Emotional exhaustion and its consequences for hotel service quality: the critical role of workload and supervisor support

ABSTRACT

This study develops and tests a research model which investigates the impact of workload and supervisor support on emotional exhaustion and analyzes its consequences for critical employee and job-related outcomes, such as affective organizational commitment, intention to leave, and the quality of service performance. A survey instrument was used to collect data from 162 hotel employees in Poland. The study findings confirmed that workload and supervisor support are both significant antecedents of emotional exhaustion, which, in turn, exerts a significant impact on hotel employees’ leaving intention that reduces the quality of service performance. This study also offers useful implications including trainings and emotional management courses to protect hotel employees from emotional exhaustion and maintain committed workforce who are likely to deliver high service quality.

调查工具用于收集波兰162名酒店员工的数据. 研究结果证实, 工作负荷和主管支持都是情绪衰竭的显著前因, 而情绪衰竭又会对酒店员工的离职意愿产生显著影响, 进而降低服务绩效. 这项研究还提供了一些有用的建议, 包括培训和情绪管理课程, 以保护酒店员工免受情绪枯竭的影响, 并维持有可能提供高质量服务的忠诚员工

KEYWORDS:

• Hotel employees
• emotional exhaustion
• affective commitment
• turnover intention
• service quality

KEYWORDS:

• 本研究发展并检验一个研究模式
• 探讨工作负荷与主管支持对员工情绪衰竭之影响
• 并分析其对关键员工及工作相关结果之影响
• 如情感组织承诺
• 离职意愿及服务品质表现

Introduction

The success of hotels largely depends on their employees who offer top-quality service to their guests (Castro-Casal et al., 2019). Yet, although highly skilled and attentive hotel employees are important to succeed (Nain, 2018), they are especially exposed to stress due to the characteristics of their jobs (Hwang et al., 2014), such as the 24/7 nature of work, anti-social and long working hours, unpredictable shifts, few breaks, excessive workload, etc. All this builds up especially high levels of stress among employees (Chan et al., 2019; Haldorai et al., 2019; Sampson & Akyeampong, 2014).

It is worth underlining that although the hospitality industry is among the most stressful work environments (e.g., Kim et al., 2007; O’Neill & Xiao, 2010), the hotel industry, as a highly interactive service industry (Li et al., 2020), is particularly prone to stress (Bora, 2017), and yet it has drawn little attention from researchers (Wen et al., 2020). There are relatively few empirical studies that have investigated the issue of work stress in the hotel context (Hwang et al., 2014; McNamara et al., 2011), making this topic still not fully understood (Sampson & Akyeampong, 2014) and explored, thus deserving further research (Chiang et al., 2010). Particularly, the phenomenon of burnout deserves more research attention (Cordes & Dougherty, 1993), including emotional exhaustion as “the most obvious manifestation” of burnout (Maslach et al., 2001, p. 403) that is not only a serious outcome in itself, resulting in depletion of resources (Yavas et al., 2008), but it also has far-reaching effects on the effectiveness of individuals and organizations.

Against this backdrop, this study aims to investigate the impact of workload and supervisor support on hotel employees’ emotional exhaustion and to explore its possible employee and job-related outcomes, including affective commitment, turnover intention and the quality of service performance.

The present study is relevant for at least several reasons. Today’s hotel industry globally faces a vast set of major challenges, such as providing and maintaining excellent and exceptional guest service, labor shortage, and retaining quality employees (Nain, 2018). This is visible also in Poland, where the quality gap in hotel services still exists (Gołąb, 2009; Grobelna & Marciszewska, 2013; Wrukowska, 2019), and the hotel industry suffers from a disproportionately higher rate of staff turnover (Ratajczyk, 2014; Węglarz & Kowalczyk, 2019). Given the fact that contact employees create one of the most valuable assets in an organization necessary to gain a sustainable competitive advantage (Karatepe et al., 2019), these challenges seriously threaten hotel’s competitiveness. This is because, despite playing a critical role in delivering high service quality, overburdened hotel contact employees are likely to experience emotional exhaustion (Li et al., 2020), which may influence their behavior and affect customers’ perception of the service quality (Castro-Casal et al., 2019) as well as induce a lower sense of belonging to the organization (Li et al., 2020) and force them to think about leaving it (Yavas et al., 2008). Therefore, identifying factors that may affect emotional exhaustion and any methods to reduce hotel employees’ stress is increasingly vital for both hotel practitioners and researchers (Chiang et al., 2010).

Additionally, previous studies referring to hotel employees’ burnout/emotional exhaustion have been conducted in different countries, e.g., Nigeria (Karatepe & Aleshinloye, 2009), Cameroon (Karatepe, 2015), Turkey (Yavas et al., 2008), Taiwan (Yang, 2010) or the USA (Kang et al., 2010; Kim et al., 2007). Moreover, although there are studies that have tested the relationships between emotional exhaustion and employees’ performance, still some of them yield mixed results (see e.g., Essawy, 2016; Karatepe & Aleshinloye, 2009; Karatepe & Uludag, 2008; Yavas et al., 2008). Thus, this study addresses the need to verify this ambiguous relationship and tests the impact of emotional exhaustion on the quality of service performance extending the research stream of burnout to other societies, which may contribute to richer insights into the work–stress process in the hospitality setting. To the best of the Author’s knowledge, this is the first study that attempts to examine simultaneously the relationships between the proposed study constructs using data from hotel employees in Poland.

Literature review

Although the tourism and hospitality industry seems to create an unusual workplace to examine stress, such a work context contains stressors for those working within it (Ross, 1997) and creates antecedents of burnout (Hon et al., 2013; O’Neill & Xiao, 2010). Stress occurs in situations where employees’ well-being is detrimentally affected by their failure to cope with the environment’s demands (Faulkner & Patiar, 1997).

