From Customer-Related Social Stressors to Emotional Exhaustion: An Application of the Demands

Abstract

This study examines emotional intelligence (EI) as an additional moderator in the interactive effect of customer-related social stressors (CSS) (demands) and job autonomy (control) on emotional exhaustion by constructing a three-way interaction (EI × CSS × Autonomy) for the regression model of emotional exhaustion. Employees from hotels and restaurants in one metropolitan city in the United States participated in this study. The interactive effect of CSS and autonomy on exhaustion is shown among employees low in EI rather than high. Low EI employees experience greater exhaustion when they perceive low autonomy (stress exacerbating effect). When CSS is high, job autonomy is low, and EI is low, employees experience the highest level of emotional exhaustion. The findings support the proposition that individual characteristics, such as EI, add significant explanatory power to the job demands–control model (JDC) to understand occupational stress. Managerial and theoretical implications are provided based on the results.

Keywords

Hospitality companies survive and succeed by continuously providing superior services. However, high employee turnover challenges this goal. In 2016, the restaurant and accommodation sector had a turnover rate of more than 70% (Bureau of Labor Statistics, 2017; National Restaurant Association, 2017). While hoteliers spend as much as 33% of revenue on labor costs (Deloitte, 2010), staff turnover could cost $5,964 on average per employee because of productivity loss, recruiting, and orientation and training (Tracey & Hinkin, 2006). Thus, hospitality researchers have been keen to identify the underlying causes of high turnover and related poor service performance among service providers.

One common antecedent that has been explored to explain both high turnover and low service performance is emotional exhaustion (EE; Maslach & Jackson, 1981). Among many forms of stressors that could result in emotionally depleted employees (e.g., organizational climate, work overload, and work–family conflict) (Karatepe, 2013; O’Neill & Xiao, 2010), social stressors from negative encounters with customers have been noted as a key stressor reflecting an inevitable phenomenon in the customer service setting (Hu, Hu, & King, 2017; Huang & Miao, 2016). Customer-related social stressors (CSS) are known to negatively affect service providers’ short- and long-term well-being (Dudenhöffer & Dormann, 2013) and service recovery performance (Kim, Paek, Choi, & Lee, 2012).

Hospitality researchers have investigated how CSS predicts EE. Several dimensions of CSS can directly increase EE (Choi, Kim, Lee, & Lee, 2012) and the chronic impact of CSS on EE can be mediated through acute negative emotion, such as anger (Weber, Bradley, & Sparks, 2017). Many approaches to lessening CSS-induced EE were identified from the individual level, incorporating self-efficacy/positive affectivity (Karatepe, 2015), intrinsic motivation (Karatepe & Aleshinloye, 2009), and customer orientation (Yoo, Kim, & Lee, 2015), to the organizational level, including perceived organizational support (Karatepe, 2011) and supervisory support/training/empowerment/rewards (Babakus, Yavas, & Karatepe, 2008).

However, these studies have focused on direct effects on EE without considering the interactive effect. More scholarly attention needs to be paid to integrate both individual and organizational approaches, and their interactions to alleviate EE, because individuals, as part of the organization, are heavily affected by company action. In hospitality literature, Chiang, Birtch, and Kwan (2010) have shown the possibility of interactive effects using two job-related moderators (i.e., three-way interactions) on job stress, but no study has yet been conducted to examine potential interactions between individual and job/organizational variables to attenuate job stress.

To fill the research gap explained above, this study explores the mechanism of how to attenuate the influence of CSS on EE under the framework of the job demands–control (JD-C) model (Karasek, 1979; Karasek & Theorell, 1990). More specifically, this study aims to shed light on how to effectively manage CSS-induced EE by enabling employees to exercise autonomy at work (organizational level) and by activating employees’ ability to handle emotions on site (individual level) in an appropriate manner. Employees’ emotional intelligence (EI) is a significant predictor of service performance (Prentice & King, 2011). Job autonomy plays a considerable role in employees’ organizational commitment, particularly for quality-focused organizations, such as service firms (Park & Searcy, 2012); furthermore, contact employees’ service innovative behavior is commensurate to job autonomy (Dhar, 2016). The JD-C model will serve to articulate the interactions between autonomy and employees’ EI and identify the conditions under which a phenomenon (EE) is more or less likely to occur. The findings of this study will contribute to the hospitality stress-related literature and assist practitioners with a way to combat contact employees’ exhaustion to ultimately enhance service performance.

Literature Review and Hypotheses Development

Job Demands–Control Model

The JD-C model was first introduced by Karasek (1979). Job demands pertain to the aspect of the job requiring employees to make sustained physical and/or mental efforts, and job control refers to decision latitude, often conceived as discretion, autonomy, or self-determination. Karasek (1979) foresaw the interaction between job demands and job control in predicting job-related outcomes and conceptualized the following four scenarios: (1) high demands–high control (active job), (2) low demands–low control (passive job), (3) high demands–low control (high-strain job), and (4) low demands–high control (low-strain job). Among these four, two scenarios have received considerable, scholarly attention—active jobs (desirable) and high-strain jobs (undesirable). Active jobs are in harmony with a stress-buffering hypothesis: High job control has a stress-alleviating effect on job strain when job demands are high. On the other hand, the most adverse situation occurs when job demands are high and job control is low (Karasek & Theorell, 1990). These high-strain jobs suggest a stress-exacerbating hypothesis.

As studies on job stress advanced from health care professionals or teachers to workers in various professions, scholars urged that the JD-C model should incorporate the measures of diversified job demands to represent different occupational groups (de Jonge, Dollard, Dormann, Le Blanc, & Houtman, 2000). In this study, we selected customer-derived social stressors as critical job demands, given that the participants of this study are hotel and restaurant employees facing an excessive number of customer contacts around the clock (Kim, 2008). Hospitality service providers’ exhaustion triggered by frequent guest contacts has been well documented (Min, Kim, & Lee, 2015).

In addition to a limited choice of occupations by JD-C scholars, the problematic area in the JD-C model is the inconsistent interaction between job control and job demands, which has been emphasized in several main reviews of the JD-C model (Doef & Maes, 1999; Hausser, Mojzisch, Niesel, & Schulz-Hardt, 2010). Explanations for the failure of the moderating hypothesis have been offered, including the mismatch of measures for job demand and job control and the absence of a third variable (second moderator). For the latter issue, proponents of the JD-C theory suggest job characteristics (Chiang et al., 2010) or individual characteristics (Fernet, Guay, & Senécal, 2004) as a feasible third variable.

Customer-Related Social Stressors and Emotional Exhaustion

CSS is defined as job stressors that service providers experience during social interactions with customers at work (Dormann & Zapf, 2004). During day-to-day social interactions with customers, hospitality service providers do “people-work” (Kim, 2008). “People-work” requires emotional work, also known as emotional labor. Emotional labor signifies an employee’s emotional expression and regulation per display rules, prescribed by the organization, to achieve organizational goals (Ashforth, 1993; Grandey, 2000; Hochschild, 1983). There are two primary emotional regulation strategies: (1) surface acting (faking facial signs of emotion) and (2) deep acting (reappraising events and modifying inner feelings). While display rules could be resources as a guideline for employees, emotional labor demanded by display rules is viewed as a “double-edged sword” that induces strain, emotional dissonance, and burnout (Kim, 2008).

