Job stress, Job Performance, and Leadership support

ABSTRACT Based on data obtained from frontline hotel employees in Northern Cyprus, this study investigated the effects of role stress and burnout on job performance. It was found that role ambiguity decreased job performance while role conﬂict enhanced job performance. The results indicated that diminished personal accomplishment exerted a signiﬁcant negative inﬂuence on job performance whereas the rest of the burnout dimensions did not. The results further showed that both role conﬂict and ambiguity exacerbated emotional exhaustion and depersonalisation. Role ambiguity was found to have a signiﬁcant positive relationship with diminished personal accomplishment whereas role conﬂict was not. Copyright © 2007 John Wiley & Sons, Ltd.

Received 1 May 2006; Revised 20 June 2007; Accepted 17 July 2007 Keywords:role stress; burnout; frontline employees; job performance; Northern Cyprus.INTRODUCTION I n today’s global competitive environment, hotel organisations realise that they have to focus on delivering exceptional service quality and creating customer satisfaction in order to acquire and retain a pool of proﬁtable loyal customers for their survival and success.

Indeed, frontline hotel employees are the most important links in this process (Karatepe et al., 2006). Hotel organisations, however, are faced with the challenging task of ﬁnding effective ways to ensure that the attitudes and behav- iours of frontline employees are congruent with the expectations of the management and customers (Hartline and Ferrell, 1996). Under these circumstances, hotel managers need to pay attention to the well-being of their front- line employees.

Yet despite well-intentions of management, frontline hotel employees are confronted with a number of problems that stem from inade- quate training and development oppor- tunities, low pay, excessive workloads, and unsocial work hours (Babin and Boles, 1998; Karatepe and Sokmen, 2006; Kim et al., 2007).

They are often susceptible to role stress, which consists of role conﬂict and role ambiguity (Hartline and Ferrell, 1996; Karatepe and Sokmen, 2006). Role conﬂict occurs when indi- viduals have incompatible job demands from various parties such as customers, co-workers, and managers and ﬁnd that they are incapable of satisfying all job demands at the same time; and role ambiguity occurs when individuals do not have clear information about the Copyright © 2007 John Wiley & Sons, Ltd. INTERNATIONAL JOURNAL OF TOURISM RESEARCH Int. J. Tourism Res.10, 111–126 (2008) Published online 2 November 2007 in Wiley InterScience (www.interscience.wiley.com) DOI:10.1002/jtr.645 Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance: Evidence from Northern Cyprus Osman M. Karatepe* and Orhan Uludag School of Tourism and Hospitality Management, Eastern Mediterranean University, Gazimagusa, Turkish Republic of Northern Cyprus via Mersin 10, Turkey *Correspondence to: O. M. Karatepe, School of Tourism and Hospitality Management, Eastern Mediterranean University, Gazimagusa, Turkish Republic of Northern Cyprus via Mersin 10, Turkey.

E-mail: [email protected] expectations of their roles in the job or the organisation (Churchill et al., 1976).

In addition, frontline hotel employees expe- rience burnout (Ledgerwood et al., 1998), which is characterised by emotional exhaus- tion, depersonalisation and diminished per- sonal accomplishment (Maslach and Jackson, 1981). Maslach and Jackson (1981, p. 99) deﬁne burnout as ‘a syndrome of emotional exhaus- tion and cynicism that occurs frequently among individuals who do “people-work” of some kind’. Emotional exhaustion is the initia- tor of the burnout syndrome (Gaines and Jermier, 1983; Cordes and Dougherty, 1993) and is related to depletion of emotional resources. Under these circumstances, individ- uals cannot give of themselves at a psycholog- ical level (Gaines and Jermier, 1983). Another aspect of burnout is depersonalisation, which is deﬁned as ‘the development of negative, cynical attitudes and feelings about one’s clients’ (Maslach and Jackson, 1981, p. 99).

Diminished personal accomplishment is the third aspect of the burnout syndrome and is related to a tendency to evaluate oneself nega- tively (Maslach and Jackson, 1981).

Against this backdrop, the current study develops and tests a research model that investigates the effects of role conﬂict and ambiguity and emotional exhaustion, deper- sonalisation, and diminished personal accom- plishment on an organisationally valued job outcome: job performance. This study also examines the relationships of role stressors with the three dimensions of burnout. The aforementioned relationships are tested using data collected from frontline hotel employees in Northern Cyprus as the study setting.

This empirical study makes several contri- butions to the tourism and hospitality litera- ture in the following ways. First, empirical evidence regarding the relationships of the three components of burnout with employees’ job performance in the extant literature is not abundant (Bakker et al., 2004; Wright and Hobfoll, 2004). This is also valid in the tourism and hospitality literature. Second, empirical evidence pertaining to the relationships of role stressors with burnout dimensions in the tourism and hospitality literature is meager.

Finally, the ﬁndings of this study may prove useful to hotel managers for business practice.The next section includes the research model and a set of hypotheses. This is followed by discussions of the method and results of the study. The study concludes with key implica- tions for research and practice.

RESEARCH MODEL AND HYPOTHESES The conservation of resources (COR) theory provides a basis to understand the relation- ships of role stressors with the three compo- nents of burnout and the impacts of these dimensions on various job outcomes (Hobfoll, 1989; Lee and Ashforth, 1996). The basic ingre- dient of the COR theory is that individuals seek to obtain, retain, protect and foster resources (Hobfoll, 1989, 2001). Resources are deﬁned as ‘those objects, personal characteris- tics, conditions, or energies that are valued by the individual or that serve as a means for attainment of these objects, personal charac- teristics, conditions, or energies’ (Hobfoll, 1989, p. 516). According to the theory, burnout is most likely to occur when the resources of individuals are threatened with loss, lost, or when individuals fail to gain resources after a signiﬁcant resource investment (Hobfoll, 2001). The theory further posits that burnout becomes a reaction to role stress when indi- viduals are incapable of coping with excessive demands on their resources (Lee and Ashforth, 1996).