Burnout is a metaphor describing a state of mental weariness (Schaufeli & Bakker, 2004) and is defined as “a syndrome of emotional exhaustion and cynicism that occurs frequently among individuals who do ‘people-work’ of some kind” (Maslach & Jackson, 1981, p. 99). Job burnout consists of three sub-constructs (Maslach & Jackson 1981, p. 99), namely emotional exhaustion, depersonalization, and diminished personal accomplishment. Emotional exhaustion (EM_EXH) refers to the feelings of energy depletion (Chen & Chen, 2012) and is defined as lack of energy and a feeling that individual’s emotional resources are used up because of excessive psychological demands (Kim et al., 2009, 2007). Depersonalization describes negative and uncaring attitudes toward others (Kang et al., 2010), whereas diminished personal accomplishment refers to the feeling of incompetence and lack of achievements in one’s work (Chen & Chen 2012). Although each sub-construct captures its unique aspect of burnout (Kim et al., 2007), emotional exhaustion is a stress dimension perceived as “the key to the experience of burnout, and the first stage of the burnout process” (Cordes & Dougherty, 1993, p. 640).

EM_EXH is mostly associated with occupations with high intensity of interpersonal interactions (Ducharme et al., 2007), such as in the hotel industry, where a high degree of interaction between hotel employees and guests takes place (Karatepe & Kilic, 2007), and guests’ satisfaction frequently depends on these encounters (Faulkner & Patiar, 1997). Therefore, hospitality employees must display organizationally expected emotions during a service encounter even in handling difficult interactions with demanding hotel guests (emotional labor) (Gursoy et al., 2011; Lam & Chen, 2012). They must also perform the required esthetic abilities and skills (esthetic labor), presenting a good image when serving customers, which may create additional stress and make employees feel particularly burdened (Tsaur & Tang, 2013). Moreover, a culture of long, unsocial working hours, and shift duties (Wong & Ko, 2009) makes the hotel work physically demanding (see Hwang et al., 2014).

Therefore, taking into consideration the emotionally, mentally, and physically demanding nature of hotel jobs, EM_EXH is an adequate indicator of burnout in the hotel setting. As noted before, EM_EXH may occur when job demands exceed employee’s abilities and resources (Cho et al., 2014) leading to many negative attitudinal and behavioral outcomes. There is empirical evidence in the tourism and hospitality industry that emotional exhaustion negatively influences job satisfaction and affective commitment (Karatepe & Uludag, 2007), customer orientation (Choi et al., 2014; Lee et al., 2012), service recovery performance (Choi et al., 2014), and that it is positively related to employees’ turnover intention (Karatepe & Aleshinloye, 2009; Karatepe & Uludag, 2007; Yavas et al., 2008).

Based on the above, better understanding the frontline employees’ EM_EXH and its consequences is still a major concern for both hospitality researchers and managers. Therefore, this study develops and tests a research model (Figure 1) where the impact of workload (W_LOAD; as a job demand) and supervisor support (S_SUPPORT; as a job resource) on emotional exhaustion (EM_EXH) is investigated. The effect of EM_EXH on critical employee and job-related outcomes, such as affective organizational commitment (AFCOM), intention to leave (IN_LEAVE), and the quality of service performance (QUALITY OF SERV_PERF), is also analyzed.

Figure 1. Research model

Note: S_SUPPORT (Supervisor Support); W_LOAD (Workload); EM_EXH (Emotional Exhaustion); AFCOM (Affective Organizational Commitment); QUALITY OF SERV_PERF (Quality of Service Performance); IN_LEAVE (Intention to Leave)

The conceptualization also considers the effect of AFCOM on IN_LEAVE and explores the relationships between AFCOM and IN_LEAVE with QUALITY OF SERV_PERF. The abovementioned relationships are tested using data from hotel employees working in guest contact positions in hotels of Tricity (a tourist destination in Northern Poland) as a setting. Operational employees were chosen as they are viewed as “more directly and immediately responsible for satisfying guests’ needs” (Faulkner & Patiar, 1997, p. 102).

Both the COR theory and the JD-R model provide feasible guidelines to develop the studied relationships. The COR theory is used as a “principal explanatory mechanism” to understand the burnout process (Hobfoll, 2001, p. 347). It states that that people seek to obtain, retain, and protect fundamental resources, such as objects, personal characteristics, conditions, or energies (Hobfoll, 2001). Stress is a reaction to the environment where resources are threatened with loss, resources are actually lost, or individuals fail to gain sufficient resources following significant resource investment (of time, energy, lost opportunities, etc.) (Hobfoll, 2001). Particularly, the resource loss is a critical ingredient in the stress process (Hobfoll, 2001) resulting in certain behavioral and attitudinal outcomes, such as turnover intentions, erosion of organizational commitment, job involvement, and satisfaction, etc. (Lee & Ashforth, 1996). Many previous studies in the hospitality setting used the conservation of resources framework to show how certain resources and demands are associated with employees’ burnout (emotional exhaustion) and/or how it influences selected outcomes (e.g., Karatepe & Aleshinloye, 2009; Karatepe & Uludag, 2007; Yavas et al., 2008).

The Job Demands-Resources (JD-R) model is also among widely known theories explaining the burnout process (Kim et al., 2009). Generally, it postulates that high job demands elevate job strain and lead to employee burnout, negatively affecting their attitudes and behaviors (Kim et al., 2009). By contrast, the availability of job resources helps employees to cope more efficiently with the demands of their work, preventing the negative outcomes, such as emotional exhaustion/burnout (Karatepe, 2010). A number of previous studies have also successfully adopted the JD-R model to investigate the antecedents and consequences of burnout/emotional exhaustion in the hospitality setting (e.g., Karatepe, 2010; Kim et al., 2009).

Summarizing, it follows from the COR theory and the JD_R model that employees confronted with high job demands might drain their scarce resources (e.g., time, energy) and develop high levels of exhaustion, which may result in certain negative attitudinal and behavioral outcomes, whereas availability of job resources may aid stress resistance and help employees to cope better with emotional exhaustion and better manage difficulties associated with it.

Development of hypotheses

Antecedents of emotional exhaustion

Social support may be critical in counterbalancing employees’ experience of occupational stress (Sampson & Akyeampong, 2014), and it is one of the most commonly desired resources by employees (Boukis et al., 2020). Particularly, supervisor support as a viable variable of social support (Karatepe & Olugbade, 2009) may weaken the emotional exhaustion of employees confronted with high levels of stress (Karatepe, 2010). By contrast, a poor working relationship with supervisors may contribute to work-related stress (Sampson & Akyeampong, 2014). A lack of support from supervisors is even more related to burnout than a lack of support from coworkers (Maslach et al., 2001).

