et al., 2010).
quality performance. This makes sense, because engaged employ- Consistent with the precepts of SET and strategic human resource
ees devote their cognitive, emotional, and physical resources to management perspective, the following hypotheses are proposed:
work roles.
Extra-role customer service refers to “discretionary behaviors of H3. Work engagement fully mediates the effects of HPWPs (as
contact employees in serving customers that extend beyond formal manifested by training, empowerment, and rewards) on (a) job
role requirements” (Bettencourt and Brown, 1997, p. 41). Hospi- performance and (b) extra-role customer service.
tality firms are also in need of frontline employees who would
frequently go out the way to help customers. This is important, 2.5. Research model
because frontline employees should serve customers in challeng-
ing service encounters and show proactive behaviors to be able to The research model that includes the hypothesized relation-
adapt to specific situations (De Jong and De Ruyter, 2004). There- ships is presented in Fig. 1. According to the model, the indicators
fore, engagement can be considered as an important determinant of HPWPs are training, empowerment, and rewards. The model
of frontline employees’ willingness to go beyond their formal role proposes that frontline employees’ perceptions of HPWPs are pos-
requirements for satisfying customers, because engaged employees itively related to their work engagement. The model also suggests
are more willing to exert efforts to go the extra mile in facilitating that work engagement functions as a full mediator of the effects of
customer needs (Moliner et al., 2008). Consequently, it is important HPWPs on job performance and extra-role customer service.
to uncover the relationship of work engagement with job perfor-
mance and extra-role customer service. 3. Method
Second, as highlighted by Christian et al. (2011), it is important
to assess whether work engagement simultaneously leads to job 3.1. Sample and procedure
performance and extra-role customer service. Such an assessment
would enable us to understand whether engaged employees tend In this empirical study data were gathered from a sample of full-
to prioritize their in-role or extra-role tasks or pay equal attention time frontline hotel employees and their managers in the Poiana
to their in-role or extra-role tasks. Brasov region which is one of the most important winter tourist
Consistent with the precepts of SET, engaged employees appear destinations in Romania. These frontline employees (e.g., front desk
to have more trusting and high quality relationships with their agents, wait staff, bell attendants, guest relations representatives,
employer and thus demonstrate positive behaviors (Saks, 2006). bartenders, door attendants) had intense face-to-face or voice-to-
Although limited, empirical evidence suggests that work engage- voice interactions with customers.
ment leads to better in-role and/or extra-role performances Most organizations in Romania are devoid of qualified man-
(Christian et al., 2011; Karatepe, 2011). Therefore, the following power and human resource expertise (cf. Ineson and Berechet,
hypotheses are proposed: 2011). According to Ciulu and Dragan (2011), low pay and inad-
equate training programs are among the critical problems in the
H2. Work engagement is positively related to frontline employ- hospitality industry in Romania, because they seem to be respon-
ees’ (a) job performance and (b) extra-role customer service. sible for poor services and high employee turnover. Insufficient
 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140 135

Training Job performance

H2(a)

H1
Empowerment High-performance Work engagement
work practices H3(a)-H3(b)

H2(b)

Extra-role
customer service
Rewards

Fig. 1. Research model.