The COR theory also provides a basis to understand the relationships of emotional exhaustion, depersonalisation and diminished personal accomplishment with job perfor- mance (Wright and Bonett, 1997; Wright and Hobfoll, 2004). Job performance is deﬁned as ‘the level of productivity of an individual employee, relative to his or her peers, on several job-related behaviors and outcomes’ (Babin and Boles, 1998, p. 82). The COR theory suggests that employees who are emotionally exhausted lack certain resources needed to enhance their job performance because these resources might already have depleted (Wright and Cropanzano, 1998). According to this theory, employees may try to minimise the loss of emotional resources by selecting strategies that help utilise and conserve their scarce resources (Wright and Hobfoll, 2004).

However, employees who detach themselves 112O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr from their customers and co-workers perform ineffectively in the organisation. As persua- sively discussed by Wright and Bonett (1997), the COR theory predicts that employees do not perform well once they hold negative self- evaluations about their job performance.

The research model guiding the present study is shown in Figure 1. The model con- tends that both role conﬂict and ambiguity amplify emotional exhaustion, depersonalisa- tion and diminished personal accomplish- ment. According to the model, the three components of burnout inﬂuence job perfor- mance deleteriously. The model further posits that role ambiguity decreases job performancewhile role conﬂict increases job performance.

The research hypotheses are discussed next.

Outcomes of role conﬂict and role ambiguity Individuals who are confronted with role con- ﬂict and ambiguity experience emotional exhaustion, depersonalisation and diminished personal accomplishment. This is consistent with the COR theory. There is empirical evi- dence to support the previously mentioned relationships. For instance, Lee and Ashforth (1996), in their meta-analytic study, reported that role conﬂict and ambiguity were Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance113 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr Role conflict (ROLEC) Emotional exhaustion (EEXHAUST) Role ambiguity (ROLEA) Depersonalization (DEP) Job performance (JPERF) Diminished personal accomplishment (DPA) Control variables Age Gender Education Organizational tenure Marital status The number of childen Hotel type H1a (+) H1b (+) H1c (+) H2a (+) H2b (+) H2c (+) H3a (+) H3b (-) H4a (-) H4b (-) H4c (-) Figure 1. Research model. signiﬁcantly and positively correlated with emotional exhaustion and depersonalisation.

They also reported that these role stressors had signiﬁcant negative correlations with personal accomplishment. Hsieh and Hsieh (2003) found evidence in the manufacturing and service sectors of Taiwan that the previously mentioned role stressors had signiﬁcant posi- tive effects on employees’ emotional exhaus- tion and depersonalisation. They further indicated that role ambiguity was signiﬁcantly and positively associated with diminished per- sonal accomplishment whereas role conﬂict was not. Recently, Piko (2006) also reported similar ﬁndings regarding the impact of role conﬂict on the three dimensions of burnout for a sample of health care employees.

Thus, the following hypotheses are proposed:

H1: Role conﬂict is positively related to front- line employees’ (a) emotional exhaustion, (b) depersonalisation and (c) diminished personal accomplishment.

H2: Role ambiguity is positively related to frontline employees’ (a) emotional exhau- stion (b) depersonalisation and (c) dimi- nished personal accomplishment.

Although a number of empirical studies have investigated the impact of role conﬂict on job performance, the individual studies have reported mixed results regarding this relation- ship. For example, Brown and Peterson (1993), in their meta-analytic study, failed to ﬁnd a signiﬁcant relationship between role conﬂict and salesperson performance. MacKenzie et al.

(1998) showed that salespeople reported lower job (in-role) performance when they were con- fronted with role conﬂict. On the other hand, various empirical studies demonstrated a pos- itive relationship between role conﬂict and job performance (Behrman and Perreault, 1984; Ross and Boles, 1994; Babin and Boles, 1996; Babakus et al., 1999). This ﬁnding can be attrib- uted to the fact that role conﬂict is one of the unavoidable aspects of boundary-spanning positions, and frontline employees may have learnt how to cope with difﬁculties emanating from role conﬂict. Therefore, this study posits that role conﬂict increases frontline hotel employees’ job performance.

Acareful examination of the extant literature indicates that the ﬁndings regarding the effectof role ambiguity on job performance have been more consistent. Speciﬁcally, a number of empirical studies (e.g., Ross and Boles, 1994; MacKenzie et al., 1998; Babakus et al., 1999) and a meta-analytic study (Brown and Peterson, 1993) demonstrate that employees having a lack of information about their job-related duties and responsibilities display decreased performance in the workplace.

Accordingly, the following hypotheses are proposed:

H3a: Role conﬂict is positively related to front- line employees’ job performance.

H3b: Role ambiguity is negatively related to frontline employees’ job performance.

Outcomes of emotional exhaustion, depersonalisation and diminished personal accomplishment Consistent with the COR theory, when employees are prone to high levels of emo- tional exhaustion, depersonalisation and dimi- nished personal accomplishment, their job per- formance starts to erode. This is evident because employees no longer possess adequate resources to cope with difﬁculties that stem from the three components of burnout.

Although there are limited empirical studies pertaining to the aforementioned relation- ships, the ﬁndings across the individual studies are largely mixed. For example, Wright and Bonett (1997) found that emotional exhaustion was signiﬁcantly and negatively associated with subsequent work performance whereas depersonalisation and diminished personal accomplishment were not. Recently, Wright and Hobfoll (2004) reported similar results concerning the relationships of the burnout dimensions with job performance.

More recently, Advani et al. (2005) found that emotional exhaustion increased the job per- formance of Indian software professionals whereas depersonalisation and diminished personal accomplishment were not signiﬁ- cantly related to their job performance. Obvi- ously, the aforementioned ﬁndings require a further testing of the relationships of burnout components with job performance.

Accordingly, the following hypotheses are proposed: 114O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr H4a: Emotional exhaustion is negatively related to frontline employees’ job performance.

H4b: Depersonalization is negatively related to frontline employees’ job performance.

H4c: Diminished personal accomplishment is negatively related to frontline em- ployees’ job performance.

THE NORTHERN CYPRUS CONTEXT Cyprus is one of the largest islands in the Mediterranean Sea, which covers an area of about 3500 square miles (Kilic and Okumus, 2005). As a result of the partition of the island in 1974, the Turkish Cypriots have been living in the north while the Greek Cypriots have been living in the south (Okumus et al., 2005).

The Turkish Cypriots declared the indepen- dence of the Turkish Republic of Northern Cyprus in 1983, which is recognised interna- tionally only by Turkey (Altinay, 2000). North- ern Cyprus, having a population of 211 000 is a small island destination (Kilic and Okumus, 2005). Despite its sensitivity to the political uncertainty, tourism industry is regarded as the lifeblood of the ﬂedgling Northern Cyprus economy.