A supervisor is the first representative of the organization for frontline hotel employees (King & Garey, 1997), who are especially sensitive to how fairly they are treated by their employers. Respectful treatment may help hotel employees to handle better challenges and stress at work (Lam & Chen, 2012). Lambert and Hogan’s study (Lambert & Hogan, 2009) confirmed that supervisor support had negative effects on job stress. Karatepe and Kilic (2007) study revealed that having supervisor support in the workplace alleviated frontline hotel employees’ work-family conflict and increased their job satisfaction. Hon et al. (2013) revealed that employees who have a lot of advice and support from their supervisors could channel work stress into creativity. The results of meta-analysis by Lee and Ashforth (1996) showed a negative correlation between supervisor support and emotional exhaustion. It is also evidenced that low supervisor support contributes to emotional exhaustion (Kalliath & Beck, 2001). Therefore, based on the literature review and empirical findings, the following hypothesis is proposed:

H1. Supervisor support is negatively related to hotel employees’ emotional exhaustion.

The experienced workload and time pressure are strongly related to burnout, particularly to the exhaustion dimension (Maslach et al., 2001). Workload is perceived as “the sheer volume of work required of an employee” (Spector & Jex, 1998, p. 358) to be done during the specific time (Cordes & Dougherty, 1993). Workload may be connected with anxiety and worry whether someone is able to perform all of the required work or with frustration, since having so much to do, employees are forced to neglect some aspects of their job and life (Spector & Jex, 1998).

According to the literature, quantitative work overload is an important determinant of emotional exhaustion (Cordes & Dougherty, 1993), as “giving too much too long” is energy-depleting (Schaufeli & Bakker, 2004, p. 297). The hotel industry belongs to those with a heavy workload (Haldorai et al., 2019; Wen et al., 2020). It is proven that hotel employees who perceive their workload as high are likely to experience work stress, particularly when the job demands do not match the employees’ capacity to handle them (Sampson & Akyeampong, 2014). Faulkner and Patiar (1997) found “too much work” to be a common stressor for hotel operational staff, and Hwang et al. (2014) identified problems related to “too much work” and “working long hours” as one of the critical aspects of work stress among hotel employees. Thus, unsurprisingly, employees who face work overload may be more likely to experience emotional exhaustion (Wen et al., 2020), which is also confirmed in empirical findings of previous studies. For example, Essawy (2016) in his study among employees of Quick Service Restaurants (QSRs) confirmed a positive association between overloaded employees and their emotional exhaustion. Based on the above, the following hypothesis is proposed:

H2. Workload is positively related to hotel employees’ emotional exhaustion.

Consequences of emotional exhaustion

The research model depicted in Figure 1 shows that emotional exhaustion decreases affective organizational commitment. It is underlined that work-related stress may lead to many negative outcomes, including a decrease in organizational commitment (Sampson & Akyeampong, 2014). Organizational commitment may be conceptualized as a three-component model, including affective, continuance, and normative commitment (Allen & Meyer, 1990). These three components reflect distinct psychological states, and employees may experience each of them to different degrees (Allen & Meyer, 1990). Affective commitment is of special interest in this study. It refers to emotional attachment to the organization (Allen & Meyer, 1990; Somers, 1995). This psychological bond between employees and the organization (Lambert & Hogan, 2009) is called “sentimental commitment” (Bilgin & Demirer, 2012, p. 471).

The COR theory strongly supports the relationship between emotional exhaustion and affective organizational commitment (Karatepe & Uludag, 2007; Lee & Ashforth, 1996). Particularly, Karatepe and Uludag (2007) showed that when hotel employees lose resources or resources are threatened with loss, they become emotionally exhausted, which in turn influences their affective commitment. Thus, the following hypothesis is proposed:

H3. Hotel employees’ emotional exhaustion is negatively related to affective organizational commitment.

The research model (Figure 1) illustrates that emotional exhaustion leads to lowering the quality of service performance. According to the COR theory, when employees are prone to a high level of burnout, erosion of job performance occurs (Karatepe & Uludag, 2008). Specifically, emotional exhaustion may impair employees’ performance and hinder effective customer service (Yavas et al., 2008). Emotionally exhausted employees will not be able to work well (Prajogo, 2019). They lack certain resources needed to enhance their job performance (Karatepe & Uludag, 2008). Thus, unsurprisingly, their customer-oriented attitudes and behaviors decline, including interest in customers’ needs (Lee et al., 2012). Such employees may be less likely to serve guests appropriately, which is critical, given the fact that delivering a high-quality service that generates a positive customer’s experience is fundamental to the success of today’s hotels (Castro-Casal et al., 2019). Evidence has shown that emotional exhaustion is closely associated with service quality. For example, Choi et al. (2014), in their study conducted in the tourism context, confirmed a negative relationship between emotional exhaustion of frontline employees (including those working in tourist hotels) and both service recovery performance and customer orientation. Similarly, Essawy (2016) proved a negative relationship between emotional exhaustion and service recovery performance of QSRs employees in Egypt, while in the recent work by Wang (2020) among service employee of international hotels in Taiwan, a negative impact of emotional exhaustion on service quality was revealed. In view of the literature review and the rationale of empirical evidence, the following hypothesis is proposed:

H4. Emotional exhaustion is negatively related to the quality of service performance.