training programs, coupled with the lack of service standards, hin- collected data from managers to evaluate frontline employees’ job
der efficient and effective customer service (cf. Carmen, 2011). performance and extra-role customer service for reducing the risk
Under these circumstances, it is obvious that the hospitality of common method bias to the magnitudes of the relationships
industry in Romania needs modern human resource management among the study variables.
methods that are associated with HPWPs for retaining a pool of 123 questionnaires were distributed to frontline employees. The
engaged employees with high quality performance in the organi- number of employee questionnaires distributed and collected can
zation. be seen in Table A.1 in Appendix A. By the cut-off date for data
There were 7 four-star hotels and only 1 five-star hotel in the collection, 114 questionnaires were returned. However, 4 ques-
Poiana Brasov region at the time of this study. Management of these tionnaires were eliminated, because there was missing information
hotels was contacted using a letter indicating the purpose of the in each of these questionnaires. Usable 110 questionnaires were
study and permission for data collection. Although management of retrieved for a response rate of 89.4%. Strong management support
all hotels agreed to participate in this study, the researcher was not and cooperation made the solid response rate possible. It is also
allowed to directly contact frontline employees. Therefore, hotel worthy to note that the response rate in this study is not neces-
managers distributed the questionnaires to their frontline employ- sarily unusual and is comparable to the response rates obtained in
ees. some prior studies (e.g., Boles et al., 2003; Karatepe and Tekinkus,
The first page of each questionnaire consisted of informa- 2006). In addition, the researcher obtained 110 questionnaires from
tion about the assurance of anonymity and confidentiality. The managers that were matched with the employee questionnaires.
employee questionnaire contained the training, empowerment, Nine percent of the respondents were between the ages of
rewards, and work engagement measures as well as items about 38–47, while the overwhelming majority of the respondents (76%)
respondents’ profile. The researcher prepared a master list con- were younger than 38. The rest were older than 47.58% of the
sisting of the name of each frontline employee in the hotel. Each respondents were male. 6% of the respondents had primary school
employee in this master list had an identification number. An iden- education, while one-half of the respondents had secondary and
tification number was also written on each questionnaire. The high school education. The rest had college education or beyond.
manager questionnaire included the job performance and extra- 66% of the respondents had tenures of 5 years or less. The rest had
role customer service measures and had an identification number. been with their hotel for more than 5 years.
Managers assessed each frontline employee’s job performance and
extra-role customer service under their supervision. The manager 3.2. Measures
questionnaires were matched with the employee questionnaires
using the identification number. Each frontline employee and man- All perceptual constructs were measured using multiple items
ager self-administered the questionnaire, sealed it in an envelope, from different studies in the extant literature. Specifically, train-
and placed it in a special box in order to keep anonymity and con- ing was measured with six items and rewards with five items
fidentiality. Then, the researcher collected the questionnaires from from Boshoff and Allen (2000). Five items adapted from Hayes
this box. (1994) were used to measure empowerment. The shortened ver-
Using data from a single source is prone to common method sion of the Utrecht Work Engagement Scale (nine items) was used
bias. In their recent review, Line and Runyan (2012) report that to operationalize work engagement (Schaufeli et al., 2006). Vigor,
less than 10% of the studies in the hospitality marketing literature dedication, and absorption each consisted of three items. Five
have obtained data from multiple sources. Therefore, consistent items adapted from Babin and Boles (1998) were used to opera-
with the suggestions made by Podsakoff et al. (2003), this study tionalize job performance. Five items from Bettencourt and Brown
136 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140