There are a number of problems in the Northern Cyprus tourism and hospitality industry that emanate from transportation difﬁculties, scarce qualiﬁed staff, inadequate infrastructure, lack of top management com- mitment to service quality, seasonality and low occupancy rates, poor service quality, pollu- tion, inadequate supplementary facilities and services, and high prices (Akis and Warner, 1994; Altinay et al., 2002; Yavas et al., 2004; Kilic and Okumus, 2005; Altinay and Bowen, 2006).

In addition, employees working in Northern Cyprus hotels are confronted with various problems including long work hours, inade- quate pay, lack of training, empowerment, and rewards, role stress, work–family conﬂict, excessive job demands, irregular and inﬂexible work schedules, and limited weekend time off (Scott, 1997; Karatepe et al., 2003; Yavas et al., 2004; Kilic and Okumus, 2005; Karatepe and Kilic, 2007). The aforementioned problems demonstrate that most of the Northern Cyprus hotels are devoid of contemporary human resource managerial practices.METHODOLOGY Sample Data for this study were collected from a judgemental sample of Turkish Cypriot front- line employees in Northern Cyprus hotels.

Judd et al. (1991) deﬁne judgemental sampling as ‘picking cases that are judged to be typical of the population in which we are interested, assuming that errors of judgement in the selec- tion will tend to counterbalance one another ’ (p. 136). Using the judgemental sampling pro- cedure, the sample of this empirical study included frontline employees in three-, four- and ﬁve-star hotels in Northern Cyprus. There are several reasons for selecting frontline employees. First, frontline employees have fre- quent interaction with customers and are expected to deal with a number of customers’ requests. Second, as indicated before, they are prone to heightened role stress and burnout.

Hotel types in the research location con- sisted of independently/family-owned and -operated hotels and chain hotels. 83.3% of the participating hotel companies in the research location were independently/family- owned and -operated hotels, while 16.7% were chain hotels.

The research team contacted the manage- ments of 28 three-, 8 four- and 6 ﬁve-star hotels in order to learn the total number of frontline employees and receive permission for data collection. The research team learnt that there were 943 frontline employees working in the previously mentioned hotels at the time of the study. However, the research team was not able to receive permission from the manage- ments of three three-star hotels and three four- star hotels for data collection. Allowing for hotels that did not participate, the number of frontline employees in the research location was 872, each of whom personally received a copy of a questionnaire. It should be noted that all frontline employees in the study sample (e.g. food servers, front desk agents and concierges) had frequent contact with cus- tomers and spent most of their time directly with customers to deal with their needs and requests. A detailed explanation on the ques- tionnaire was provided, and all participants were assured of the conﬁdentiality and Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance115 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr anonymity of their responses. Participants were requested to ﬁll out the questionnaires in a self-administered manner. Among the 872 questionnaires distributed, 677 were retrieved by the cut-off date for data collection, yielding a response rate of 77.6%.

Measurement The relevant writings in the extant literature were canvassed in order to operationalise the constructs depicted in Figure 1.

Role conﬂict and role ambiguity.Role conﬂict and role ambiguity were measured using eight and six items from Rizzo et al. (1970) respectively.

There are some concerns regarding the valid- ity of the role conﬂict and role ambiguity scales developed by Rizzo et al. (1970) (McGee et al., 1989). However, these scales have received considerable empirical attention in the extant literature (Jackson and Schuler, 1985; Brown and Peterson, 1993), and there is empirical support for the validity of these scales (Smith et al., 1993). Responses to the items in role con- ﬂict and role ambiguity were elicited on ﬁve- point scales ranging from ‘5 =strongly agree’ to ‘1 =strongly disagree’.

Burnout. As mentioned in the preceding parts, burnout consists of emotional exhaustion, depersonalisation and diminished personal accomplishment. Emotional exhaustion was measured using eight items, depersonalisation was measured via ﬁve items and diminished personal accomplishment was operationalised using eight items from Maslach and Jackson (1981). Responses to the items in emotional exhaustion, depersonalisation and diminished personal accomplishment were elicited on ﬁve-point scales ranging from ‘5 =strongly agree’ to ‘1 =strongly disagree’.

Job performance.Five items from Babin and Boles (1998) were adapted to operationalise job performance. Although the use of self-report measure may result in inﬂated results, Churchill et al. (1985) cogently discuss that measuring job performance using a self-report measure does not necessarily lead to system- atic bias. There are also a number of empirical studies that have used self-report measures inorder to operationalise job performance (e.g.

Babin and Boles, 1998; Babakus et al., 1999).

Each of the job performance items used a ﬁve- point scale ranging from ‘5 =strongly agree’ to ‘1 =strongly disagree’.

The questionnaire was originally prepared in English and then translated into Turkish via the back-translation method (McGorry, 2000).

The survey instrument was tested with a pilot sample of 30 frontline employees in the research location. The results of the pilot study revealed that frontline employees did not have any difﬁculty in understanding the items.

Consequently, no changes were made in the survey instrument.

Age, gender, education, organisational ten- ure, marital status, the number of children and hotel type were statistically controlled in this study. Age and education were measured using ﬁve-point scales. Organisational tenure was measured using a six-point scale. The number of children was measured using a four-point scale. Gender was coded as a binary variable (0 =male and 1 =female). Marital status was recorded as a binary variable (0 = single or divorced and 1 =married). Hotel type was also coded as a binary variable (0 =chain hotel and 1 =independently/family-owned and -operated hotel).

All measures were subjected to conﬁrmatory factor analysis to provide support for the issues of dimensionality, convergent and dis- criminant validity (Anderson and Gerbing, 1988). The research hypotheses were tested using LISREL 8.30 through path analysis (Joreskog and Sorbom, 1996).

RESULTS Sample The majority of the respondents were male (57.3%). Most of the respondents (50.7%) were between the ages of 18–27, 30.6% were aged between 28 and 37, and the rest were older than 37. Approximately one-third (31.9%) of the respondents had secondary and high school education. 56.2% of the respondents had university education. The rest had pri- mary school education. Twenty-nine percent of the respondents had tenures of less than one year and almost half of the respondents 116O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr (47.9%) had tenures of one to ﬁve years. The rest had tenures of above ﬁve years. 40.2% of the respondents were married whereas the rest were single or divorced. Almost two-thirds (65%) of the respondents had no children, 34.6% of the respondents had one to four chil- dren and the rest had more than four children.