The research model (Figure 1) also demonstrates that emotional exhaustion is positively related to the turnover intention. There are many empirical findings in hospitality confirming that job stress may be a reason for employees’ increased intention to leave (e.g., Fong et al., 2018; Gök et al., 2017; Huang et al., 2018). According to the COR theory, employees are willing to leave the organization to retain their scarce and limited resources; thus, for those who experience emotional exhaustion (Karatepe & Uludag, 2007), turnover may be a final solution to minimize the threats to resource loss (Walsh, 2011) and to achieve a better balance (Wen et al., 2020). Understandably, employees with high emotional exhaustion will not try to keep their jobs (Prajogo, 2019). Emotional exhaustion was also found to affect significantly and positively turnover intentions in previous studies conducted among hotel employees in Turkey (Yavas et al., 2008), Nigeria (Karatepe & Aleshinloye, 2009) or Northern Cyprus (Karatepe & Uludag, 2007). Therefore, based on the theoretical framework and empirical evidence, the following hypothesis is proposed:

H5. Hotel employees’ emotional exhaustion is positively related to their intention to leave.

The degree of organizational commitment is also a crucial factor in the decision to stay or leave (Choi, 2006). In a three-dimensional model of organizational commitment, common to all of its components, there is a link between an employee and the organization which reduces the probability of staff turnover; however, the nature of this link is different in each of these approaches (Allen & Meyer, 1990). In the affective attachment approach, organizational commitment refers to employees’ accepting the organizational goals and values, to their willingness to make an effort on behalf of the organization and to a strong desire to maintain their membership in the organization (Mowday et al., 1979) due to strong, emotional bonds with it (Sajjad & Nas, 2013). Thus, given the fact that affective commitment constitutes a dimension of happiness at work (Castro-Casal et al., 2019), naturally, employees with high affective commitment stay because they want to. There is some empirical evidence, also in hospitality, of a negative relationship between affective organizational commitment and turnover intention (Zopiatis et al., 2014; Yang, 2010); hence, the following hypothesis is proposed:

H6. Hotel employees’ affective organizational commitment is negatively related to their intention to leave.

As seen in Figure 1, affective organizational commitment leads to heightened service quality performance. Committed employees are more likely to perform work well and to put forth effort on behalf of their organization (Lambert & Hogan, 2009), resulting in an increased quality of service. Employees with strong affective commitment are motivated to a higher level of performance, making a more meaningful contribution than those who expressed other types of commitment (Alnıaçık et al., 2013). Thus, affective commitment may be perceived as the most appropriate predictor of employees’ performance (Sajjad & Nas, 2013). Empirical evidence in a study by Babakus et al. (2003) demonstrated a positive relationship between frontline employees’ affective organizational commitment and service recovery performance in the banking industry. Similarly, in the hospitality setting, the results of a study by Castro-Casal et al. (2019), conducted among hotel employees in Spain, proved a positive association between affective commitment and extra-role service of frontline hotel employees. Based on the above, the following hypothesis is proposed:

H7. Hotel employees’ affective organizational commitment is positively related to the quality of service performance.

It is underlined that employees with turnover intention deliver poor service and decrease organizational competitiveness (Karatepe & Ngeche, 2012), as they may lack motivation to perform well (Gardner & Hom, 2016). It is empirically proved that employees with leaving intentions concentrate on work less than usual, are disinterested in cooperation with customers, perform only the required minimum, and decrease their productivity (Gardner & Hom, 2016). Generally, such employees “do not have their hearts in the work being done” (Lambert & Hogan, 2009, p. 465), which may finally have a detrimental effect on their performance. As discussed in a study by Javed et al. (2014), intention to leave is negatively related to performance, as also confirmed by some previous hospitality studies. For example, in a study by Al-Badarneh et al. (2019), among hotel employees in Jordan, intention to leave had a detrimental impact on their job performance. Therefore, the following hypothesis is proposed:

H8. Turnover intention among hotel employees is negatively related to the quality of service performance.

Methodology

Data collection procedure

The hypotheses were tested among contact employees working in hotels located in the Pomeranian Voivodeship, Northern Poland, which is one of the most visited tourist destinations in Poland and also one of the biggest accommodation centers in the country (Regionalny Program Strategiczny w zakresie atrakcyjności kulturalnej i turystycznej. Pomorska Podróż, 2013). The above-average attractiveness and well-developed tourist infrastructure contribute to increasing tourist arrivals in the region. In 2017, Pomerania was visited by nearly 9 million tourists (Gospodarka Turystyczna Regionu Pomorskiego, 2018). Understandably, the economic potential of the region is also based on dynamic tourism development. Unfortunately, although the quality of service has become a crucial factor in gaining a sustainable competitive advantage in the hotel industry (Al-Ababneh, 2016), common problems in this industry (e.g., long working hours, high labor intensity, low pay, and other demands) also affect hotels in Poland (Employment, wages and salaries in national economy, 2019; Wszendybył-Skulska, 2014). This includes the Pomeranian Region, where the accommodation and catering sector (including hotels) has the lowest salaries of any sections of economy in the region (Statistical Bulletin of Pomorskie Voivodship – II quarter, 2017). Understandably, work is very stressful, and there is high staff turnover (Olszewski-Strzyżowski & Dróżdż, 2014), which may have its consequences in hotel employees’ attitudes and behaviors influencing guests’ perception of service (Szafrański & Piasta, 2019).

This study was conducted in hotels of a three-star standard and above. This was dictated by greater access to contact staff, as in the Polish administrative regulation those hotels (a three-star standard and above) are full-service establishments with different operational departments where many additional services demanding personal contact (e.g., restaurant services) are obligatory (Rozporządzenie Ministra Gospodarki i Pracy z dnia 19 sierpnia, 2004). Moreover, data collected from various departments may reduce a single-department bias, further supporting the external validity of the study (Chang & Teng, 2017). A convenience sample of hotels was used. Seven hotels were chosen as the target group due to their availability and previous cooperation and/or familiarity between the researcher and management of these hotels (Kim, 2008; Lam & Chen, 2012). According to the information received from hotel managers, the total number of contact employees who have frequent and extensive contact with customers, spending most working time dealing with guests, was 235. A detailed explanation of both the study aim and the questionnaire was given to hotel managers, who were also instructed to distribute survey questionnaires to contact employees in their hotels. All questionnaires included information about the assurance of confidentiality and were distributed to hotels in a pack containing a cover letter, an appropriate number of questionnaires, and return envelopes. Respondents were requested to fill in anonymous questionnaires in a self-administrated manner. A total of 235 questionnaires were distributed, and 162 valid questionnaires were collected at a response rate of 68.9%.