(1997) were also employed to measure extra-role customer ser- The fully mediated model was compared with the partially
vice. Responses to items in training, empowerment, rewards, job model based on the 2 difference test (p < .01) (Chen et al., 2005).
performance, and extra-role customer service were rated on five- The 2 difference test for the fully (2 = 290.40, df = 182) and
point scales ranging from 5 (strongly agree) to 1 (strongly disagree). partially (2 = 288.06, df = 180) mediated models showed a non-
Response options for items in work engagement (vigor, dedica- significant difference in fit (2 = 2.34, df = 2). The fully mediated
tion, absorption) ranged from 6 (always) to 0 (never). Higher scores model appeared to provide a better fit to the data than did the
indicated higher levels of each construct (e.g., training, job perfor- partially mediated model. The fully mediated model fit the data
mance). adequately based on a number of fit statistics: (2 = 290.40, df = 182;
All items were originally prepared in English and then translated 2 /df = 1.60; CFI = .93; IFI = .94; RMSEA = .074; SRMR = .076). The
into Romanian using the back-translation method (Parameswaran results of SEM for the fully mediated model are presented in
and Yaprak, 1987). Specifically, the employee and manager ques- Fig. 2.
tionnaires were prepared in English. Then, two bilingual individuals The results of SEM indicated that all estimates were signifi-
(fluent in both Romanian and English) participated independently cant. The indicators of HPWPs were reliable. Specifically, rewards
in the translation process. Finally, the researcher further checked ( 31 = .84, t = 7.71) appeared to be the most reliable indicator, fol-
the two versions of the employee and manager questionnaires in lowed by training ( 11 = .79, t = 7.27) and empowerment ( 21 = .58,
English for any inconsistencies. The employee questionnaire was t = 5.65). The results demonstrated that HPWPs significantly and
tested with a pilot sample of ten frontline hotel employees. The positively influenced work engagement ( 41 = .82, t = 8.30). There-
manager questionnaire was tested using a pilot sample of five fore, Hypothesis 1 was supported. According to the results of
managers. No changes were made in the employee and manager SEM, work engagement exerted a significant positive effect on job
questionnaires, because frontline employees and their managers performance (ˇ54 = .44, t = 4.23) and extra-role customer service
did not have any difficulty understanding items. (ˇ64 = .57, t = 5.08). Hence, Hypotheses 2(a) and (b) were supported.
The results regarding the indirect effects were significant based
on Sobel test. Specifically, the results in Fig. 2 showed that
3.3. Data analysis
the indirect effect of HPWPs (standardized indirect effect = .36,
t = 3.93) on job performance through work engagement was
This study employed confirmatory factor analysis (CFA) and SEM
significant and positive. Hence, Hypothesis 3(a) received empir-
through LISREL 8.30 (Joreskog and Sorbom, 1996). Specifically, in
ical support. The results also indicated that the indirect effect
the current study a two-step approach containing CFA and SEM
of HPWPs (standardized indirect effect = .47, t = 4.58) on extra-
was employed (Anderson and Gerbing, 1988). In the first step,
role customer service via work engagement was significant and
the measurement model was assessed in terms of convergent and
positive. Therefore, Hypothesis 3(b) was supported. The results
discriminant validity (Anderson and Gerbing, 1988; Fornell and
accounted for 55% of the variance in HPWPs, 66% in work engage-
Larcker, 1981).
ment, 20% in job performance, and 32% in extra-role customer
The second step was associated with a comparison of two mod-
service.
els. It should be noted that HPWPs (as manifested by training,
empowerment, and rewards) were represented as a second-order
latent variable. In addition, average item scores of vigor, dedication,
5. Discussion
and absorption were calculated as the indicators of work engage-
ment (Schaufeli et al., 2006). The fully and partially mediated
5.1. Strengths of the study
models were compared based on the 2 difference test. Specifi-
cally, the partially mediated model consisted of the direct effects
This study proposed and tested a research model that investi-
of HPWPs on job performance and extra-role customer service as
gated work engagement as a mediator variable in the relationship
well as their indirect effects on job performance and extra-role cus-
between HPWPs and performance outcomes. The strengths of the
tomer service through work engagement. The fully mediated model
study are given below.
included the indirect effects of HPWPs on job performance and
First, this study examined the relationship between HPWPs and
extra-role customer service via work engagement. Having no sig-
work engagement. HPWPs have been conceptualized and mea-
nificant difference in fit suggests that the partially mediated model
sured as a second-order latent variable that is manifested by
does not improve fit. Then, the results in the fully mediated model
training, empowerment, and rewards. This is significant, because
are used for assessing the relationships.
training, empowerment, and rewards are among the effective
The hypothesized relationships were tested using SEM. The
human resource practices (Pfeffer, 1994), are in the top six human
overall 2 measure, CFI [comparative fit index], IFI [incremental
resource practices (Boselie et al., 2005), and have been regarded
fit index], RMSEA [root mean square error of approximation], and
among the most important indicators of management commit-
SRMR [standardized root mean square residual] were used to assess
ment to service quality (e.g., Kim et al., 2009). These HPWPs have
model fit. Although the sample size of this study is small, it appears
also been considered to be the critical implications for hospitality
to be consistent with the majority of suggestions regarding the min-
managers for business practice (Kusluvan et al., 2010).
imum sample size of 100 for SEM (Hair et al., 2010). The internal
Second, this study responds to Christian et al.’s (2011) research
consistency reliability was assessed using the cut-off value of .70.
call by testing work engagement that may simultaneously result in
The mediation effects were assessed using the guidelines provided
job performance and extra-role customer service. Third, the present
by Baron and Kenny (1986).
study adds to the existing knowledge base by testing work engage-
ment as a full mediator between HPWPs and performance outcomes
4. Results using data gathered in frontline service jobs in the hospitality
industry (e.g., Karatepe, 2011). Finally, the results from SEM suggest
Measurement results are reported in Table B.1 in Appendix B. that all hypotheses are supported and the research model is viable.
Table 1 presents means, standard deviations, and correlations of While causality cannot be determined due to the cross-sectional
study constructs. In light of the results reported in Table 1, all direct nature of the study, the previously mentioned HPWPs have been
associations were significant. Such results met the conditions for demonstrated to influence work engagement that in turn leads to
employing a mediation analysis. high levels of job performance and extra-role customer service.
 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140 137

Table 1
Means, standard deviations, and correlations of study variables.

Variables Mean SD 1 2 3 4 5 6 7 8

1. Training 3.07 .85 1.000

2. Empowerment 2.93 1.04 .518 1.000
3. Rewards 3.24 .94 .578 .434 1.000
4. Vigor 3.18 1.16 .524 .435 .644 1.000
5. Dedication 3.39 1.12 .553 .406 .576 .790 1.000
6. Absorption 3.36 1.14 .502 .272 .432 .666 .701 1.000
5. Job performance 3.38 .88 .284 .407 .370 .361 .324 .338 1.000
6. Extra-role customer service 3.28 .89 .421 .437 .411 .465 .417 .418 .682 1.000

Note: Composite scores for each construct were calculated by averaging respective item scores. Correlations are significant at the .01 level.