Hospitality researchers have noted hospitality service providers’ suffering and exhaustion from emotional work, originating from CSS (Choi et al., 2012; Karatepe, Yorganci, & Haktanir, 2009). CSS has four subfacets: (1) disproportionate customer expectations, (2) customer verbal aggression, (3) disliked customers, and (4) ambiguous customer expectations (Dormann & Zapf, 2004). Karatepe et al. (2009) and Kim, Shin, and Swanger (2009) reported a significant, positive relationship between customers’ verbal aggression and EE using a sample of lodging employees in Northern Cyprus and a sample of Subway restaurant employees in the United States. Choi et al. (2012) found that three components of CSS (ambiguous customer expectations, disliked customers, and customer verbal aggression) are positively related to exhaustion among frontline personnel in travel agencies, tourist hotels, and tourist restaurants in South Korea. Given the previous findings, it seems reasonable to assume CSS as the antecedent of EE in the hospitality work context:

Hypothesis 1₀: There will be no effect of CSS on EE.

Hypothesis 1_a: There will be a main effect of CSS on EE. CSS will be positively associated with EE.

Job Autonomy and EE and the Moderating Role of Autonomy in the Relationship Between CSS and Exhaustion

Hackman and Oldham (1975) define job autonomy as the degree to which the job provides employees with substantial independence and decision latitude in their work pace and phases. The occupational stress literature demonstrates that job autonomy is crucial for employee well-being because of the opportunities that employees may have to cope with stressful situations (Jenkins, 1991). For example, in the service job context, managers may not always be available, and customers are impatient with organizational hierarchy or being referred to other departments or units (Hart, Heskett, & Sasser, 1990). From the service providers’ viewpoint, satisfying customers sooner rather than later can lessen their own stress. In other words, job autonomy enables employees to solve problems in a timely fashion. The hospitality literature has yielded support for the positive link between autonomy and employees’ better response to work role (Ross, 1997) and the negative link between autonomy and employees’ perception of EE (Kim, 2008). Therefore, the following hypothesis is proposed for the relationship between job autonomy and EE.

Hypothesis 2₀: There will be no effect of job autonomy on emotional exhaustion.

Hypothesis 2_a: There will be a main effect of job autonomy on emotional exhaustion. Job autonomy will be negatively associated with emotional exhaustion.

At the same time, abiding by the core principle of the JD-C model, this study postulates that employees’ use of job autonomy (control) may potentially mitigate the magnitude of the relationship between CSS (demands) and EE (job strain) because ample decision latitude may help improve the quality of customer service. Kim and Stoner (2008) surveyed social workers in California and found that social workers with more job autonomy report lower levels of burnout. Bakker, Demerouti, and Euwema (2005) provided evidence that autonomy can effectively diminish the unfavorable influences of work overload and emotional demands on burnout through interactive effects.

As hospitality jobs involve role stress, work overload, and emotional demands, dealing with customers’ complaints, impoliteness, and intimidation, hospitality employees can benefit from the specific job control—job autonomy. In other words, when service providers perceive a high level of job autonomy, EE originating from CSS may be diminished. This is in line with the stress-buffering hypothesis. The opposite relationship (stress-exacerbating hypothesis) can be predicted with the presence of low job control. When hospitality employees perceive a low level of autonomy, EE resulting from CSS may be increased. Given the empirical support for the JD-C-moderating hypothesis (Bourbonnais, Comeau, & Vézina, 1999; Van Yperen & Hagedoorn, 2003) and the belief that job autonomy and CSS are well-matched control and demand variables in the high customer-contact work setting such as hotels and restaurants, this study posits the following:

Hypothesis 3₀: There will be no two-way interactive effect of CSS and job autonomy on emotional exhaustion.

Hypothesis 3_a: There will be a two-way interactive effect of CSS and job autonomy on emotional exhaustion. High job autonomy will be stress buffering, and low job autonomy will be stress exacerbating in the relationship between CSS and emotional exhaustion. The highest level of exhaustion will occur when CSS is high and job autonomy is low.

EI and Its Moderating Role in the Interactive Effect of CSS and Autonomy on EE

The trait of EI is defined as a constellation of behavioral dispositions and self-perceptions concerning one’s ability to recognize, process, and utilize emotion-laden information (Petrides & Furnham, 2003). In a meta-analysis, Schutte, Malouff, Thorsteinsson, Bhullar, and Rooke (2007) argue that the individual characteristic of EI, which accounts for variability in emotional capabilities of both intrapersonal (e.g., regulating one’s own emotion) and interpersonal (e.g., using others’ emotion) dimensions, may favorably affect one’s physical and psychosomatic well-being. Other scholars reveal health-conscious behaviors by people with high EI. Those high in EI tend to seek heath professionals’ assistance and conform to their advice (Ciarrochi & Deane, 2001). People with high EI are better equipped to resist peer pressure to drink alcohol and smoke (Austin, Saklofske, & Egan, 2005; Trinidad & Johnson, 2002). Salovey, Bedell, Detweiler, and Mayer (2000) suggest that EI functions as a coping mechanism that may assist individuals in regulating themselves toward desired ends.

As mentioned earlier, the JD-C model has been controversial because of mixed findings on the interaction between job demands and job control (de Jonge, van Vegchel, Shimazu, Schaufeli, & Dormann, 2010; Doef & Maes, 1999; Hausser et al., 2010) and individual differences may offer the most feasible answer to the failure (Györkös, Becker, Massoudi, de Bruin, & Rossier, 2015; Parker & Sprigg, 1999). For example, Meier, Semmer, Elfering, and Jacobshagen’s (2008) study shows the significance of individual characteristics in the JD-C model. They found that for people with high levels of self-efficacy or internal locus of control, job control diminished the effect of job stressors on affective strain and musculoskeletal pain. Fernet et al. (2004) reported the three-way interaction under the JD-C model after including an individual characteristic—self-determination. Their results show that job control buffers the unhealthy effect of job demands on EE only for employees with high levels of work self-determination. In other words, certain individuals are susceptible to the use of job control and, thus, enhance the traditional JD-C model with more refined, theoretical accounts.

EI may influence the interaction effect between CSS and job autonomy on exhaustion in a number of ways. First, a high level of EI predisposes individuals to recognize emotion in a customer’s nonverbal expression, which allows service employees to process CSS and capture the accurate emotions of customers (Buck, 1984). Second, with high EI, service employees may take less effort to develop knowledge about emotions to facilitate certain types of problem solving (Erez & Isen, 2002). It is thus expected that high EI people would use a proper emotion to react to customer emotions. Third, when employees are faced with a difficult customer, the increased ability to manage one’s own emotion should promote a calm situation and, at the same time, they think and act in ways to use job autonomy in a timely manner. As it can be imagined, reasonable actions based on the service recovery protocol can hardly be executed when emotions are out of control. Finally, the growth of emotional self-management enhances the protective role of EI in stress in the workplace (Mikolajczak, Menil, & Luminet, 2007).