Measurement results All items were subjected to a series of conﬁr- matory factor analyses for a rigorous psycho- metric assessment (Joreskog and Sorbom, 1996). The initial results of the conﬁrmatory factor analysis provided low model ﬁt statis- tics. Therefore, several items were deleted because of low standardised loadings (<0.50) or correlation measurement errors. Speciﬁcally, two items from role conﬂict and one item each from role ambiguity and diminished personal accomplishment were removed from further analysis. The ﬁnal results of the conﬁrmatory factor analysis demonstrated a moderate ﬁt of the six-factor model to the data on the basis of a number of ﬁt statistics (e.g., χ 2=2500.60, d.f.

=579; Goodness of Fit Index =0.83; Compara- tive Fit Index (CFI) =0.83; Root Mean Square Residual (RMR) =0.059). As Table 1 indicates, the magnitudes of the standardised loadings ranged from 0.51 to 0.86, and all t-values were signiﬁcant (>2.00). The previously mentioned results provided support for the issue of con- vergent validity (Anderson and Gerbing, 1988). As shown in Table 1, all coefﬁcient alphas were found to be greater than the cut- off level of 0.70 (Nunnally, 1978).

To assess discriminant validity, a series of pairwise conﬁrmatory factor analyses was employed. A two-dimensional model for each pair of study constructs was ﬁrst developed, and then items representing each factor were forced into a single factor solution. The results of various ﬁt statistics were better for a two- factor model. The results conﬁrm that each set of items represents a single underlying con- struct, and thus provide evidence of discrimi- nant validity (Anderson and Gerbing, 1988).

Composite scores for each measure were obtained by averaging scores across items rep- resenting that measure. Table 2 shows the cor- relations among the study variables. All control variables excluding gender and organ-isational tenure had signiﬁcant positive corre- lations with the study variables. The results indicate that older, better-educated and married employees have more emotional exhaustion and depersonalisation. This is also valid for those who work in independently/ family-owned and -operated hotels. The results show that better-educated employees experience higher role conﬂict and diminished personal accomplishment. The results further indicate that older employees have less per- sonal accomplishment while the ones with more children have higher depersonalisation.

As the results in Table 2 illustrate, married employees and the ones working in indepen- dently/family-owned and -operated hotels are confronted with higher role ambiguity.

Of the control variables, age, gender, organ- isational tenure and hotel type also had signiﬁ- cant negative correlations with the study vari- ables. Speciﬁcally, employees with longer tenure are susceptible to less role ambiguity, emotional exhaustion, depersonalisation and diminished personal accomplishment. The results reveal that female employees experi- ence less role conﬂict and older employees and the ones in independently/family-owned and -operated hotels display an ineffective performance.

Test of research hypotheses The correlation matrix depicted in Table 2 was used as input to test the hypotheses using LISREL 8.30 through path analysis (Joreskog and Sorbom, 1996). The results indicate that the model ﬁts the data well (e.g. χ 2=6.45, d.f.

=1; CFI =1.00; RMR =0.013). As shown in Table 3, the variances explained in role conﬂict, role ambiguity, emotional exhaustion, deper- sonalisation, diminished personal accomplish- ment, and job performance were 6, 4, 35, 27, 29 and 30% respectively.

Of the 11 hypotheses, 8 were supported.

Table 3 shows that role conﬂict is signiﬁcantly and positively related to emotional exhaustion and depersonalisation. H1a and H1b are there- fore supported. There is, however, no signiﬁ- cant relationship between role conﬂict and diminished personal accomplishment. H1c is therefore not supported. In addition, the results demonstrate that role ambiguity is Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance117 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr 118O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr Table 1. Scale items, reliabilities and conﬁrmatory factor analysis results (n=677) Scale items Standardised loadingst-values Alpha Role conﬂict0.80 I receive an assignment without adequate 0.65 17.42 resources and materials to execute it.

I receive an assignment without the manpower 0.66 17.68 to complete it.

I do things that are apt to be accepted by one 0.76 21.35 person and not accepted by others.

I work with two or more groups who operate 0.51 12.99 differently.

I receive incompatible requests from two or 0.69 18.69 more people.

I have to do things that should be done differently. 0.55 14.13 Role ambiguity0.78 I have clear, planned goals and objectives for my job.* 0.61 15.98 I know exactly what is expected of me.* 0.68 18.37 I know what my responsibilities are.* 0.80 22.82 I feel certain about how much authority I have.* 0.65 17.40 I know I have divided my time properly.* 0.54 13.94 Emotional exhaustion0.90 I feel emotionally drained from my work. 0.66 18.83 I feel used up at the end of the workday. 0.59 16.41 I feel fatigued when I get up in the morning and 0.70 20.45 have to face another day on the job.

Working with people all day is really a strain for me. 0.83 25.85 I feel burned out from my work. 0.86 27.79 I feel frustrated by my job. 0.84 26.60 I feel I am working too hard on my job. 0.57 15.85 I feel like I am at the end of my rope. 0.76 22.94 Depersonalisation0.85 I feel I treat some customers as if they were 0.79 23.88 impersonal objects.

I have become more callous towards people 0.84 26.05 since I took this job.

I worry that this job is hardening me emotionally. 0.84 26.36 I do not really care what happens to some customers. 0.66 18.67 I feel customers blame me for some of their problems. 0.53 14.18 Diminished personal accomplishment0.82 I deal effectively with the problems of my customers.* 0.52 13.60 I feel I am positively inﬂuencing other people’s lives 0.55 14.53 through my work.* I feel very energetic. 0.70 19.35 I can easily create a relaxed atmosphere with my customers.* 0.66 18.25 I feel exhilarated after working closely with my customers.* 0.74 21.19 I have accomplished many worthwhile things in this job.* 0.64 18.12 In my work, I deal with emotional problems very calmly.* 0.59 15.76 Job performance0.77 I am a top performer. 0.59 15.13 I am in the top 10% of frontline employees here. 0.62 16.13 I get along better with customers than do others. 0.73 19.85 I know more about services delivered to customers than others. 0.64 16.77 I know what my customers expect better than others. 0.61 15.84 Each item is measured on ﬁve-point scales. All loadings are signiﬁcant at the 0.01 or better.