5.2. Assessment of findings Under these circumstances, employees feel obliged to respond to
the organization through work engagement.
The results suggest that the indicators of HPWPs are reli- The results also suggest that work engagement simultaneously
able. Specifically, rewards seem to be the most reliable indicator leads to job performance and extra-role customer service. As
of HPWPs, followed by training and empowerment. This study SET contends, employees who are engaged in their work are
provides empirical evidence that suggests a positive relationship likely to have more trusting and high quality relationships with
between HPWPs and work engagement. This is consistent with their employer (Karatepe, 2011; Saks, 2006). Such employees
internal fit (Wall and Wood, 2005) and SET (Cropanzano and in turn carry out their tasks effectively and go the extra-mile in
Mitchell, 2005). It seems that the presence of rewards mecha- dealing with customer problems and facilitating customer needs.
nism in the workplace sends powerful signals to employees that According to the results of SEM, work engagement appears to
their efforts are appreciated, recognized and rewarded in the orga- have a stronger relationship with extra-role customer service than
nization. Training programs that focus on improving employees’ with job performance. This finding suggests that employees tend
task-related and behavioral skills and teaching employees the use to prioritize extra-role tasks for responding to customer requests
of empowerment in challenging service encounters are also impor- and problems promptly. The results pertaining to the relationship
tant signals concerning management commitment’s to HPWPs. The between work engagement and performance outcomes are not
joint presence of such HPWPs creates synergy in the workplace. only consonant with limited empirical evidence in the relevant

.47,

CFI = comparative fit index; IFI = incremental fit index; RMSEA = root mean square error of
approximation; SRMR = standardized root mean square residual.

Fig. 2. Structural model results.

138 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140

literature (Christian et al., 2011; Karatepe, 2011), but also con- light on their relationships with work engagement and perfor-
sistent with the precepts of SET (Cropanzano and Mitchell, 2005; mance outcomes.
Takeuchi et al., 2007). Third, job performance and extra-role customer service were
According to strategic human resource management perspec- used as the performance outcomes in this study. There are also
tive, HPWPs should motivate employees to demonstrate positive other important performance outcomes in frontline service jobs.
behavioral outcomes (Huselid, 1995). The results of this study One of them is creative performance. Creative performance refers
provide support for this claim. As a motivational construct, work to the amount of new ideas generated and novel behaviors dis-
engagement plays a full mediator role between HPWPs and played by employees in carrying out job-related tasks (Wang and
performance outcomes. Specifically, the availability of HPWPs, Netemeyer, 2004). In future studies incorporating creative perfor-
as manifested by training, empowerment, and rewards, makes mance into the research model would provide a better picture of
employees stay engaged in their work. Such employees in turn dis- the mediating role of work engagement in the relationship between
play high quality job performance and willingly go out of their way HPWPs and relevant performance outcomes. Fourth, in future stud-
for satisfying customers. ies using cross-national samples would be beneficial for evaluating
the applicability of the research model to other countries (e.g.,
5.3. Management implications Nigeria, Turkey, China).
Fifth, hotel managers did not permit the researcher to directly
The results of this study delineate several useful implications contact frontline employees for data collection. Therefore, hotel
regarding the management of HPWPs and retention of engaged managers coordinated the data collection. The researcher used a
employees in the workplace. Work engagement is a long-term and special box for questionnaires to ensure anonymity and confiden-
continuous process. Therefore, managers need to understand the tiality. Despite such a precaution, in future studies gathering data
critical role of social exchange for work engagement and its perfor- directly from frontline employees would be more useful. Finally,
mance outcomes (Saks, 2006). replication studies with large sample sizes in different hospital-
Managers should provide employees with training, empower- ity and tourism settings in Romania and other Eastern European
ment, and rewards that would make them feel obliged to respond to countries would allow for broadening the database in this research
the organization via elevated levels of work engagement and per- stream.
formance outcomes. Specifically, management of the hotels should
make sure that appropriate reward policies are in place. Manage- 6. Conclusion
ment of the hotels should also ensure that they have continuous
training programs for improving employees’ technical and behav- The current study proposed and tested a research model that
ioral skills and provide them with the adequate responsibility and examined work engagement as a mediator between HPWPs and
authority for managing customer requests and delivering effec- performance outcomes. The results showed that the availability of
tive service recoveries. The presence of continuous and effective HPWPS, as manifested by training, empowerment, and rewards, led
training and empowerment practices, coupled with employees’ to work engagement. Work engagement in turn enhanced employ-
perceptions of fairness in rewards, would enable managers to retain ees’ job performance and extra-role customer service. In technical
a pool of engaged employees who can deal with customer requests terms, work engagement fully mediated the effects of HPWPs on
and complaints effectively and go beyond their formal role require- job performance and extra-role customer service. Under these cir-
ments for satisfying customers. As stated by Bakker et al. (2008), it is cumstances, hotel managers should invest in these HPWPs to retain
important to retain engaged employees in the workplace, because a pool of engaged employees that can demonstrate high quality
they often have positive emotions and better psychological and performance in the workplace.
physical health than disengaged employees. In today’s global competitive market environment, a richer and
On the other hand, it should be noted that the acceptance of deeper understanding of various HPWPs that may be linked to
empowerment by frontline employees is not a sure thing, because work engagement and employee performance outcomes will con-
it may result in stress (Yavas et al., 2010). With this realization, a sig- tinue to be important. In closing, it is hoped that the results of
nificant implication for managers is the need for training employees this study can inspire other researchers to focus on the mediating
successfully to accept responsibility and authority in challenging role of work engagement in the relationship between HPWPs and
service encounters. On a closing note, the aforementioned impli- various performance outcomes using data obtained from multiple
cations would also be useful for hospitality managers in Romania sources.
due to the fact that there is a lack of qualified manpower in most
organizations that are devoid of contemporary human resource
Appendix A.
management practices.