Like people high in self-efficacy and internal locus of control, we expect high EI individuals to have a potential to actively use job control such as job autonomy because of the role of EI as a coping mechanism (Salovey et al., 2000). Several coping styles are possible to manage the potential stress and negative consequences of emotional labor, driven by CSS, including emotion-focused coping, avoidance coping, and problem-solving coping (Jung & Yoon, 2016; Kim & Agrusa, 2011). In general, compared with emotion- or avoidance-oriented coping, active and problem-focused coping produces desirable outcomes in the workplace (Endler & Parker, 1994).

Kim and Agrusa (2011) found that EI is significantly related to active, task-oriented coping in the hospitality workplace. High-EI individuals have the ability to relate to others thanks to their sensitivity to emotion-laden information (Mikolajczak, Nelis, Hansenne, & Quoidbach, 2008). This sensitivity to others is likely to motivate high-EI employees to be task-oriented, problem-solving copers (Kim & Agrusa, 2011) by energetically using their job autonomy to reduce others’ stress, which ultimately lowers their own stress. High job autonomy is, therefore, likely to serve as a buffer against CSS for high-EI hospitality employees. What happens to high-EI individuals when they perceive low autonomy? Despite high-EI individuals’ inclination to use job autonomy, we do not necessarily expect low autonomy to worsen high-EI employees’ exhaustion because of the empirical findings on their resilience to burnout (Lee & Ok, 2012); in other words, the stress-exacerbating effect of low autonomy may not take place under a high-EI condition.

If the nature of low EI is opposite of that of high EI, job autonomy may not be actively sought by low-EI workers (less frequent use of autonomy). This implies that the stress-buffering effect of autonomy may not be salient under a low-EI condition. However, everyone strives to minimize stress at work. When low-EI employees do feel that autonomy is needed while confronting demanding or challenging situations, the lack of (or low) autonomy may create far worse consequences for low-EI employees than for high-EI workers. Low-EI employees are found to be more susceptible to job stress than high-EI employees (Huang, Chan, Lam, & Nan, 2010). The conservation of resources (COR) theory asserts that for individuals with a low personal resource (low-EI person in this study), the lack of other resources (job autonomy in this study) could worsen the stress level (Hobfoll, 1989; Hobfoll & Shirom, 1993). In summary, this study predicts that high EI is well suited to the stress-buffering hypothesis while low EI fits well into the stress-exacerbating hypothesis in the JD-C framework. These rationales lead to the following three-way interaction hypotheses (Figure 1):

Hypothesis 4₀: There will be no three-way interactive effect of CSS, job autonomy, and EI on emotional exhaustion.

Hypothesis 4_a: There will be a three-way interactive effect of CSS, job autonomy, and EI on emotional exhaustion.

Hypothesis 4(1)_a: For employees with high EI (compared with low EI), high job autonomy will be stress buffering in the relationship between CSS and emotional exhaustion.

Hypothesis 4(2)_a: For employees with low EI (compared with high EI), low job autonomy will be stress exacerbating in the relationship between CSS and emotional exhaustion. The highest level of exhaustion will occur when CSS is high, job autonomy is low, and EI is low.

Figure 1

Proposed Model

Method

Procedure and Sample

Data were collected from employees working in four mid-to-upscale hotels and three full-service restaurants located in one metropolitan city in the United States. After gaining approval from the top manager of each establishment (i.e., general manager or owner), employees and supervisors/managers in the front of the house were invited to this research project: hotels (personnel from front desk, concierge, sales, and banquets) and restaurants (servers and bartenders). Top managers, who approved data collection, are alumni or recruiters who are familiar with the lead author’s hospitality program. Thus, it is a convenient sampling method. Although the responsibility of each department may differ, participating employees and supervisors in these departments make extensive face-to-face interactions with customers, thereby likely experiencing CSS.

On the cover letter of the survey, we emphasized the importance of this study and guaranteed anonymity. Participants were allowed to take part in this study voluntarily during their break or after a daily meeting of the unit. Participants returned their surveys to the research assistant in an envelope. Two research assistants visited their assigned properties twice. Data collection took about 2 weeks. Of 480 surveys, 191 surveys were collected (response rate: 40%). After the removal of poorly answered questionnaires, 173 surveys were used for data analyses. The majority of respondents (60.1%) were females (n = 104) and 39.9% were males (n = 69). The respondents’ ages ranged from 18 to 60 years. One-third of respondents (30.6%, n = 53) were supervisors or managers, and the remaining (69.4%, n = 120) were nonsupervisors. The average work experience was 6.5 years for hotels and 2.9 years for restaurants.

Measures

The CSS scale, developed by Dormann and Zapf (2004), was used as an indicator for job demands in this study: ambiguous customer expectations (4 items), disproportionate customer expectations (8 items), verbal aggression (5 items), and disliked customers (4 items). Job autonomy was assessed by the subscale of Hackman and Oldham’s (1975) job diagnostic survey (3 items). EI was assessed by Wong and Law’s (2002) EI scale: self-emotion appraisal (4 items), others’ emotion appraisal (4 items), uses of emotion (4 items), and regulation of emotion (4 items). EE was measured with the subscale of Maslach Burnout Inventory (9 items; Maslach & Jackson, 1981). All CSS items were rated on a five-point Likert-type scale (1 = never, 5 = very often), and the rest of the constructs including autonomy, EI, and EE were rated on a five-point Likert-type scale (1 = strongly agree, 5 = strongly disagree). For the purpose of descriptive analyses, age, gender, and departments were inquired at the end of the survey. Age was answered in years. Gender (male or female) and work areas (department names) were listed and asked as categorical variables. Reliability and validity of all scales are discussed in detail in the later, measurement model section.

Analysis and Results

Common Method Bias

Because data were collected from a single source (i.e., common rater), they may be vulnerable to common method bias. First, Harman’s one-factor test was performed to test common method bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The results of the analyses showed that the largest, extracted component accounts for only 24.9% of total variance, indicating a minimal possibility of common method bias. Second, despite its popularity, Harman’s one-factor test has been perceived as a weak, conceptually flawed statistical procedure by several researchers (Chang, van Witteloostuijn, & Eden, 2010; MacKenzie & Podsakoff, 2012). Given this criticism, the unmeasured latent method factor approach (Min, Park, & Kim, 2016; Podsakoff et al., 2003), was conducted. All paths (factor loading values) were compared between two measurement models: (1) a model with a common method factor and (2) a model without a common method factor. Because factor loading differences between the two measurement models were trivial, ranging from −.096 to .009, the results confirmed that the possibility of common method bias is indeed slim.

Measurement Model

Confirmatory factor analysis was conducted with Mplus 7 (Muthén & Muthén, 2012) to examine the psychometric properties of the measurement model. Prior to testing the measurement model, we averaged items in each subdimension of CSS and EI and treated subdimensions as indicators of their respective higher order latent variables. The fit of the four-factor model appears to be satisfactory except for the chi-square test: χ² (161) = 269.010, p ≤ .001; comparative fit index (CFI) = .923, Tucker–Lewis index (TLI) = .909, standardized root mean square residual (SRMR) = .06, root mean square error of approximation (RMSEA) = .06. Note that the chi-square test is sensitive to the sample size, frequently producing a significant result (Kline, 1998). Thus, researchers consider the results of other fit indices.