*These items were reverse scored. Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance119 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr Table 2. Correlations, means, and standard deviations of composite measures of model constructs and control variables Variables 1 234567 8910111213 1Age 1.000 2Gender−0.090* 1.000 3Education 0.048 0.016 1.000 4Organisational tenure 0.527**−0.080*−0.005 1.000 5Marital status 0.457**−0.098*−0.016 0.248** 1.000 6The number of children 0.558**−0.093*−0.012 0.378** 0.606** 1.000 7Hotel type 0.046−0.075 0.004 0.054 0.029 0.035 1.000 8Role conﬂict−0.045−0.122** 0.194**−0.066−0.005−0.025−0.006 1.000 9Role ambiguity 0.027−0.042 0.040−0.077* 0.122** 0.047 0.089* 0.110** 1.000 10 Emotional exhaustion 0.111**−0.063 0.318**−0.111** 0.090* 0.062 0.154** 0.435** 0.289** 1.000 11Depersonalisation 0.114**−0.010 0.241**−0.098* 0.131** 0.092* 0.171** 0.377** 0.234** 0.753** 1.000 12 Diminished personal 0.090* 0.025 0.155**−0.135** 0.049 0.023 0.023 0.122** 0.486** 0.430** 0.321** 1.000 accomplishment 13 Job performance−0.127**−0.044 0.008 0.014−0.069 0.005−0.111** 0.248**−0.318**−0.010−0.034−0.363** 1.000 Mean 1.75 0.43 2.74 2.02 0.40 1.42 0.55 3.21 2.01 2.72 2.64 2.16 3.79 Standard deviation 0.94 0.50 1.05 0.90 0.49 0.62 0.50 0.78 0.67 0.97 0.99 0.67 0.69 Composite scores for each measure were obtained by averaging scores across items representing that measure. The scores range from 1 to 5. Age and education were measured using ﬁve-point scales. Organisational tenure was measured using a six-point scale. The number of children was measured using a four-point scale. Higher scores indicated older age, more educated, longer tenure and more children. Gender was coded as a binary variable (0 =male and 1 =female). Marital status was coded as a binary variable (0 =single or divorced and 1 =married). Hotel type was also coded as a binary variable (0 =chain hotel and 1 =independently/family- owned and -operated hotel).

*Correlations are signiﬁcant at the 0.05 level.

** Correlations are signiﬁcant at the 0.01 level. Corrrelations without any asterisks are insigniﬁcant. signiﬁcantly and positively related to emo- tional exhaustion, depersonalisation and diminished personal accomplishment, provid- ing empirical support for H2a, H2b and H2c.

As predicted, role conﬂict has a signiﬁcant pos- itive impact on job performance while role ambiguity exerts a signiﬁcant negative effect on job performance. H3a and H3b are therefore supported.

Contrary to our predictions, the results demonstrate that emotional exhaustion is sig- niﬁcantly and positively related to job perfor- mance. H4a therefore receives no empirical support. The relationship between depersonal- isation and job performance is negative but not signiﬁcant. H4b is therefore not supported.

Consistent with our predictions, diminished personal accomplishment has a signiﬁcant negative effect on job performance. H4c is therefore supported.

As depicted in Table 3, several control vari- ables were found to be signiﬁcantly associated with the study variables. Speciﬁcally, gender is negatively related to role conﬂict. This ﬁnding may be attributed to the fact that female employees might have been disciplined using various tools such as counseling and coaching while male employees might have been disci- plined through verbal and written reprimands (cf. Schul and Wren, 1992). According to the results of the path analysis, education is posi- tively related to role conﬂict. This ﬁnding indi- cates that better-educated employees might have had great conﬁdence in their skills and abilities and thus might have not asked for clearer guidance regarding incompatible demands from multiple sources. In addition, organisational tenure is negatively related to role ambiguity while marital status is posi- tively associated with role ambiguity. The ﬁnding regarding the relationship between tenure and role ambiguity suggests that employees with longer tenure might have obtained more information concerning their job-related duties and responsibilities and thus might have experienced lower role ambiguity.

The results in Table 3 indicate that age and education have positive relationships with emotional exhaustion, depersonalisation and diminished personal accomplishment while organisational tenure is negatively associated with the three dimensions of burnout. Theseﬁndings suggest that older and better- educated employees might have been fed up with managing customers’ requests and com- plaints while employees with longer tenure might have learnt how to handle the difﬁcul- ties emanating from emotional exhaustion, depersonalisation and diminished personal accomplishment.

The results reveal that age is negatively asso- ciated with job performance while the number of children has a positive association with job performance. The negative relationship between age and job performance might be attributed to the fact that older employees might have displayed lower job performance because of a number of problems associated with heavy workloads and unsocial work hours. In addition, employees in indepen- dently/family-owned and -operated hotels have high levels of role ambiguity, emotional exhaustion, and depersonalisation and display lower performance in the workplace. This is not surprising because employees in frontline service jobs are highly stressed and may not be capable of carrying out job-related duties and responsibilities effectively because of the lack of training, development and mentoring pro- grammes, and well-prepared job descriptions in such hotels relative to those in chain hotels.

DISCUSSION As discussed before, frontline hotel employees are highly stressed and experience burnout (Ledgerwood et al., 1998; Karatepe and Sokmen, 2006). Little, however, is known about the relationships of role stress and burnout on an organisationally valued job outcome: job performance. This is also true for the effect of role stress on burnout. With this recognition, building on and extending prior research and ﬁlling in voids in the tourism and hospitality literature, this study developed and tested a research model that investigated the effects of role stress and burnout on job per- formance by using a sample of frontline hotel employees in Northern Cyprus as the study setting.