5.4. Limitations and avenues for future research

Table A.1
There are several limitations to the present study. First, The number of employee questionnaires distributed and collected.
this study used cross-sectional data for evaluating the research Distributed Collected
hypotheses. This practice does not permit firm conclusions in terms
Hotel (I) (5-star) 25 25
of causality. Although the hypothesized relationships have been
Hotel (II) (4-star) 20 20
developed based on SET and empirical evidence, in future stud- Hotel (III) (4-star) 20 20
ies using longitudinal data for testing the relationships reported Hotel (IV) (4-star) 20 20
in this study would be beneficial. Second, training, empowerment, Hotel (V) (4-star) 10 10
Hotel (VI) (4-star) 10 6
and rewards were selected as the indicators of HPWPs. How-
Hotel (VII) (4-star) 6 3
ever, there are also other indicators of HPWPs such as internal Hotel (VIII) (4-star) 12 10
career/promotion opportunities, teamwork, employment security,
Total 123 114
staffing selectivity, and work-family balance (e.g., Boselie et al.,
2005; Kusluvan et al., 2010; Tang and Tang, 2012). The inclusion Note: 4 Questionnaires from the Hotel (VIII) (4-star) were eliminated because of
missing information. Therefore, the total number of usable questionnaires was 110.
of other relevant HPWPs in the research model would shed further
 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140 139

Table B.1 Appendix B.

Confirmatory factor analysis results.