As for reliabilities, coefficient H (Hancock & Mueller, 2001), a measure of composite reliability displayed the evidence of construct reliability: .86 for CSS, .90 for EE, .75 for EI, and .63 for job autonomy (Table 1). Although the reliability value for autonomy is lower than other constructs, the modest level of .63 is acceptable (Nunnally & Bernstein, 1994). All measurement items of the four constructs loaded significantly on their respective constructs (p < .001) with the standardized factor loading values ranging from .46 to .85 (Table 1). This demonstrates a reasonable evidence of convergent validity of the measurement model (Anderson & Gerbing, 1988).

Table 1

Confirmatory Factor Analysis for Measurement Items

Items	Standardized Factor Loadings	Coefficient H
Customer-related social stressors		.86
1. Ambiguous customer expectations	.85
2. Disproportionate customer expectations	.76
3. Verbal aggression	.73
4. Disliked customers	.74
Job autonomy		.63
1. My job lets me be left on my own to do my own work.	.66
2. I am able to act independently of my supervisor in performing my job function.	.60
3. My job provides me with the opportunity of self-directed flexibility of work hours.	.50
Emotional intelligence		.75
1. Self-emotion appraisal	.76
2. Others’ emotion appraisal	.54
3. Use of emotion	.46
4. Regulation of emotion	.69
Emotional exhaustion		.90
1. I feel frustrated by my job.	.82
2. I feel emotionally drained from my work.	.63
3. I feel used up at the end of the workday.	.61
4. I feel tired when I get up in the morning and have to face another day on the job.	.62
5. Working with people all day is really stressful to me.	.64
6. I feel burned out from my work.	.82
7. I feel I am working too hard on my job.	.77
8. Working with people directly puts too much stress on me.	.67
9. I feel like I am at the end of my rope.	.50

Note: Model fit indices: χ² (161) = 269.010, p ≤ .001; CFI (comparative fit index) = .923, TLI (Tucker–Lewis index) = .909, SRMR (standardized root mean square residual) = .06, RMSEA (root mean square error of approximation) = .06.

Discriminant validity can be assessed by examining correlations among latent variables. Correlations among latent variables ranged from −.21 to .45 (Table 2), implying that constructs are not highly correlated with one another and therefore each construct is distinctive. To further examine discriminant validity, average variance extracted (AVE) values (Table 2) were compared with squared correlations between study constructs (Anderson & Gerbing, 1988; Fornell & Larcker, 1981). AVE values for the four study constructs (.36 ≤ AVEs ≤ .60) were greater than squared interconstruct correlations (.0001 < r²s < .20), reassuring the distinctiveness of each latent variable in this study.

Table 2

Means, Standard Deviations (SD), Average Variance Extracted (AVE), and Correlations

Variables	Mean	SD	AVE	1	2	3	4	5
1. Gender	0.52	0.50	—
2. Age	23.8	4.51	−	−0.30**
3. CSS	2.80	0.74	0.60	−0.03	−0.12
4. Job autonomy	3.41	0.90	0.37	0.05	0.03	0.04
5. Emotional intelligence	3.87	0.53	0.40	0.01	0.08	−0.21**	0.19*
6. Emotional exhaustion	2.31	0.77	0.47	0.05	−0.08	0.45**	−0.15^†	−0.32**

Note: n = 173; gender is coded 1 = female and 0 = male; CSS =customer-related social stressors. All measures are rated on a 5-point Likert-type scale except for gender and age.

†

p < .1. *p < .05. **p < .01.

Preliminary Analyses

We reviewed the mean ratings of four study constructs and zero-order correlations among the four constructs (Table 2). The mean values of study constructs are: for EI, 3.87; for CSS, 2.80; for job autonomy, 3.41; and for EE, 2.31. Given that the highest possible rating is 5.0, respondents’ EI is moderately high. This indicates that the hospitality industry attracts workers who are emotionally intelligent. Respondents’ perceived CSS level is close to 3.0. This suggests that hospitality employees do experience CSS. Despite the possible, frequent encounters of CSS, the level of job autonomy (3.41) appears to be limited.

CSS is positively correlated with EE (r = .45, p < .01) as expected (H1_a); job autonomy is negatively correlated with EE (r = −.15, p = .07) as expected (H2_a). A few other noteworthy correlations are a negative association between EI and EE (r = −.32, p < .01), a negative association between EI and CSS (r = −.21, p < .01), and a positive association between EI and autonomy (r = .45, p < .01).

Hierarchical Regression

We performed hierarchical regression analyses to test research hypotheses. All interaction variables were mean-centered to reduce multicollinearity (Aiken & West, 1991). Gender and age were used as control variables because Maslach and Jackson’s (1981) seminal work showed the profound effect of gender and age on experienced burnout among many sociodemographic characteristics. The six steps of entering predictors, moderators, and interaction terms, and the results of regression models are shown in Table 3. After controlling for gender and age, CSS is positively associated with exhaustion (β = .49, p < .01), whereas job autonomy is negatively associated with exhaustion (β = −.20, p< .01; Model 2). These results reject null hypotheses H1₀ and H2₀, lending support to alternative hypotheses (H1_a and H2_a). As for the interaction between CSS and autonomy, the two-way interactive effect (CSS × Autonomy) on exhaustion is not significant (β = −.04, n.s.; Model 3), leading to the rejection of alternative hypothesis H3_a.

Table 3

Results of Moderated Regression Analyses

Variables	Emotional Exhaustion
Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Control variables
Gender	0.04	0.18	0.17	0.16	0.18	0.14
Age	−0.02	−0.004	−0.005	−0.005	−0.007	−0.01
Job demand and job control
CSS		0.49**	0.50**	0.46**	0.47**	0.41**
Job autonomy		−0.20**	−0.20**	−0.17**	−0.17**	−0.14*
Two-way interaction
CSS × Job autonomy			−0.04	−0.02	−0.11	−0.16
Individual characteristic
Emotional intelligence				−0.22*	−0.17	−0.21^†
Two-way interactions of emotional intelligence
CSS × Emotional intelligence					0.23	0.20
Job autonomy × Emotional intelligence					−0.19	−0.16
Three-way interaction
CSS × Job autonomy × Emotional intelligence						0.34*
F	0.96	10.24**	8.51**	8.04**	6.76**	6.65**
R	0.02	0.27	0.27	0.29	0.31	0.33
ΔR		0.25**	0.001	0.02*	0.02	0.02*

Note: CSS = customer-related social stressors.

†

p < .1. *p < .05. **p < .01.

The final hypothesis focuses on the moderating role of EI in the relationship of the interactive effect of CSS and autonomy on exhaustion. The three-way interactive effect (CSS × Autonomy × EI) on exhaustion is found to be significant (β = .34, p < .05; Model 6). Thus, the null hypothesis H4₀ is rejected. To understand the nature of the moderating relationship (alternative hypotheses H4(1)_a and H4(2)_a), the pattern of the interaction (Figure 2) was plotted according to the procedure recommended by Aiken and West (1991). In addition, simple slopes analyses (Jaccard, Wan, & Turrisi, 1990) and the slope differences test (Dawson & Richter, 2006) were utilized to interpret the details of interaction effects.