The results were largely supportive of the research hypotheses. Speciﬁcally, two major job demands, role conﬂict and ambiguity, were found to exacerbate frontline employees’ 120O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance121 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr Table 3. Model test results Control variables Standardised parameter and hypotheses estimatest-values Supported/not supported (I) Impact on ROLEC Control variables Age →ROLEC−0.05−0.93 — Gender →ROLEC−0.13−3.50* — Education →ROLEC 0.20 5.19* — Organizational tenure →ROLEC−0.06−1.31 — Marital status →ROLEC 0.02 0.39 — The number of children →ROLEC 0.00 0.04 — Hotel type →ROLEC−0.02−0.42 — R 2=0.06 (II) Impact on ROLEA Control variables Age →ROLEA 0.03 0.64 — Gender →ROLEA−0.03−0.74 — Education →ROLEA 0.04 1.09 — Organisational tenure →ROLEA−0.14−3.03* — Marital status →ROLEA 0.14 2.82* — The number of children →ROLEA−0.01−0.10 — Hotel type →ROLEA 0.09 2.34* — R 2=0.04 (III) Impact on EEXHAUST Control variables Age →EEXHAUST 0.18 4.37* — Gender →EEXHAUST 0.00 0.14 — Education →EEXHAUST 0.23 7.34* — Organisational tenure →EEXHAUST−0.18−4.89* — Marital status →EEXHAUST 0.02 0.62 — The number of children→EEXHAUST 0.01 0.25 — Hotel type →EEXHAUST 0.13 4.31* — Hypotheses ROLEC →EEXHAUST 0.37 11.68* Supported ROLEA→EEXHAUST 0.20 6.48* Supported R 2=0.35 (IV) Impact on DEP Control variables Age →DEP 0.16 3.49* — Gender →DEP 0.06 1.67 — Education →DEP 0.16 4.83* — Organisational tenure →DEP−0.18−4.59* — Marital status →DEP 0.08 1.76 — The number of children →DEP 0.03 0.67 — Hotel type →DEP 0.16 4.91* — Hypotheses ROLEC →DEP 0.34 9.97* Supported ROLEA→DEP 0.14 4.29* Supported R 2=0.27 (V) Impact on DPA Control variables 122O. M. Karatepe and O. Uludag Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res. 10, 111–126 (2008) DOI: 10.1002/jtr Table 3. (Continued) Control variables Standardised parameter and hypotheses estimatest-values Supported/not supported Age →DPA 0.20 4.60* — Gender →DPA 0.05 1.49 — Education →DPA 0.11 3.31* — Organisational tenure →DPA−0.19−4.78* — Marital status →DPA−0.03−0.80 — The number of children →DPA−0.02−0.40 — Hotel type →DPA−0.02−0.52 — Hypotheses ROLEC →DPA 0.05 1.48 Not supported ROLEA→DPA 0.47 14.12* Supported R 2=0.29 (VI) Impact on JPERF Control variables Age →JPERF−0.16−3.44* — Gender →JPERF−0.01−0.31 — Education →JPERF−0.01−0.26 — Organisational tenure →JPERF 0.02 0.57 — Marital status →JPERF−0.05−1.21 — The number of children →JPERF 0.14 3.16* — Hotel type →JPERF−0.09−2.67* — Hypotheses ROLEC →JPERF 0.27 7.15* Supported ROLEA→JPERF−0.21−5.54* Supported EEXHAUST →JPERF 0.15 2.62* Not supported DEP→JPERF−0.06−1.22 Not supported DPA→JPERF−0.31−7.59* Supported R 2=0.30 Age and education were measured using ﬁve-point scales. Organisational tenure was measured using a six-point scale.

The number of children was measured using a four-point scale. Higher scores indicated older age, more educated, longer tenure and more children. Gender was coded as a binary variable (0 =male and 1 =female). Marital status was coded as a binary variable (0 =single or divorced and 1 =married). Hotel type was also coded as a binary variable (0 =chain hotel and 1 =independently/family-owned and -operated hotel).

ROLEC, role conﬂict; ROLEA, role ambiguity; EEXHAUST, emotional exhaustion; DEP, Depersonalisation; DPA, dimin- ished personal accomplishment; JPERF, job performance.

*The t-values demonstrate a statistically signiﬁcant relationship at the 0.05 level or better. The other t-values without any asterisks are insigniﬁcant.

emotional exhaustion and depersonalisation.

The results reported here are consonant with those of Babakus et al. (1999), Hsieh and Hsieh (2003) and Lee and Ashforth (1996). The results indicating that the impact of role conﬂict on emotional exhaustion and depersonalisation appeared to be stronger than that of role ambi- guity lend credence to previous empirical studies (Babakus et al., 1999; Hsieh and Hsieh, 2003). Such ﬁndings elucidate that frontline hotel employees experience emotional exhaus- tion and depersonalisation because they losetheir valued resources (e.g. time and energy) by trying to cope with various problems that emanate from role conﬂict and ambiguity.

The result demonstrating that role ambigu- ity was signiﬁcantly and positively related to diminished personal accomplishment is con- sistent with that of Hsieh and Hsieh (2003).

According to this ﬁnding, frontline employees who are devoid of speciﬁc information concerning their job-related duties and responsibilities display reduced personal accomplishment. In contrast, the results did not lend any empirical support to the relation- ship between role conﬂict and diminished per- sonal accomplishment. As discussed earlier, role conﬂict is one of the unavoidable aspects of frontline service jobs. Under these circum- stances, frontline employees might not have had negative self-evaluations when they had received incompatible demands from multiple sources such as customers and managers. The aforementioned result is also congruent with that of Hsieh and Hsieh (2003).

Consistent with our predictions, it was found that role conﬂict increased frontline employees’ job performance while role ambi- guity had a detrimental impact on their per- formance in the workplace. The ﬁndings reported here corroborate those of prior empir- ical studies (Ross and Boles, 1994; Babin and Boles, 1996). Two conclusions surface from these ﬁndings. First, because of the nature of their boundary-spanning roles, frontline employees are confronted with heightened role conﬂict. Therefore, they might have learnt how to overcome difﬁculties arising from role conﬂict and might have displayed higher job performance (Ross and Boles, 1994). Second, frontline employees having inadequate infor- mation pertaining to their job-related duties and responsibilities cannot perform effectively in the workplace.