Scale items Standardized t-Values B.1. Measurement results

loading

Training Several items were dropped during CFA due to correlation

I receive continued training to provide good .79 9.69 measurement error. Specifically, two items each from training,
service empowerment, job performance, and extra-role customer service
I received extensive customer service training – –
and one item each from vigor and absorption were removed from
before I came into contact with customers
I receive training on how to serve customers .89 11.73 further analysis. After scale purification, the proposed measure-
better ment model fit the data well according to the following model
I receive training on how to deal with .89 11.79 fit statistics (2 = 354.65, df = 247; 2 /df = 1.44; CFI = .94; IFI = .94;
complaining customers
RMSEA = .063; SRMR = .052).
I receive training on dealing with customer .88 11.52
problems As depicted in Table B.1, the magnitudes of the standardized
I was trained to deal with customer complaints – – loadings ranged from .67 to .96, and all t-values were significant.
The average variance extracted by each underlying latent variable
Empowerment
I am empowered to solve customer problems – – was above .50 and was as follows: training .75; empowerment .85;
I am encouraged to handle customer problems – – rewards .77; vigor .77; dedication .70; absorption .52; job per-
by myself formance .66; and extra-role customer service .66. None of the
I do not have to get management’s approval .87 11.42
shared variances was larger than the average variance extracted
before I handle customer problems
I am allowed to do almost anything to solve .92 12.49
by each construct. The internal consistency reliabilities were as
customer problems follows: training .92; empowerment .94; rewards .94; vigor .87;
I have control over how I solve customer .96 13.53 dedication .87; absorption .68; job performance .84; and extra-role
problems customer service .85. The internal consistency reliability for absorp-
Rewards
tion was slightly below .70. However, the internal consistency
If I improve the level of service I offer .79 9.76
customers, I will be rewarded reliability for work engagement (vigor, dedication, absorption) was
The rewards I receive are based on customer .86 11.28 .92. Collectively, these results demonstrated that the measures
evaluations of service were reliable and there was evidence of convergent and discrim-
I am rewarded for serving customers well .87 11.42 inant validity (Anderson and Gerbing, 1988; Fornell and Larcker,
I am rewarded for dealing effectively with .95 13.32
customer problems
1981).
I am rewarded for satisfying complaining .90 12.03 Though data were obtained from frontline employee–manager
customers dyads for reducing the potential threat of common method bias,
Vigor a formal test of Harman’s single-factor test was also employed
At my work, I feel bursting with energy .88 11.32
for employee and manager data separately. The six-factor model
At my job I feel strong and vigorous .88 11.35
When I get up in the morning, I feel like going – – (training, empowerment, rewards, vigor, dedication, absorption)
to work was compared with the single-factor model using the 2 differ-
Dedication ence test (p < .01). The result for a single-factor model was a 2
I am enthusiastic about my job .92 12.29 value of 1169.96 (df = 152). The results indicated that the six-
My job inspires me .87 11.26
I am proud of the work that I do .71 8.39
factor model (2 = 205.05, df = 137) was superior to the single-factor
model based on the 2 difference test (215 = 964.91). The two-
Absorption
factor model (job performance, extra-role customer service) was
I feel happy when I am working intensely .76 8.32
I am immersed in my work – – compared with the single-factor model. The results demonstrated
I get carried away when I am working .67 7.29 that the two-factor model fit (2 = 16.71, df = 8) was better than the
single-factor model (2 = 57.40, df = 9) based on the 2 difference
Job performance
This employee is a top performer – – test (2 1 = 40.69).
This employee is in the top 10% of frontline – –
employees here
This employee gets along better with .83 10.17
customers than do others References
This employee knows more about services .87 10.86
delivered to customers than others Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a
This employee knows what his/her customers .71 8.17 review and recommended two-step approach. Psychological Bulletin 103 (3),
expect better than others 411–423.
Extra-role customer service Ashill, N.J., Rod, M., Carruthers, J., 2008. The effect of management commitment to
This employee voluntarily assists customers – – service quality on frontline employees’ job attitudes, turnover intentions and
even if it means going beyond job service recovery performance in a new public management context. Journal of
requirements Strategic Marketing 16 (5), 437–462.
This employee helps customers with problems .70 7.96 Babakus, E., Yavas, U., Karatepe, O.M., Avci, T., 2003. The effect of management com-
mitment to service quality on employees’ affective and performance outcomes.
beyond what is expected or required
Journal of the Academy of Marketing Science 31 (3), 272–286.
This employee often goes above and beyond – –
Babin, B., Boles, J.S., 1998. Employee behavior in a service environment: a model and
the call of duty when serving customers test of potential differences between men and women. Journal of Marketing 62
This employee willingly goes out of his/her .87 10.98 (2), 77–91.
way to make a customer satisfied Bakker, A.B., 2011. An evidence-based model of work engagement. Current Direc-
This employee frequently goes out the way to .86 10.82 tions in Psychological Science 20 (4), 265–269.
help a customer Bakker, A.B., Demerouti, E., 2008. Towards a model of work engagement. Career
Model fit statistics: 2 = 354.65; df = 247; 2 /df = 1.44; CFI = .94; IFI = .94; Development International 13 (3), 209–223.
RMSEA = .063; SRMR = .052 Bakker, A.B., Schaufeli, W.B., Leiter, M.P., Taris, T.W., 2008. Work engagement: an
emerging concept in occupational health psychology. Work and Stress 22 (3),
Note: All loadings are significant at the .01 level. CFI = comparative fit index; 187–200.
IFI = incremental fit index; RMSEA = root mean square error of approximation; Baron, R.M., Kenny, D.A., 1986. The moderator-mediator variable distinction in
SRMR = standardized root mean square residual. social psychological research: conceptual, strategic, and statistical considera-
(–) Dropped during confirmatory factor analysis. tions. Journal of Personality and Social Psychology 51 (6), 1173–1182.
140 O.M. Karatepe / International Journal of Hospitality Management 32 (2013) 132–140