Figure 2

Three-Way Interaction Plots

As for employees high in EI, simple slope tests show that the two significant slopes are parallel between high job autonomy (Slope 1 = .54, p < .01) and low job autonomy (Slope 2 = .51, p < . 01). This seems to reject the stress-buffering effect of high autonomy on the relationship between CSS and exhaustion for high-EI personnel (H4(1)_a). The result of slope difference tests (Table 4) further validates this speculation; when EI is high, the slope difference (Slopes 1 and 2) is insignificant. Specifically, under a high-EI condition, when autonomy is high, exhaustion scores get higher as CSS becomes higher: from 1.93 to 2.72 (Slope 1 = .54). A similar pattern is observed with low autonomy. Under a high-EI condition, when autonomy is low, exhaustion ratings become higher as CSS becomes higher: from 2.35 to 3.09 (slope 2= .51). This indicates that for high-EI workers, autonomy (job control) and CSS (job demands) predict exhaustion mostly in an additive manner (main effects).

Table 4

Results of Slope Difference Tests for the Three-Way Interaction

Pair of Slopes	t
1 and 2	0.16
1 and 3	2.48*
1 and 4	−0.39
2 and 3	−1.78
2 and 4	−0.46
3 and 4	−2.27*

Note: 1 = high autonomy, high emotional intelligence; 2 = low autonomy, high emotional intelligence; 3 = high autonomy, low emotional intelligence; 4 = low autonomy, low emotional intelligence.

p < .05. **p < .01.

For low EI, the interaction between CSS and autonomy is found to be significant. Simple slope analyses reveal a significant, upward effect of CSS on EE when low-EI employees perceive low job autonomy (Slope 4 = .63, p < .01) compared with high job autonomy (Slope 3 = −.01, n.s.) (stress-exacerbating hypothesis). Moreover, the highest level of exhaustion occurs when EI is low, CSS is high, and job autonomy is low. All these results are supportive of alternative hypothesis H4(2)_a. Congruent with the results of simple slope tests, slope difference tests show that under a low EI condition, the two slopes (Slopes 3 and 4) are significantly different (p < . 05), implying that autonomy is indeed a salient moderator in the relationship between CSS and exhaustion. More specifically, under a low-EI condition, when job autonomy is low, exhaustion resulting from CSS becomes higher (from 2.35 to 3.25, Slope 4 = .63), while when job autonomy is high, exhaustion triggered by CSS remains almost the same or goes down slightly (from 2.71 to 2.69, Slope 3 = −.01). That is to say, CSS is positively related to exhaustion only when autonomy is low for those with low EI. In summary, the first alternative hypothesis H4(1)_a is rejected and the second alternative hypothesis H4(2)_a is supported.

Although tests comparing Slopes 1 and 2, and Slopes 3 and 4 are central to interpret H4(1)_a and H4(2)_a, respectively, the results of other slope difference tests confirm and enhance our understanding of the prominence of autonomy for low EI. For example, the significant slope difference between Slope 3 (low EI, high autonomy) and Slope 1 (high EI, high autonomy) evidently demonstrates that (as discussed before) thanks to high autonomy, the exhaustion level for a low-EI person remains the same from low CSS to high CSS, but high autonomy is not of much benefit to a high-EI person, as the exhaustion level increases along with CSS.

Discussion

This study shows that high demands are associated with a high level of strain (EE) employing specific job stressors, social interactions with customers. This is in line with the result of prior research that has used service samples, such as sales clerks in a shoe store (Dormann & Zapf, 2004) and employees in hotels and restaurants (Choi et al., 2012). It is worth contemplating the score of CSS (mean = 2.80) (Table 2); more close analyses show that around one third of employees (31%, n = 53) rate CSS between 3 (sometimes) and 4 (often), and about one tenth of employees (8%, n = 12) rate CSS greater than 4 (often to very often). This statistic reflects the harsh reality of CSS that hospitality employees frequently face.

In terms of the compatibility of job demands and control, we thought that autonomy, the job control variable chosen for this study, would be a good match with CSS, job demands. Nonetheless, the two-way interaction (CSS × Autonomy) is insignificant. This result suggests that the JD-C theory, as noted by many previous scholars, may need to be expanded, and with the inclusion of the third variable such as individual characteristics, the theory becomes more viable.

It is intriguing to see no stress-buffering effects of job autonomy for high EI in the linkage between CSS and exhaustion. The outcome of no interaction (buffering) effect may reflect the fact that the characteristic of EI is rather complex (Zeidner, Matthews, & Roberts, 2004). It is possible that in CSS situations, high-EI employees are able to comfort or relieve customers’ upsetting feelings or anxiety even before exerting job autonomy. This rationale can be explained by the theory of the COR (Hobfoll, 1989); the potential or actual loss of resources leads to negative states or even exhaustion if no action is taken to protect depleted resources (Hobfoll & Shirom, 1993). Individual characteristics are personal resources that affect the way in which individuals react to and help protect themselves from stress; the COR theory suggests that those with fewer resources suffer more when their resources are depleted. High-EI employees may have a “reservoir” of emotional resources that they can draw on during problematic interactions with customers and enable customers to feel and behave better. Thus, those with high EI may be less stressed and may not consume as many of their other resources, such as job autonomy.

The COR theory also suits the stress-exacerbating hypothesis under a low-EI condition. Employees who are low in EI are likely to be in need of additional resources to overcome CSS and maintain their job because they are short of personal resources to maneuver stressful situations. Therefore, EE increases more dramatically for low-EI individuals when autonomy (additional resource) is scarce. Last, it should be noted that for high-EI individuals, despite no interactive effects (CSS × Autonomy) on exhaustion, the main (direct) effect of autonomy on exhaustion is salient. This indicates that autonomy is an important job resource for high-EI employees as well.

Theoretical Implications

The first contribution of this article lies in the JD-C literature. The JD-C model has been widely used for decades as an important framework for occupational stress; however, it is apparent that there is room for further applications of the JD-C framework as the variety of jobs continues to grow. For hospitality services, intensive social interactions with customers are a distinguishable stressor from other general physical or psychological stressors (e.g., work overload and time pressure). This study advances the depth of the JD-C literature by addressing previously untapped demands, CSS in a specific occupational group such as hospitality.

At the same time, this study enriches hospitality burnout literature given that very few hospitality scholars have adopted the JD-C model. Building on the direct relationship between CSS and EE that hospitality scholars have previously established, this study proposes possible moderating hypotheses (stress buffering and stress exacerbating) to advance the theory, utilizing the JD-C framework. In addition, we constructed a three-way interaction with two moderators to provide further insight into the ongoing debate regarding the solidity of the JD-C model.

Next, EI continues to draw scholarly attention because of its impact on employee performance and well-being (Joseph, Jin, Newman, & O’Boyle, 2015; Laborde, Lautenbach, Allen, Herbert, & Achtzehn, 2014). The effect of EI on work stress is fairly well documented (Lee & Ok, 2012), but little is known about the interactive role of EI in work stress. This study illustrates an example of how job/organizational characteristics and individual dispositions work together to reduce job strain, utilizing CSS (strain), autonomy (job variable), and EI (individual variable). In summary, this study enhances our understanding of EI in the workplace and contributes to the EI literature.