Measures

Each of the constructs shown in Figure 1, except the quality of service performance, was operationalized using scales derived from previous extensive research available in the literature. Supervisor support was assessed using a four-item scale taken from Karatepe and Olugbade (2009), who based on Beehr et al. (1990), and used this scale in their study conducted among hotel employees of Abuja (Nigeria). Workload was measured using a five-item scale developed by Spector and Jex (1998). This scale was also adopted in previous studies in the tourism and hospitality setting (e.g., Kim et al., 2007). Emotional exhaustion was operationalized using eight items in line with Karatepe and Uludag (2007), who based on Maslach and Jackson (1981) and conducted their study among hotel employees in Northern Cyprus. Affective organizational commitment was measured using five items adopted from Mowday et al. (1979). Intention to leave was assessed by three items from Boshoff and Allen (2000). This measure was also used in other studies in the hotel setting (e.g., Karatepe & Sokmen, 2006; Park & Gursoy, 2012).

Taking into account that the perceived service quality is closely related to employee performance (Lundberg et al., 2009) and that the efficiency of customer service is evaluated through customer feedback (Niu, 2010), the scale items used to measure the quality of service performance were developed and operationalized by the Author. Five items were identified through careful literature review and individual in-depth interviews conducted prior to the study with hotel professionals, supported by consultation with management and marketing academics from the tourism and hospitality fields. As a result, the following five statements reflecting the commonly observed hotel guests’ feedback on high-quality service experience were identified and accepted by both academics and practitioners:

1. Hotel guests often praise me for the excellent quality of service delivered to them.

2. I rarely get complaints related to my work from guests.

3. Hotel guests who come to me with requests/problems often thank me for the satisfactory solutions and help.

4. Hotel guests whom I have served often declare that they will recommend our hotel for good service.

5. Hotel guests whom I have served often return to our hotel.

The quality of service performance was measured by self-reports from employees, as despite a large amount of literature on service quality, there is still a call for more research on employees’ perception of that point (Al-Ababneh, 2016). Moreover, although it is argued that the use of a self-report measure may lead to inflated results (Karatepe & Kilic, 2007; Karatepe & Sokmen, 2006), and some other measures are recommended (e.g., guests’ or supervisory assessments), there is evidence that self-reported measures of performance do not yield significantly different results from other “more objective” ones (Churchill et al., 1985, p. 117). It is underlined that front-line employees are “in the best position to evaluate performance outcomes, and their perceptions typically converge with those of the customers” (Babakus et al., 2003, p. 278). This may be because, as tipped employees, they are particularly sensitive to guests’ evaluation (King & Garey, 1997). Indeed, studies by both King and Garey (1997) and Yee et al. (2008) emphasized that employees’ evaluation of service quality strongly correlated with guests’ perception, which proved that service employees are relevant informants of service quality and capable of evaluating it in the customers’ terms (Yee et al., 2008). Moreover, the validity of performance assessments provided by employees’ self-reports may be supported by the fact that the data are collected anonymously (Babakus et al., 2009).

All scale items were measured on a five-point scale, ranging from “1 = strongly disagree” to “5 = strongly agree,” except workload, where items were rated on a five-point scale ranging from “1 = never” to “5 = always.”

The survey instrument was originally prepared in English and then translated into Polish via the back-translation method by competent translators with fluent, written, and spoken, knowledge of both languages in translation. Feedback from a pilot test group comprising hotel employees from operational departments confirms the understanding and adequacy of well-worded items; therefore, no specific changes were made in the questionnaire.

According to the two-step procedure suggested by Anderson and Gerbing (1988), confirmatory factor analysis (CFA) was performed to assess the measurement model, and then structural equation modeling (SEM) was applied to test the research hypotheses in the structural model.

As study data were collected through self-report questionnaires, the Common Method Bias may have influenced the results (Chen et al., 2014); therefore, a combination of Lindell and Whitney’s test and Harman’s one-factor analysis was also applied.

The study used IBM SPSS Statistics 21 for descriptive statistical and reliability analysis, and R version 3.0.2., package lavaan 0.5–15 was used for confirmatory factor analysis and structural equation modeling (Rosseel, 2012).

Results

Respondents’ profile

Female respondents accounted for the majority (64.2%) of all the surveyed employees. Among the respondents, 74.1% were less than 41 years old; 50.4% had higher education, followed by 42.6% who graduated from secondary school. With respect to the total work experience, 56.8% of the respondents had been working in the hospitality industry for 1 to 10 years, followed by 34.6% with working experience of over 10 years. Respondents with such long working experience in the industry may greatly enhance the feedback value on the survey questions. 47.5% had been with the current organization for between 1 and 5 years, followed by 33.3% working in the current hotel for over 10 years and 19.1% working for up to 1 year.

Measurement model

All measures were subjected to confirmatory factor analysis to provide support for the issue of dimensionality, convergent, and discriminant validity (Karatepe & Sokmen, 2006). The measurement items with standardized loadings below <0.7 were removed (Hair et al., 2010; Hamid et al., 2011). Specifically, two items from both S_SUPPORT and QUALITY OF SERV_PERF were deleted, and three items from W_LOAD had to be dropped.

After scale purification, the results indicated a relatively reasonable fit of the six-factor model to the data on the basis of a variety of fit statistics applied in this study, in accordance with the literature recommendation (Bagozzi & Yi, 2012; Hooper et al., 2008) and the previous empirical research (e.g., Babakus et al., 2003; Karatepe & Uludag, 2007). The model fit statistics are as follows: χ2 = 313.726; df = 174; χ2/df (the normed chi-square) = 1.80; GFI (Goodness of Fit Index) = 0.85; NFI (Normed Fit Index) = 0.88; NNFI (Non-normed Fit Index) = 0.93; CFI (Comparative Fit Index) = 0.94; RMSEA (Root Mean Square Error of Approximation) = 0.07 and, additionally, RMR (Root Mean Square Residual) = 0.05. The magnitudes of standardized loadings ranged from 0.741 to 0.939, and all t-values, ranging from 9.877 to 15.097, were significant (Table 1). The abovementioned results, especially the magnitudes of the loadings and significant t-values provide support for convergent validity.