Bettencourt, L.A., Brown, S.W., 1997. Contact employees: relationships among Huselid, M.A., Jackson, S.E., Schuler, R.S., 1997. Technical and strategic human
workplace fairness, job satisfaction and prosocial service behaviors. Journal of resource management effectiveness as determinants of firm performance.
Retailing 73 (1), 39–61. Academy of Management Journal 40 (1), 171–188.
Bettencourt, L.A., Brown, S.W., MacKenzie, S.B., 2005. Customer-oriented boundary- Ineson, E.M., Berechet, G., 2011. Employee loyalty in hotels: Romanian expe-
spanning behaviors: test of a social exchange model of antecedents. Journal of riences. Journal of Human Resources in Hospitality and Tourism 10 (2),
Retailing 81 (2), 141–157. 129–149.
Boles, J.D., Wood, J.A., Johnson, J., 2003. Interrelationships of role conflict, role ambi- Joreskog, K., Sorbom, D., 1996. LISREL 8: User’s Reference Guide. Scientific Software
guity, and work-family conflict with different facets of job satisfaction and the International, Inc., Chicago.
moderating effects of gender. Journal of Personal Selling and Sales Management Karatepe, O.M., 2011. Procedural justice, work engagement, and job outcomes: evi-
28 (2), 99–113. dence from Nigeria. Journal of Hospitality Marketing and Management 20 (8),
Boselie, P., Dietz, G., Boon, C., 2005. Commonalities and contradictions in HRM and 855–878.
performance research. Human Resource Management Journal 15 (3), 67–94. Karatepe, O.M., Olugbade, O.A., 2009. The effects of job and personal resources on
Boshoff, C., Allen, J., 2000. The influence of selected antecedents on frontline staff’s hotel employees’ work engagement. International Journal of Hospitality Man-
perceptions of service recovery performance. International Journal of Service agement 28 (4), 504–512.
Industry Management 11 (1), 63–90. Karatepe, O.M., Tekinkus, M., 2006. The effects of work-family conflict, emotional
Carmen, S., 2011. Tourism and its influence upon macro-environment in Romania. exhaustion, and intrinsic motivation on job outcomes of front-line employees.
Annales Universitatis Apulensis Series Oeconomica 13 (1), 142–154. International Journal of Bank Marketing 24 (3), 173–193.
Chen, Z.X., Aryee, S., Lee, C., 2005. Test of a mediation model of perceived organiza- Karatepe, O.M., Uludag, O., 2007. Conflict, exhaustion, and motivation: a study of
tional support. Journal of Vocational Behavior 66 (3), 457–470. frontline employees in Northern Cyprus hotels. International Journal of Hospi-
Chi, C.G., Gursoy, D., 2009. Employee satisfaction, customer satisfaction, and finan- tality Management 26 (3), 645–665.
cial performance: an empirical examination. International Journal of Hospitality Kim, H.J., Tavitiyaman, P., Kim, W.G., 2009. The effect of management commitment
Management 28 (2), 245–253. to service on employee service behaviors: the mediating role of job satisfaction.
Chiang, F.F.T., Birtch, T.A., 2008. Achieving task and extra-task-related behaviors: Journal of Hospitality and Tourism Research 33 (3), 369–390.
a case of gender and position differences in the perceived role of rewards in Kusluvan, S., Kusluvan, Z., Ilhan, Z., Buyruk, L., 2010. The human dimension: a review
the hotel industry. International Journal of Hospitality Management 27 (4), of human resources management issues in the tourism and hospitality industry.
491–503. Cornell Hospitality Quarterly 51 (2), 171–214.
Chiang, F.F.T., Birtch, T.A., 2011. Reward climate and its impact on service quality Line, N.D., Runyan, R.C., 2012. Hospitality marketing research: recent trends and
orientation and employee attitudes. International Journal of Hospitality Man- future directions. International Journal of Hospitality Management 31 (2),
agement 30 (1), 3–9. 477–488.
Cho, S., Woods, R.H., (Shawn) Jang, S., Erdem, M., 2006. Measuring the impact Moliner, C., Martínez-Tur, V., Romos, J., Cropanzano, R., 2008. Organizational
of human resource management practices on hospitality firms’ performances. justice and extrarole customer service: the mediating role of well-being
International Journal of Hospitality Management 25 (2), 262–277. at work. European Journal of Work and Organizational Psychology 17 (3),
Christian, M.S., Garza, A.S., Slaughter, J.E., 2011. Work engagement: a quantitative 327–348.
review and test of its relations with task and contextual performance. Personnel Murphy, K.S., Murrmann, S., 2009. The research design used to develop a high perfor-
Psychology 64 (1), 89–136. mance management system construct for US restaurant managers. International
Ciulu, R., Dragan, L., 2011. Hospitality industry’s competition in terms of attracting Journal of Hospitality Management 28 (4), 547–555.