Practical Implications

The findings of this study have managerial implications for work design and organizational interventions to reduce job stress among hospitality frontline employees. As indicated in the discussion section, stress occurs when dealing with difficult customers. To reduce the negative effects of CSS on service providers, preventative strategies are recommended. First, hospitality companies should realize that “customers are not always right,” which may be against the traditional mentality of the service company. For example, a zero-tolerance policy toward customer abuse of frontline employees along with supervisors’ prompt intervention in such situations could be implemented. Such policies or procedures are likely to send a powerful signal that companies genuinely care about the well-being of their service employees.

Second, short breaks after stressful encounters and opportunities to briefly express concerns about misbehaving customers to experienced coworkers or supervisors can ease line employees’ EE before they return to high-quality service. In other words, coworkers and supervisors should be trained in how to support line employees, who have negative interactions with customers, by sharing successful experiences and strategies in problematic situations. This way, coworkers and supervisors become a helpful resource for stressed service employees (Kim, Hur, Moon, & Jun, 2017).

Third, although hiring applicants with high EI is ideal for hospitality companies, in reality, hospitality firms may not always secure top candidates with high EI. The most critical finding of this study is that even low-EI personnel have a chance to thrive in the hospitality workplace with training on the proper use of job autonomy, which in turn helps regulate stress. For example, companies could use training programs of “lens-of-the-customers” or empathy-type to enhance knowledge about customers’ unmet needs and their intention and behavior (Sliter, Jex, Wolford, & McInnerney, 2010), potentially increasing low-EI employees’ emotional capacity. Companies can then teach employees how to exert their job autonomy and take appropriate discretionary actions. In addition to face-to-face training, in this digital age, it is feasible that all different service scenarios based on real stories can be collected from previous or present frontline employees and shared online (with suggested actions or actual actions taken). Employees, such as low EI ones, could refer to this collection of stories whenever needed. For cost-conscious, small hospitality firms, this online tool alone may work effectively.

One of the best hospitality companies that has mastered the importance of job autonomy is Ritz-Carlton; Ritz’s employees are given plentiful independence and allowed to spend $2,000 (per employee, per guest) to resolve any customer complaints (Brooks, 2000). This discretionary fund signifies the company’s trust in and empowerment of its employees. It is often quoted that no two days are the same in hospitality. Given this situation, as seen in Ritz-Carlton, top managers should be confident in their frontline employees’ discretionary actions and boost employees’ confidence with well-crafted service training in a supportive work environment so that they can handle various challenges effectively.

Limitations and Future Research

Data of this study are from hotels and restaurants in one major city in the United States. In the future, it is recommended that hospitality researchers examine Karasek’s interactive JD-C model using various segments and different regions or countries. For example, service failures in the airline industry are devastating compared with service problems in hotels or restaurants. Missed or cancelled flights often bring serious consequences to customers, not to mention the high price of air travel. Employees’ ability to resolve service issues as soon as possible is therefore essential in airlines. Thus, interactions between autonomy and CSS may be more pronounced, lending support to Karasek’s original moderating hypothesis, regardless of individual characteristics. From a data analysis perspective, although the measurement model of this study is sound with a good model fit and reasonable reliabilities and validities of study constructs, a relatively small sample size may preclude some significant effects (lack of power). Future study is warranted to validate the weak, two-way moderation effect with larger samples.

Also, it would be interesting to see how autonomy influences frontline employees in collective Asian culture where supervisors play a more prominent role in handling complaints; autonomy may have little impact on or actually increase frontline employees’ stress rather than buffer or decrease stress. EI is treated as one construct with a combination of four dimensions combined in this study. Future study could benefit from investigating the moderating role of each dimension of EI. Last, it would be worthwhile to test other individual characteristics as a third variable in the hospitality work setting. For example, the dispositional coping style is a good candidate as a third variable in the JD-C model (three-way interactions) to see which type of copers utilize autonomy more actively to reduce exhaustion or stress.

Concluding Remarks

In conclusion, this study investigates how CSS, job autonomy, EI, and EE are related to one another and highlights the important challenges that are associated with work design and organizational interventions. From a theoretical point of view, this study addresses a relatively unexplored area (JD-C model with a three-way interaction), providing empirical evidence of how the joint effect of individual and organizational resources can ease employees’ exhaustion from frequent social interactions with customers. From a practical perspective, the findings suggest that, in addition to recruiting hospitality-fit people (e.g., high EI), hospitality companies must develop specific polices, trainings, and practices for frontline employees to be resilient to CSS.

Footnotes

Zihui Ma, PhD (e-mail: zihui.ma@unlv.edu), is an assistant professor in residence in the William F. Harrah College of Hospitality, University of Nevada, Las Vegas, NV. Hyun Jeong Kim, PhD (e-mail: jennykim@wsu.edu), is a professor in the School of Hospitality Business Management, Washington State University, Pullman, WA. Kang Hyun Shin, PhD (e-mail: khs9933@ajou.ac.kr), is a professor at the Division of Social Sciences, Ajou University, Seoul, Republic of Korea.

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.

Anderson

J. C.

Gerbing

D. W.

(1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423. doi:10.1037/0033-2909.103.3.411

Ashforth

B. E.

(1993). Emotional labor in service roles: The influence of identity. Academy of Management Review, 18, 88-115. doi:10.5465/AMR.1993.3997508

Austin

E. J.

Saklofske

D. H.

Egan

(2005). Personality, well-being and health correlates of trait emotional intelligence. Personality and Individual Differences, 38, 547-558.

Babakus

Yavas

Karatepe

O. M.

(2008). The effects of job demands, job resources and intrinsic motivation on emotional exhaustion and turnover intentions: A study in the Turkish hotel industry. International Journal of Hospitality & Tourism Administration, 9, 384-404. doi:10.1080/15256480802427339

Bakker

A. B.

Demerouti

Euwema

M. C.

(2005). Job resources buffer the impact of job demands on burnout. Journal of Occupational Health Psychology, 10, 170-180.

Bourbonnais

Comeau

Vézina

(1999). Job strain and evolution of mental health among nurses. Journal of Occupational Health Psychology, 4, 95-107.

Brooks

(2000). Why loyal employees and customers improve the bottom line. Journal for Quality and Participation, 23, 40-44.

Buck

(1984). The communication of emotion. New York, NY: Guilford Press.

10.

Bureau of Labor Statistics. (2017). Job openings and labor turnover survey (home page). Retrieved from https://www.bls.gov/jlt/

11.

Chang

S.-J.

van Witteloostuijn

Eden

(2010). From the editors: Common method variance in international business research. Journal of International Business Studies, 41, 178-184. doi:10.1057/jibs.2009.88

12.

Chiang

F. F. T.

Birtch

T. A.

Kwan

H. K.

(2010). The moderating roles of job control and work-life balance practices on employee stress in the hotel and catering industry. International Journal of Hospitality Management, 29, 25-32. doi:10.1016/j.ijhm.2009.04.005

13.

Choi

C. H.

Kim

Lee

S. K.

(2012). Testing the stressor-strain-outcome model of customer-related social stressors in predicting emotional exhaustion, customer orientation and service recovery performance. International Journal of Hospitality Management, 36, 272-285.

14.

Ciarrochi

J. V.

Deane

F. P.

(2001). Emotional competence and willingness to seek help from professional and nonprofessional sources. British Journal of Guidance and Counselling, 29, 233-246.

15.