Table 1. Measurement of the study constructs in the light of CFA

Scale items	Standardized loadings	t-value	CR	AVE
Supervisor support (S_SUPPORT)			0.89	0.75
My supervisor is willing to listen to my personal problems.	0.793	11.765
My supervisor is easy to talk to.	0.939	15.097
I can depend on my supervisor for help when things get tough at work.	0.852	13.018
Affective Organizational Commitment (AFCOM)			0.89	0.72
I really care about the future of this hotel	0.874	13.653
I am proud to tell others that I work for this hotel.	0.851	13.099
I am willing to put in a great deal of effort beyond that normally expected in order to help the hotel to be successful.	0.814	12.243
For me, this is the best of all possible organizations for which to work.	0.741	10.687
Quality of Service Performance (QUALITY OF SERV_PERF)			0.80	0.58
Hotel guests who come to me with requests/problems often thank me for the satisfactory solutions and help.	0.742	9.877
Hotel guests whom I have served often declare that they will recommend our hotel for good service.	0.781	10.499
Hotel guests whom I have served often return to our hotel.	0.751	10.011
Intention to leave (IN_LEAVE)			0.92	0.79
I often think about leaving this hotel.	0.887	14.152
It would not take much to make me leave this hotel.	0.904	14.612
I will probably be looking for another job soon.	0.868	13.684
Emotional Exhaustion (EM_EXH)			0.92	0.66
I feel emotionally drained from my work.	0.802	12.124
I feel fatigued when I get up in the morning and have to face another day on the job.	0.751	11.033
I feel burned out from my work.	0.777	11.572
I feel frustrated by my job.	0.875	13.903
I feel I am working too hard on my job.	0.768	11.379
I feel like I am at the end of my rope.	0.884	14.148
Workload (W_LOAD)			0.83	0.74
How often does your job leave you with little time to get things done?	0.789	10.204
How often do you have to do more work than you can do well?	0.890	11.562

Model fit statistics: χ2 = 313.726; df = 174; χ2/df = 1.80; GFI = 0.85; NFI = 0.88; NNFI = 0.93; CFI = 0.94; RMSEA = 0.07; RMR = 0.05 Note: All loadings are significant at the level < 0.001

The average variance extracted for each factor was between 0.52 and 0.99. Composite reliabilities of constructs ranged from 0.83 to 0.92, which was higher than the standard of 0.6 (Chiang & Hsieh, 2012). Cronbach’s alpha was also employed to ascertain the internal consistency of items (Table 1); it has become commonplace to report the reliabilities of measures in a study, even when SEMs are used (Bagozzi & Yi, 2012). Results of the alpha coefficient for each study construct were greater than the cutoff value of 0.70 (Nunnally, 1978), indicating an acceptable level of internal reliability of all the study constructs. Means, standard deviations and correlations are also provided in Table 2.

Table 2. Correlations, means, standard deviations, and reliabilities of the study constructs

	1	2	3	4	5	6
1. S_SUPPORT	–
2. AFCOM	0.206*	–
3. QUALITY OF SERV_PERF	0.155	0.500**	–
4. IN_LEAVE	−0.242**	−0.679**	−0.209*	–
5. EM_EXH	−0.374**	−0.519**	−0.222*	0.780**	–
6. W_LOAD	−0.397**	−0.112	−0.032	0.313**	0.541**	–
Mean	4.00	3.85	4.03	2.17	2.15	3.08
Standard deviation	1.05	0.98	0.73	1.13	1.05	0.91
Alpha	0.89	0.89	0.80	0.92	0.92	0.83

*p < 0.05;**p < 0.01

S_SUPPORT (Supervisor Support); W_LOAD (Workload); EM_EXH (Emotional Exhaustion); AFCOM (Affective Organizational Commitment); QUALITY OF SERV_PERF (Quality of Service Performance); IN_LEAVE (Intention to Leave)

To assess the common method bias, Harman’s one-factor test (Podsakoff et al., 2003; Podsakoff & Organ, 1986) was employed. Principal component analysis was used in the test with all variables in the model. Both with the Varimax rotation and without it, the share of variance explained by one factor amounted to 32% and showed no evidence of the common method bias. Additionally, CMB assessment was conducted using a marker-variable technique (Lindell & Whitney, 2001). The tests confirmed results from Harman’s one-factor test. Therefore, it was concluded that CMB was not problematic.

Structural model and tests of hypotheses

The hypothesized relationships were assessed with SEM in R version 3.0.2., package lavaan v. 05.-15 (Rosseel, 2012). According to the empirical results, the model fits the data relatively well (χ2 8.784; df = 6; χ2/df = 1.46; GFI = 0.98; NFI = 0.97; NNFI = 0.98; CFI = 0.97; RMSEA = 0.05; RMR = 0.04). Six of the eight hypothesized structural paths were statistically significant in the predicted directions (Figure 2).

Figure 2. Path results of the research model

Model fit statistics: χ2 = 8.784; df = 6; χ2/df = 1.46; GFI = 0.98; NFI = 0.97; NNFI = 0.98; CFI = 0.97; RMSEA = 0.05; RMR = 0.04.Note: Numbers in parentheses are t-statistics; * p < .05; **p < .01; *** p < .001.S_SUPPORT (Supervisor Support); W_LOAD (Workload); EM_EXH (Emotional Exhaustion); AFCOM (Affective Organizational Commitment); QUALITY OF SERV_PERF (Quality of Service Performance); IN_LEAVE (Intention to Leave)

Hypothesis 1 stated that S_SUPPORT negatively influences hotel employees’ emotional exhaustion. The path estimate was consistent with this prediction, thus supporting Hypothesis 1. Results of the path analysis showed that workload was significantly and positively related with EM_EXH. Thus, Hypothesis 2 was also supported. S_SUPPORT and W_LOAD jointly explained 27% of the variance in emotional exhaustion. The results of path analysis demonstrated that EM_EXH does not exert a significant and negative impact on AFCOM and QUALITY OF SERV_PERF. Thus, Hypotheses 3 and 4 were not accepted. Hypothesis 5 predicted that EM_EXH significantly and positively influences IN_LEAVE. The path analysis showed that EM_EXH had a significant positive impact on employees’ IN_LEAVE. Therefore, Hypothesis 5 was supported. Path analysis also showed that AFCOM was negatively and significantly related to IN_LEAVE, confirming Hypothesis 6. EM_EXH explained 23% of the variance in AFCOM, whereas AFCOM and EM_EXH jointly explained 94% of the variance in IN_LEAVE. The result concerning the path from AFCOM to QUALITY OF SERV_PERF was significant and positive. Thus, Hypothesis 7 was supported. It was also observed that IN_LEAVE negatively affected QUALITY OF SERV_PERF providing support for Hypothesis 8. EM_EXH, AFCOM, and IN_LEAVE jointly explained 27% of variance in QUALITY OF SERV_PERF.