and retaining valuable HR in Eastern Europe – the case of Romania. Journal of Parameswaran, R., Yaprak, A., 1987. A cross-national comparison of con-
Tourism 11, 55–63. sumer research measures. Journal of International Business Studies 18 (1),
Combs, J., Liu, Y., Hall, A., Ketchen, D., 2006. How much do high-performance work 35–49.
practices matter? A meta-analysis of their effects on organizational perfor- Pfeffer, J., 1994. Competitive Advantage through People: Unleashing the Power of
mance. Personnel Psychology 59 (3), 501–528. the Work Force. HBS Press, Boston.
Cropanzano, R., Mitchell, M.S., 2005. Social exchange theory: an interdisciplinary Podsakoff, P.M., MacKenzie, S.B., Lee, J-Y., Podsakoff, N.P., 2003. Common method
review. Journal of Management 31 (6), 874–900. biases in behavioral research: a critical review of the literature and recom-
Cropanzano, R., Rupp, D.E., Byrne, Z.S., 2003. The relationship of emotional mended remedies. Journal of Applied Psychology 88 (5), 879–903.
exhaustion to work attitudes, job performance, and organizational citizenship Saks, A.M., 2006. Antecedents and consequences of employee engagement. Journal
behaviors. Journal of Applied Psychology 88 (1), 160–169. of Managerial Psychology 21 (7), 600–619.
De Jong, A., De Ruyter, K., 2004. Adaptive versus proactive behavior in service recov- Schaufeli, W.B., Bakker, A.B., Salanova, M., 2006. The measurement of work engage-
ery: the role of self-managing teams. Decision Sciences 35 (3), 457–491. ment with a short questionnaire: a cross-national study. Educational and
DiPietro, R.B., Pizam, A., 2008. Employee alienation in the quick service restaurant Psychological Measurement 66 (4), 701–716.
industry. Journal of Hospitality and Tourism Research 32 (1), 22–39. Schaufeli, W.B., Salanova, M., Gonzáles-Romá, V., Bakker, A.B., 2002. The measure-
Forrester, R., 2000. Empowerment: rejuvenating a potent idea. Academy of Manage- ment of engagement and burnout: a two sample confirmatory factor analytic
ment Executive 14 (3), 67–80. approach. Journal of Happiness Studies 3 (1), 71–92.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unob- Slåtten, T., Mehmetoglu, M., 2011. Antecedents and effects of engaged frontline
servable variables and measurement error. Journal of Marketing Research 18 employees: a study from the hospitality industry. Managing Service Quality 21
(1), 39–50. (1), 88–107.
Guchait, P., Kim, M.G., Namasivayam, K., 2012. Error management at different orga- Takeuchi, R., Lepak, D.P., Wang, H., Takeuchi, K., 2007. An empirical examination of
nizational levels – frontline, manager, and company. International Journal of the mechanisms mediating between high-performance work systems and the
Hospitality Management 31 (1), 12–22. performance of Japanese organizations. Journal of Applied Psychology 92 (4),
Hair Jr., J.F., Black, W.C., Babin, B.J., Anderson, R.E., 2010. Multivariate Data Analysis: 1069–1083.
A Global Perspective, 7th ed. Pearson Education, Upper Saddle River, NJ. Tang, T.-W., Tang, Y.-Y., 2012. Promoting service-oriented organizational citizenship
Halbesleben, J.R.B., Harvey, J., Bolino, M.C., 2009. Too engaged? A conservation of behaviors in hotels: the role of high-performance human resource practices and
resources view of the relationship between work engagement and work inter- organizational social climates. International Journal of Hospitality Management
ference with family. Journal of Applied Psychology 94 (6), 1452–1465. 31 (3), 885–895.
Hayes, B.E., 1994. How to measure empowerment. Quality Progress 27 (February), Wall, T.D., Wood, S.J., 2005. The romance of human resource management and
41–46. business performance, and the case for big science. Human Relations 58 (4),
Henry, B., Butcher, W., Browne, Y., Hinds, M., Jayawardena, C., 2004. Future human 429–462.
resource challenges in the Caribbean hospitality industry. International Journal Wang, G., Netemeyer, R.G., 2004. Salesperson creative performance: conceptualiza-
of Contemporary Hospitality Management 16 (7), 419–423. tion, measurement, and nomological validity. Journal of Business Research 57
Hughes, J.C., Rog, E., 2008. Talent management: a strategy for improving employee (8), 805–812.
recruitment, retention and engagement within hospitality organizations. Inter- Yavas, U., Karatepe, O.M., Babakus, E., 2010. Relative efficacy of organizational
national Journal of Contemporary Hospitality Management 20 (7), 743–757. support and personality traits in predicting service recovery and job perfor-
Huselid, M.A., 1995. The impact of human resource management practices on mances: a study of frontline employees in Turkey. Tourism Review 65 (3),
turnover, productivity, and corporate financial performance. Academy of Man- 70–83.
agement Journal 38 (3), 635–672.