Dawson

J. F.

Richter

A. W.

(2006). Probing three-way interactions in moderated multiple regression: Development and application of a slope difference test. Journal of Applied Psychology, 91, 917-926.

16.

de Jonge

Dollard

M. F.

Dormann

Le Blanc

P. M.

Houtman

I. L. D

. (2000). The demand-control model: Specific demands, specific control, and well-defined groups. International Journal of Stress Management, 7, 269-287.

17.

de Jonge

van Vegchel

Shimazu

Schaufeli

Dormann

. (2010). A longitudinal test of the demand–control model using specific job demands and specific job control. International Journal of Behavioral Medicine, 17, 125-133.

18.

Deloitte. (2010). Hospitality 2015, Game changers or spectators (Research report). London, England: Author.

19.

Dhar

R. L.

(2016). Ethical leadership and its impact on service innovative behavior: The role of LMX and job autonomy. Tourism Management, 57, 139-148.

20.

van der

Doef, M.

Maes

(1999). The job demand-control (-support) model and psychological well-being: A review of 20 years of empirical research. Work & Stress, 13, 87-114. doi:10.1080/026783799296084

21.

Dormann

Zapf

(2004). Customer-related social stressors and burnout. Journal of Occupational Health Psychology, 9, 61-82.

22.

Dudenhöffer

Dormann

(2013). Customer-related social stressors and service providers’ affective reactions. Journal of Organizational Behavior, 34, 520-539.

23.

Endler

N. S.

Parker

J. D. A.

(1994). Assessment of multidimensional coping: Task, emotion, and avoidance strategies. Psychological Assessment, 6, 50-60. doi:10.1037/1040-3590.6.1.50

24.

Erez

Isen

A. M.

(2002). The influence of positive affect on the components of expectancy motivation. Journal of Applied Psychology, 87, 1055-1067.

25.

Fernet

Guay

Senécal

(2004). Adjusting to job demands: The role of work self-determination and job control in predicting burnout. Journal of Vocational Behavior, 65, 39-56. doi:10.1016/S0001-8791(03)00098-8

26.

Fornell

Larcker

D. F.

(1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 39-50.

27.

Grandey

A. A.

(2000). Emotional regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5, 95-110.

28.

Györkös

Becker

Massoudi

de Bruin

G. P.

Rossier

(2015). The impact of personality and culture on the job demands-control model of job stress. Swiss Journal of Psychology/Schweizerische Zeitschrift Für Psychologie/Revue Suisse de Psychologie, 71, 21-28. Retrieved from http://econtent.hogrefe.com/doi/full/10.1024/1421-0185/a000065

29.

Hackman

J. R.

Oldham

G. R.

(1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60, 159-170. doi:10.1037/h0076546

30.

Hancock

G. R.

Mueller

R. O.

(2001). Rethinking construct reliability within latent variable systems. Structural Equation Modeling: Present and Future, 195-216.

31.

Hart

C. W. L.

Heskett

J. L.

Sasser

W. E.

Jr. (1990). The profitable art of service recovery. Harvard Business Review, 68, 148-156.

32.

Hausser

J. A.

Mojzisch

Niesel

Schulz-Hardt

(2010). Ten years on: A review of recent research on the job demand-control (-support) model and psychological well-being. Work & Stress, 24, 1-35. doi:10.1080/02678371003683747

33.

Hobfoll

S. E.

(1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513-524.

34.

Hobfoll

S. E

Shirom

(1993). Stress and burnout in the workplace: Conservation of resources. In Golembiewski

R. T.

(Ed.), Handbook of organizational behavior (Vol. 1, pp. 41-60). New York, NY: M. Dekker.

35.

Hochschild

(1983). The managed heart: Commercialization of human feeling. Berkeley: University of California Press.

36.

H.-H.

H.-Y.

King

(2017). Impacts of misbehaving air passengers on frontline employees: Role stress and emotional labor. International Journal of Contemporary Hospitality Management, 29, 1793-1813. doi:10.1108/IJCHM-09-2015-0457

37.

Huang

Chan

S. C.

Lam

Nan

(2010). The joint effect of leader–member exchange and emotional intelligence on burnout and work performance in call centers in China. International Journal of Human Resource Management, 21, 1124-1144.

38.

Huang

Miao

(2016). Illegitimate customer complaining behavior in hospitality service encounters: A frontline employee perspective. Journal of Hospitality & Tourism Research, 40, 655-684. doi:10.1177/1096348013515916

39.

Jaccard

Wan

C. K.

Turrisi

(1990). The detection and interpretation of interaction effects between continuous variables in multiple regression. Multivariate Behavioral Research, 25, 467-478.

40.

Jenkins

(1991). Demographic aspects of stress. In Personality and stress: individual differences in the stress process (pp. 107-132). Oxford, England: Wiley.

41.

Joseph

D. L.

Jin

Newman

D. A.

O’Boyle

E. H.

(2015). Why does self-reported emotional intelligence predict job performance? A meta-analytic investigation of mixed EI. Retrieved from http://psycnet.apa.org/psycinfo/2014-39897-001/

42.

Jung

H. S.

Yoon

H. H.

(2016). Why is employees’ emotional intelligence important? The effects of EI on stress-coping styles and job satisfaction in the hospitality industry. International Journal of Contemporary Hospitality Management, 28, 1649-1675.

43.

Karasek

(1979). Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly, 24, 285-308.

44.

Karasek

Theorell

(1990). Healthy work : Stress, productivity, and the reconstruction of working life. New York, NY: Basic Books.

45.

Karatepe

O. M.

(2011). Do job resources moderate the effect of emotional dissonance on burnout? A study in the city of Ankara, Turkey. International Journal of Contemporary Hospitality Management, 23, 44-65.

46.

Karatepe

O. M.

(2013). The effects of work overload and work-family conflict on job embeddedness and job performance: The mediation of emotional exhaustion. International Journal of Contemporary Hospitality Management, 25, 614-634.

47.

Karatepe

O. M.

(2015). Do personal resources mediate the effect of perceived organizational support on emotional exhaustion and job outcomes? International Journal of Contemporary Hospitality Management, 27, 4-26.

48.

Karatepe

O. M.

Aleshinloye

K. D.

(2009). Emotional dissonance and emotional exhaustion among hotel employees in Nigeria. International Journal of Hospitality Management, 28, 349-358. doi:10.1016/j.ijhm.2008.12.002

49.

Karatepe

Yorganci

Haktanir

(2009). Outcomes of customer verbal aggression among hotel employees. International Journal of Contemporary Hospitality Management, 21, 713-733.

50.

Kim

H. J.

(2008). Hotel service providers’ emotional labor: The antecedents and effects on burnout. International Journal of Hospitality Management, 27, 151-161. doi:10.1016/j.ijhm.2007.07.019

51.

Kim

H. J.

Agrusa

(2011). Hospitality service employees’ coping styles: The role of emotional intelligence, two basic personality traits, and socio-demographic factors. International Journal of Hospitality Management, 30, 588-598. doi:10.1016/j.ijhm.2010.11.003

52.

Kim

H. J.

Hur

W.-M.

Moon

T.-W.

Jun

J.-K.

(2017). Is all support equal? The moderating effects of supervisor, coworker, and organizational support on the link between emotional labor and job performance. BRQ Business Research Quarterly, 20, 124-136.