Discussion

The study developed and tested a research model that examined the effects of job resources (supervisor support) and job demands (workload) on emotional exhaustion and explored the relationships between emotional exhaustion and its critical employee and job-related outcomes.

The study findings confirmed that workload and supervisor support significantly influence emotional exhaustion. Specifically, the results reveal that employees who are not able to cope with difficulties arising from excessive workload are especially prone to experience a heightened level of emotional exhaustion. Similar findings are in a study by Kim et al. (2007), where quantitative workload was significantly and positively related to exhaustion among hotel employees of chain hotels located in Washington State (USA).

Results of this study also confirmed that supervisor support might play an important role in decreasing employees’ experience of emotional exhaustion. Similarly, Karatepe’s (2010) study also revealed that availability of adequate support from supervisors could be viable in protecting hotel employees from feeling exhausted.

It is worth noting that the impact of workload (job demand) on emotional exhaustion was stronger than the effect of supervisory support (job resource), which may be explained by the Conservation of Resources (COR) Theory, specifically principle 1: “The Primacy of Resource Loss,” which states that resource loss is disproportionally more salient than the resource, and with equal amounts of loss and gain, loss will have a greater impact (Hobfoll, 2001). Therefore, as Lee and Ashforth (1996) showed in their meta-analysis, and as proved in this study, hotel employees are more sensitive to the demands that are placed on them, e.g., excessive workload, than to the received resources, such as supervisor support.

This study also evidenced that emotional exhaustion significantly enhanced hotel employees’ intention to leave. Similar positive relationships between emotional exhaustion and turnover intention were found by Karatepe and Aleshinloye (2009), Karatepe and Uludag (2007), and Yavas et al. (2008).

This study failed to find any significant relationship between emotional exhaustion and the quality of service performance. This is in line with previous studies by Yavas et al. (2008), Karatepe and Aleshinloye (2009), and Karatepe and Uludag (2008), where a relationship between hotel employees’ emotional exhaustion and job performance was not proved. Similarly, the relationship between emotional exhaustion and affective commitment appeared to be insignificant. One of the plausible explanations for this unexpected finding may result from the fact that employee’s affective commitment may be influenced by a variety of different factors, such as demographic variables (Lee & Chen, 2013), job characteristics (Ozturk et al., 2014), social (e.g., coworkers support, Limpanitgul et al., 2014) and motivational factors (Alnıaçık et al., 2012; Karatepe & Uludag, 2007) and different personality dimensions (e.g., Rathi & Lee, 2016). Therefore detrimental impact of emotional exhaustion on hotel employee’s affective commitment may be attenuated by other impactful variables, not covered by this research model. Thus for better understanding, this finding needs further empirical attention.

The results also revealed that both affective organizational commitment and turnover intention were significant predictors of the quality of service performance. Specifically, the positive relationship between affective commitment and the quality of service performance has been confirmed. Babakus et al. (2003) also provided empirical support for a positive relationship between affective organizational commitment and service recovery performance. By contrast, as revealed in this study and confirmed by previous ones (e.g., Al-Badarneh et al., 2019), turnover intention was significantly and negatively related to the quality of service performance.

In this study, an affective organizational commitment was also significantly and negatively related to turnover intention, which is consistent with findings of previous studies conducted among hotel employees (e.g., Zopiatis et al., 2014; Yang, 2010), but in contrast with Karatepe and Kilic’s study (Karatepe & Kilic, 2007), where this relationship was not confirmed.

Summing up, the study findings confirmed that workload and supervisor support are both significant antecedents of EM_EXH, which, in turn, exerts a significant impact on hotel employees’ leaving intention that reduces the quality of service performance.

Implications

Although stress is perceived as a “one of the most serious problems” affecting hotel employees (Hwang et al., 2014, p. 71), only few studies have tried to investigate occupational stress in that setting (Chiang et al., 2010; Hwang et al., 2014; McNamara et al., 2011). Therefore, this study may contribute to a better understanding of the work-stress process in the hotel context bridging several gaps in the current knowledge.

Firstly, although previous studies shed important light on the understanding of the effects of job demands and resources on EM_EXH and/or its selected outcomes (e.g., Karatepe & Uludag, 2007; Kim et al., 2007; Yavas et al., 2008), none of them tested the joint effects of workload and supervisor support on hotel employees’ EM_EXH and its employee and job-related outcomes, such as affective commitment, intention to leave and, primarily, the quality of service performance, linking this specific job characteristics with critical employees’ outcomes, both attitudinal and behavioral via emotional exhaustion. By assessing the abovementioned relationships, this paper fills the void in the hospitality management literature and extends the study results to a new social and cultural context of the hospitality industry in East-Central Europe, which significantly enriches the research insights into the burnout phenomenon.

Secondly, since “a guest experience is a service (…)” (Ford et al., 2012, p. 23) and frontline hotel employees must continually manifest high-quality job performance (Karatepe & Douri, 2012), the quality of service performance was measured as a critical variable for frontline service jobs in the hospitality industry. Specifically, the 5-item scale was developed and validated by the Author. Therefore, this study may serve as a frame of reference for future research which may use and test this measurement tool in different hospitality settings and socio-cultural backgrounds and may constitute a basis for comparing the obtained results and their discussion given different contexts of further studies.