In addition, the result demonstrating that diminished personal accomplishment exerted a signiﬁcant negative effect on job performance is not concordant with that of Wright and Bonett (1997). This ﬁnding suggests that front- line employees’ performance in the workplace starts to erode as a result of resource loss and reduced personal accomplishment. The results did not provide any empirical support for the signiﬁcant negative impact of depersonalisa- tion on job performance. The absence of such a signiﬁcant ﬁnding is also evident in prior research (Wright and Bonett, 1997; Wright and Hobfoll, 2004). Surprisingly, it was found that emotional exhaustion ampliﬁed frontline employees’ job performance. This intriguing ﬁnding is consistent with that of Advani et al.

(2005). One tenable explanation for this ﬁnding can be made. The sample of this study con- sisted of frontline employees such as front desk agents, concierges and food servers who were tipped by hotel guests. Therefore,emotional exhaustion might have made them focus more on their job-related activities such as serving guests in a courteous manner and showing empathy and concern, and dealing with their complaints effectively (cf. Van Dyne et al., 2002).

Age, gender, education, organisational tenure, marital status, the number of children and hotel type were treated as control vari- ables in this study to avoid statistical con- founds. The results revealed that none of the control variables confounded the hypothesised relationships.

Managerial implications Several useful implications for hotel managers emerge from the study ﬁndings. First, Maslach (2005) states: ‘Burnout is not a problem of people but of the social environment in which they work’ (p. 99). Therefore, it is important to make speciﬁc changes in working conditions of hotel organisations. In apparent recognition of this, managers need to employ mentors who are likely to deal with a number of problems that stem from role stress and burnout as expe- rienced by frontline employees. Such mentors may hold meetings with department managers (perhaps on a monthly basis) to discuss various problematic cases and come up with effective solutions from which both the organ- isation and the employee will beneﬁt.

Second, most of the hotel organisations in Northern Cyprus are devoid of contemporary human resource managerial practices (Karatepe and Kilic, 2007). Thus, managers are supposed to prepare appropriate job descrip- tions for frontline service positions/revise the existing ones and clearly communicate their expectations to their frontline employees through continuous training programmes.

Such efforts may potentially decrease frontline employees’ role stress in the workplace.

Finally, there is evidence that individual characteristics such as self-efﬁcacy and intrin- sic motivation diminish employees’ burnout (Low et al., 2001; Perrewé et al., 2002). With this realisation, managers should establish ap- propriate selection criteria and offer career opportunities in order to be able to attract self-efﬁcacious and/or intrinsically motiv- ated employees to their organisations. Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance123 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr Consequently, burnout can be reduced once such individuals are employed in hotel organisations.

Limitations and directions for future research The aforementioned results and their implica- tions should be considered with several limi- tations. First, the current study used only one service setting in order to investigate the rela- tionships shown in the research model. Thus, it would be useful if future studies could inves- tigate hypothesised relationships in various tourism and hospitality service settings such as travel agencies and airlines.

Second, all variables were measured using self-report data. This approach is prone to common method variance (Doty and Glick, 1998). Future studies should use multiple sources to avoid such a potential problem. For example, frontline employees’ job perfor- mance can be measured using their supervi- sors’ assessment.

In closing, incorporating emotional disso- nance and supervisor support as the predictors of burnout and job performance into the research model would shed further light on our understanding of the previously men- tioned relationships.

REFERENCES Advani JY, Jagdale SC, Garg AK, Kumar R. 2005.

Antecedents and consequences of ‘burnout’ in services personnel: a case of Indian software pro- fessionals. South Asian Journal of Management 12(3): 21–34.

Akis S, Warner J. 1994. A descriptive analysis of North Cyprus tourism. Tourism Management 15(5): 379–388.

Altinay L. 2000. Possible impacts of a federal solution to the Cyprus problem on the tourism industry of North Cyprus. International Journal of Hospitality Management19(3): 295–309.

Altinay L, Bowen D. 2006. Politics and tourism interface: the case of Cyprus. Annals of Tourism Research33(4): 939–956.

Altinay L, Altinay M, Bicak HA. 2002. Political sce- narios: the future of the North Cyprus tourism industry. International Journal of Contemporary Hospitality Management14(4): 176–182.Anderson JC, Gerbing DW. 1988. Structural equa- tion modeling in practice: a review and recom- mended two-step approach. Psychological Bulletin 103(May): 411–423.

Babakus E, Cravens DW, Johnston M, Moncrief WC.

1999. The role of emotional exhaustion in sales force attitude and behavior relationships. Journal of the Academy of Marketing Science27(1): 58–70.

Babin BJ, Boles JS. 1996. The effects of perceived co-worker involvement and supervisor support on service provider role stress, performance and job satisfaction. Journal of Retailing72(1): 57–75.

Babin BJ, Boles JS. 1998. Employee behavior in a service environment: a model and test of poten- tial differences between men and women. Journal of Marketing62(April): 77–91.

Bakker AB, Demerouti E, Verbeke W. 2004. Using the job demands-resources model to predict burnout and performance. Human Resource Management43(1): 83–104.

Behrman DN, Perreault, WD, Jr. 1984. A role stress model of the performance and satisfaction of industrial salespersons. Journal of Marketing 48(Fall): 9–21.

Brown SP, Peterson RA. 1993. Antecedents and con- sequences of salesperson job satisfaction: meta- analysis and assessment of causal effects. Journal of Marketing Research30(February): 63–77.

Churchill GA, Jr, Ford NM, Walker OC, Jr. 1976.

Organizational climate and job satisfaction in the salesforce. Journal of Marketing Research 13(November): 323–332.

Churchill GA, Jr, Ford NM, Hartley SW, Walker OC, Jr. 1985. The determinants of salesperson perfor- mance: a meta-analysis. Journal of Marketing Research22(May): 103–118.

Cordes CL, Dougherty TW. 1993. A review and integration of research on job burnout. Academy of Management Review18(4): 621–656.

Doty DH, Glick WH. 1998. Common method bias: does common method variance really bias results? Organizational Research Methods1(4):

374–406.

Gaines J, Jermier JM. 1983. Emotional exhaustion in a high stress organization. Academy of Manage- ment Journal26(4): 567–586.

Hartline MD, Ferrell OC. 1996. The management of customer-contact service employees: an empirical investigation. Journal of Marketing60(October):

52–70.

Hobfoll SE. 1989. Conservation of resources: a new attempt at conceptualizing stress. American Psy- chologist44(3): 513–524.

Hobfoll SE. 2001. The inﬂuence of culture, commu- nity, and the nested-self in the stress process:

advancing conservation of resources theory.