53.

Kim

H. J.

Shin

K. H.

Swanger

(2009). Burnout and engagement: A comparative analysis using the Big Five personality dimensions. International Journal of Hospitality Management, 28, 96-104. doi:10.1016/j.ijhm.2008.06.001

54.

Kim

Stoner

(2008). Burnout and turnover intention among social workers: Effects of role stress, job autonomy and social support. Administration in Social Work, 32(3), 5-25.

55.

Kim

T. T.

Paek

Choi

C. H.

Lee

(2012). Frontline service employees’ customer-related social stressors, emotional exhaustion, and service recovery performance: Customer orientation as a moderator. Service Business, 6, 503-526.

56.

Kline

(1998). Principles and practice of structural equation modeling. New York, NY: Guilford Press.

57.

Laborde

Lautenbach

Allen

M. S.

Herbert

Achtzehn

(2014). The role of trait emotional intelligence in emotion regulation and performance under pressure. Personality and Individual Differences, 57, 43-47.

58.

Lee

J. J.

(2012). Reducing burnout and enhancing job satisfaction: Critical role of hotel employees’ emotional intelligence and emotional labor. International Journal of Hospitality Management, 31, 1101-1112.

59.

MacKenzie

S. B.

Podsakoff

P. M.

(2012). Common method bias in marketing: Causes, mechanisms, and procedural remedies. Journal of Retailing, 88, 542-555. doi:10.1016/j.jretai.2012.08.001

60.

Maslach

Jackson

S. E.

(1981). Maslach burnout inventory manual. Palo Alto, CA: Consulting Psychologists Press.

61.

Meier

L. L.

Semmer

N. K.

Elfering

Jacobshagen

(2008). The double meaning of control: Three-way interactions between internal resources, job control, and stressors at work. Journal of Occupational Health Psychology, 13, 244-258. doi:10.1037/1076-8998.13.3.244

62.

Mikolajczak

Menil

Luminet

(2007). Explaining the protective effect of trait emotional intelligence regarding occupational stress: Exploration of emotional labour processes. Journal of Research in Personality, 41, 1107-1117.

63.

Mikolajczak

Nelis

Hansenne

Quoidbach

(2008). If you can regulate sadness, you can probably regulate shame: Associations between trait emotional intelligence, emotion regulation and coping efficiency across discrete emotions. Personality and Individual Differences, 44, 1356-1368.

64.

Min

Kim

H. J.

Lee

S.-B.

(2015). Extending the challenge–hindrance stressor framework: The role of psychological capital. International Journal of Hospitality Management, 50, 105-114.

65.

Min

Park

Kim

H. J.

(2016). Common method bias in hospitality research: A critical review of literature and an empirical study. International Journal of Hospitality Management, 56, 126-135.

66.

Muthén

(2012). Mplus. The comprehensive modelling program for applied researchers: User’s guide (5th ed.). Los Angeles, CA: Muthén & Muthén.

67.

National Restaurant Association. (2017). Hospitality employee turnover rate edged higher in 2016. Retrieved from https://www.restaurant.org/Articles/News/Batch2_201901/Hospitality-employee-turnover-rate-edged-higher-in

68.

Nunnally

J. C.

Bernstein

I. H.

(1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill.

69.

O’Neill

J. W.

Xiao

(2010). Effects of organizational/occupational characteristics and personality traits on hotel manager emotional exhaustion. International Journal of Hospitality Management, 29, 652-658.

70.

Park

Searcy

(2012). Job autonomy as a predictor of mental well-being: The moderating role of quality-competitive environment. Journal of Business and Psychology, 27, 305-316.

71.

Parker

S. K.

Sprigg

C. A.

(1999). Minimizing strain and maximizing learning: The role of job demands, job control, and proactive personality. Journal of Applied Psychology, 84, 925-939.

72.

Petrides

K. V.

Furnham

(2003). Trait emotional intelligence: Behavioural validation in two studies of emotion recognition and reactivity to mood induction. European Journal of Personality, 17, 39-57.

73.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J.-Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. doi:10.1037/0021-9010.88.5.879

74.

Prentice

King

(2011). The influence of emotional intelligence on the service performance of casino frontline employees. Tourism and Hospitality Research, 11, 49-66.

75.

Ross

G. F.

(1997). Career stress responses among hospitality employees. Annals of Tourism Research, 24, 41-51. doi:10.1016/S0160-7383(96)00032-1

76.

Salovey

Bedell

B. T.

Detweiler

J. B.

Mayer

J. D.

(2000). Current directions in emotional intelligence research. In Lewis

Haviland-Jones

J. M.

(Eds.), Handbook of emotions (2nd ed., pp. 504-520). New York, NY: Guilford Press.

77.

Schutte

N. S.

Malouff

J. M.

Thorsteinsson

E. B.

Bhullar

Rooke

S. E.

(2007). A meta-analytic investigation of the relationship between emotional intelligence and health. Personality and Individual Differences, 42, 921-933. doi:10.1016/j.paid.2006.09.003

78.

Sliter

Jex

Wolford

McInnerney

(2010). How rude! Emotional labor as a mediator between customer incivility and employee outcomes. Journal of Occupational Health Psychology, 15, 468-481.

79.

Tracey

J. B.

Hinkin

T. R.

(2006). The costs of employee turnover: When the devil is in the details. Cornell Hospitality Report, 6(15), 4-16.

80.

Trinidad

D. R.

Johnson

C. A.

(2002). The association between emotional intelligence and early adolescent tobacco and alcohol use. Personality and Individual Differences, 32, 95-105.

81.

Van Yperen

N. W.

Hagedoorn

. (2003). Do high job demands increase intrinsic motivation or fatigue or both? The role of job control and job social support. Academy of Management Journal, 46, 339-348. doi:10.2307/30040627

82.

Weber

Bradley

G. L.

Sparks

(2017). Stressor effects of negative online reviews on anger and burnout in the restaurant industry. International Journal of Contemporary Hospitality Management, 29, 2847-2866.

83.

Wong

C. S.

Law

K. S.

(2002). The effects of leader and follower emotional intelligence on performance and attitude: An exploratory study. The leadership quarterly, 13(3), 243-274.

84.

Yoo

J. J.-E.

Kim

T. T.

Lee

(2015). When customers complain: The value of customer orientation in service recovery. Cornell Hospitality Quarterly, 56, 411-426. doi:10.1177/1938965514566070

85.

Zeidner

Matthews

Roberts

R. D.

(2004). Emotional intelligence in the workplace: A critical review. Applied Psychology, 53, 371-399.

From Customer-Related Social Stressors to Emotional Exhaustion: An Application of the Demands–Control Model

Abstract

Keywords

Literature Review and Hypotheses Development

Job Demands–Control Model

Customer-Related Social Stressors and Emotional Exhaustion

Job Autonomy and EE and the Moderating Role of Autonomy in the Relationship Between CSS and Exhaustion

EI and Its Moderating Role in the Interactive Effect of CSS and Autonomy on EE

Method

Procedure and Sample

Measures

Analysis and Results

Common Method Bias

Measurement Model

Preliminary Analyses

Hierarchical Regression

Discussion

Theoretical Implications

Practical Implications

Limitations and Future Research

Concluding Remarks

Footnotes

References