Thirdly, empirical evidence regarding the relationship of burnout with employees’ job performance is scarce in the existing tourism and hospitality literature (Karatepe & Uludag, 2008), and the results of individual studies remain mixed (see e.g., Choi et al., 2014; Karatepe & Uludag, 2008; Yavas et al., 2008). Therefore, by investigating the relationship between emotional exhaustion and the quality of service performance, this study enriches limited findings in this research stream and adds to the modest body of hospitality research.

This paper also offers useful practical implications. The results of this study showed that if hotel managers want to maintain committed workforce who delivers high service quality, they should find appropriate means to alleviate emotional exhaustion. Specifically, because workload proved to be an influential job stressor leading to hotel employees’ emotional exhaustion, any methods to reduce excessive workload are strongly recommended under managerial attention. Therefore, firstly, as hospitality jobs are characterized by long, unsocial working hours in variable shifts, including nights and weekends (Chan et al., 2019; Wong & Ko, 2009), employees should have more flexibility in setting their own working schedule, and their preferences in this regard should be then incorporated into the schedule (Schiffinger & Braun, 2020). Secondly, given seasonality in hospitality jobs and high peaks in demand, hotel operational departments should be optimally staffed to avoid excessive workload and time pressure (Schiffinger & Braun, 2020); otherwise, the pressure to perform well in understaffed situations makes the work particularly stressful and exasperating for employees.

Results of this study also confirmed that supervisors should be aware of their supporting role in decreasing employees’ exhaustion and managing their emotional labor. However, to do it effectively, they need to be sensitive to their employees’ emotional needs (Haldorai et al., 2019), empathize with them (Genc & Genc, 2018) and play a role of their mentors transferring knowledge on how to do tasks efficiently to alleviate the work pressure (Schiffinger & Braun, 2020). Supervisory mentoring may also play a critical role in increasing employees’ affective commitment and their retention (Lapointe & Vandenberghe, 2017), which is particularly important given the fact that both affective commitment and turnover intention may influence the quality of service performance, as proved in this study. Additionally, to retain efficient hotel employees and strengthen their attachment to the hotel organization, it is also important to motivate them effectively to plan for a long-term career with the organization. Therefore, each and every employee should be made to understand how their work is aligned with organizational goals and how their unique contribution will help in organizational growth (Haldorai et al., 2019). Furthermore, more activities to enhance team cohesion and create a good working environment should be offered, which may also ease the employees’ emotions and help them to relieve the emotional pressure (Wen et al., 2019).

Having in mind the findings of this study, hotel management should also regularly check their frontline employees’ level of job stress to protect them from experiencing emotional exhaustion and its negative consequences. Thus, if the job stress reaches a certain threshold, managers should take adequate actions, such as, for example, allowing employees to take time off to recharge (Choi et al., 2019; Wen et al., 2020). In this context, it is important to provide some work–life balance practices and encourage employees to cultivate their hobbies to relieve stress (Cheng et al., 2018). Likewise, it is vital to train employees to manage their emotions and, instead of personally absorbing the work stress, repurpose this energy more constructively (Boukis et al., 2020) giving support for service performance. Thus, emotional management courses on how to regulate and deal with emotional exhaustion should be offered (see Cheng et al., 2018; Essawy, 2016; Wen et al., 2020).

Finally, when the job demands are incongruent with employees’ individual abilities, such employees cannot cope effectively with job difficulties and are likely to deliver poor services and display voluntary turnover (Karatepe et al., 2014). Therefore, hotel managers should recruit and select the most suitable individuals for frontline service positions who will be likely to succeed in hotel organizations, e.g., extroverts who may deal better with emotional demands of their hotel work (Haldorai et al., 2019). Therefore, the recruitment process should involve some psychometric tests to ensure that the candidates have adequate personality profiles. Managers may also use holistic data regarding a candidate’s employment history to determine if he or she would fit into the organization in terms of skills and personality (Haldorai et al., 2019).

Limitations and conclusions

Since all information in this research was generally collected at one time and based on the respondents’ self-reports, self-reported bias could be a concern in this study. However, although CMB appeared not to be problematic in this research, future studies should collect data from multiple sources (e.g., the customers’ real evaluations of the quality of service performance) and at different periods to avoid such potential bias.

This research has involved only selected hotels from Poland, which limits the possibility to generalize the results. Future research should be directed to investigate the relationships with both larger samples and different hospitality settings for the sake of generalization. Moreover, the presence of a representative assigned by the management of hotels during the data collection for this study could put some pressure on employees’ participation. Therefore, it is recommended that future studies collect data directly from contact hotel employees without the presence of persons appointed by the management (Karatepe, 2013).

Furthermore, due to the cross-sectional nature of the data, it is also impossible to conclude causal relationships between the study variables (Tromp & Blomme, 2012). Thus, a longitudinal approach should be applied in future studies.

Testing the external validity of the research model in other service sectors could also be interesting, and a cross-cultural study is highly recommended to reveal any socio-cultural differences. The future research may also take many other directions. For example, taking into consideration the nature of hospitality jobs, it would be appropriate to focus not only on quantitative workload but also on other, physical and mental aspects of workload as influencing factors of emotional exhaustion. Therefore, qualitative workload (emotional and social) deserves future empirical attention. Moreover, it could be interesting to test the differences between employees of different hotel departments to identify if any of them is subject to experiencing emotional exhaustion more frequently. This may help hotel managers to determine and understand which factors may affect a particular group of employees the most and why. Hence, they will have a better foundation for understanding what works in the hotel well and what needs immediate improvements. Additionally, individual differences should also be considered in future research, as different employees may potentially have different coping mechanisms and resources. Particularly, a relationship between personality traits and EM_EXH has not received adequate attention yet (Kim et al., 2007; O’Neill & Xiao, 2010). This information could be of great importance both for employers to recruit the “right” employees with proper personality profiles and for individuals to find an adequate workplace to cope effectively with work stress (Kim et al., 2007).

Additional information

Funding

This work was partly supported by the Gdynia Maritime University project [No.WPIT/2020/PZ/07].