337–421.

Hsieh Y-M, Hsieh A-T. 2003. Does job standardiza- tion increase job burnout? International Journal of Manpower24(5): 590–614.

Jackson SE, Schuler RS. 1985. A meta-analysis and conceptual critique of research on role ambiguity and role conﬂict in work settings. Organizational Behavior and Human Decision Processes36(1):

16–78.

Joreskog, K, Sorbom, D. 1996. LISREL 8: User’s Reference Guide. Scientiﬁc Software International, Inc.: Chicago, IL.

Judd CM, Smith ER, Kidder LH. 1991. Research Methods in Social Relations. 6th edn. Holt, Rinehart and Winston, Inc.: Fort Worth, TX.

Karatepe OM, Kilic H. 2007. Relationships of super- visor support and conﬂicts in the work-family interface with the selected job outcomes of frontline employees. Tourism Management28(1):

238–252.

Karatepe OM, Sokmen A. 2006. The effects of work role and family role variables on psychological and behavioral outcomes of frontline employees.

Tourism Management27(2): 255–268.

Karatepe OM, Avci T, Karatepe T, Canozer S. 2003.

The measurement of job satisfaction: an empiri- cal study of frontline employees in the Northern Cyprus hotel industry. International Journal of Hospitality and Tourism Administration4(1): 69– 85.

Karatepe OM, Uludag O, Menevis I, Hadzi- mehmedagic L, Baddar L. 2006. The effects of selected individual characteristics on frontline employee performance and job satisfaction.

Tourism Management27(4): 547–560.

Kilic H, Okumus F. 2005. Factors inﬂuencing pro- ductivity in small island hotels: evidence from Northern Cyprus. International Journal of Contem- porary Hospitality Management17(4): 315–331.

Kim HJ, Shin KH, Umbreit WT. 2007. Hotel job burnout: the role of personality characteristics.

International Journal of Hospitality Management 26(2): 421–434.

Ledgerwood CE, Crotts JC, Everett AM. 1998.

Antecedents of employee burnout in the hotel industry. Progress in Tourism and Hospitality Research4(1): 31–44.

Lee RT, Ashforth BE. 1996. A meta-analytic exami- nation of the correlates of the three dimensions of job burnout. Journal of Applied Psychology81(2):

123–133.

Low GS, Cravens DW, Grant K, Moncrief WC. 2001.

Antecedents and consequences of salesperson burnout. European Journal of Marketing35(5/6):

587–611.MacKenzie SB, Podsakoff PM, Ahearne M. 1998.

Some possible antecedents and consequences of in-role and extra-role salesperson performance.

Journal of Marketing62(July): 87–98.

Maslach C, Jackson SE. 1981. The measurement of experienced burnout. Journal of Occupational Behavior2(2): 99–113.

Maslach CM. 2005. Understanding burnout: work and family issues. In From Work–Family Balance to Work–Family Interaction: Changing the Metaphor, Halpern D (ed). Lawrence Erlbaum Associates, Inc: Mahwah, NJ; 99–114.

McGee G, Ferguson C, Seers A. 1989. Role conﬂict and role ambiguity: do the scales measure these two constructs? Journal of Applied Psychology74(5):

815–818.

McGorry SY. 2000. Measurement in a cross-cultural environment: survey translation issues. Qualita- tive Market Research: an International Journal3(2):

74–81.

Nunnally JC. 1978. Psychometric Theory. 2nd edn.

McGraw-Hill Book Company: New York.

Okumus F, Altinay M, Arasli H. 2005. The impact of Turkey’s economic crisis of February 2001 on the tourism industry in Northern Cyprus. Tourism Management26(1): 95–104.

Perrewé PL, Hochwarter WA, Ralston DA, Westman M, Rossi AM, Vollmer G, Wallace A, Tang M, Wan P, Maignan I, Van Deusen CA, Castro SL. 2002. Are work stress relationships universal? A nine-region examination of role stressors, general self-efﬁcacy, and burnout.

Journal of International Management8(2): 163–187.

Piko BF. 2006. Burnout, role conﬂict, job satisfaction and psychosocial health among Hungarian health care staff: a questionnaire survey. Interna- tional Journal of Nursing Studies43(3): 311– 318.

Rizzo JR, House RJ, Lirtzman SI. 1970. Role conﬂict and ambiguity in complex organizations. Admin- istrative Science Quarterly15(2): 150–163.

Ross LE, Boles JS. 1994. Exploring the inﬂuence of workplace relationships on work-related atti- tudes and behaviors in the hospitality work envi- ronment. International Journal of Hospitality Management13(2): 155–171.

Schul PL, Wren BM. 1992. The emerging role of women in industrial selling: a decade of change.

Journal of Marketing56(July): 38–54.

Scott J. 1997. Chances and choices: women and tourism in Northern Cyprus. In Gender, Work and Tourism, Sinclair MT (ed.). Routledge: London; 60–90.

Smith CS, Tisak J, Schmieder RA. 1993. The mea- surement properties of the role conﬂict and role ambiguity scales: a review and extension of the Role Stress, Burnout and Their Effects on Frontline Hotel Employees’ Job Performance125 Copyright © 2007 John Wiley & Sons, Ltd.Int. J. Tourism Res.10, 111–126 (2008) DOI: 10.1002/jtr empirical research. Journal of Organizational Behav- ior14(1): 37–48.

Van Dyne L, Jehn KA, Cummings A. 2002. Differ- ential effects of strain on two forms of work per- formance: individual employee sales and creativity. Journal of Organizational Behavior23(1):

57–74.

Wright TA, Bonett DG. 1997. The contribution of burnout to work performance. Journal of Organi- zational Behavior18(5): 491–499.

Wright TA, Cropanzano R. 1998. Emotional ex- haustion as a predictor of job performance andvoluntary turnover. Journal of Applied Psychology 83(3): 486–493.

Wright TA, Hobfoll SE. 2004. Commitment, psy- chological well-being and job performance: an examination of conservation of resources (COR) theory and job burnout. Journal of Business and Management9(4): 389–406.

Yavas U, Karatepe OM, Babakus E, Avci T. 2004.

Customer complaints and organizational responses: a study of hotel guests in Northern Cyprus. Journal of Hospitality and Leisure Market- ing11(2/3): 31–46.