The effects of emotional labour on the outcomes of the job and the organization: Do the differences in age and the manager’s emotional intelligence have any impact in the hotel business?

In the high competition of the hospitality and service industry, hotels are in the process of intensely managing their employees’ behaviour and emotions, and this can have a detrimental impact on frontline employees’ work attitudes, job performance, and behaviour within the organization. In Thailand, there are still pressing questions regarding these issues. The objectives of this study were to examine the relationships among emotional labour, burnout, job satisfaction, turnover intention, and organizational commitment, and to compare the moderating roles of the employee’s age and manager emotional intelligence affecting emotional labour concerning the job satisfaction and burnout of hotel frontline employees in Thailand. A total of 509 subjects were investigated. With the structural equation modelling (SEM) and multiple group analysis technique used in this study, the positive and negative significant effects of emotional labour concerning job and organizational outcomes, and the moderating roles of age and manager emotional intelligence, were determined. The managerial implications from the results of the study can have beneficial outcomes for the hospitality and service industry, particularly regarding the human resource management process in terms of dealing and custom treat with employees’ emotional labour.


Introduction
In today's increasing forceful rivalry in the hospitality and service industry, highly-competitive businesses are in the process of intensively managing their employees' emotional behaviour (Hofmann and Stokburger-Sauer, 2017;Lee and Hwang, 2016). Emotional labour is the concept of how employees manage their emotions as part of their job demands (Hochschild, 2012). The concept concerns regulating both one's feelings and the expression of those feelings to adhere to an organization's policies and entails employees using emotion regulation strategies at work Hur et al., 2015) in order to sustain the goals of the organization (Lee and Ok, 2012). Employees, especially frontline hotel staff, are inclined to regulate the display of their emotions, and this results in two types of emotional labour behaviours, surface and deep acting, depending on the efforts and methods used to modify their emotions (Liu, 2017). When engaging in many tasks with challenging service encounters, employees might encounter the effect of both negative and positive sides of emotional labour (Humphrey et al., 2015;Rathi and Lee, 2016), potentially leading to a detrimental impact on an individual's work attitudes (Zhang et al., 2018), job performance, and behaviour within the organization.
Within the emotional labour model, burnout has been labelled as the biggest occupational hazard (Lee and Ok, 2015). Burnout is linked to work stress as a psychological fatigue condition resulting in the accumulation of decreased physical energy (Shani and Pizam, 2009), lowered immunity, more work dissatisfaction and pessimism, frequent work absence, work ineffectiveness, and ultimately job dissatisfaction (Lu and Gursoy, 2016). These could lead to unwanted outcomes, such as lower organizational commitment and increased absenteeism and turnover intentions (Deery, 2008;Low et al., 2001), which are among the distinctive structures (Carbery et al., 2003;Payne and Huffman, 2005)of an organization. In order to solve such problems, the organization should heighten employees' job satisfaction and strengthen their organizational commitment so that they can be retained and so that the turnover intention rate can be reduced (Wu, 2011). Employees with higher commitment will produce more highly-effective performances and exhibit lower levels of turnover intention (Karatepe, 2014). This will in turn help create business competitiveness as well as service quality, which are the heart of a business. In addition, enhancing the quality of life of employees and increasing their job satisfaction could help to reduce their stress at work and represent good corporate governance (Davis and Newstrom, 2014).
The differences in moderating factors, such as the age of the manager and his or her emotional intelligence, could also result in a change in the level of the influence of emotional labour on its outcomes (Allen et al., 2010;Chen et al., 2019). Age acts as a significant moderator, having different results regarding emotional labour and its outcomes (Cheung and Tang, 2010;Hur et al., 2014;Walsh and Bartikowski, 2013). The support of the manager's emotional intelligence also moderates the effect of emotional labour on job outcomes, and the relationship between emotional labour and burnout, as well as that between supervisory support and emotional labour and the job ), which will influence the level of the impact of emotionally-charged labour concerning relevant variables (Han et al., 2017;Wolfe and Kim, 2013). From the abovementioned findings, differences in moderating factors such as age and manager emotional intelligence could result in a change in the level of the influence of emotional labour regarding its outcomes.
The strong growth rates in the tourism industry in Thailand have resulted in significant investments, particular in the country's hotel industry. The number of tourists visiting Phuket, a province of the South of Thailand, is likely to increase continuously with more than 13 million Thai and foreign tourists going there each year. In 2016, 9,641,703 foreign tourists visited, accounting for 71.0 percent of all the tourists in Thailand (Phuket Provincial Statistical Office, 2017). Additionally, the number of hotels in southern Thailand has increased to 13,327 in 2019. This represents a steady increase to 5,874 rooms from 2016 to 2019, with the number of new hotels being mostly 4-5-star ones located throughout Phuket. The visiting of tourists to Phuket has increased the number of new operators, resulting in high competition in the area. The increase in the number of tourists both domestic and international is reflected by higher hotel occupancy rates. However, the current number of skilled workers in the south are scarce and Thai employees have relatively high turnover rates (Bank of Thailand, 2018).
In the evolution of the accommodation sector, changes and growth in labour markets have become more visible in hotels, and employment patterns responsive to skills development are needed for the Thailand 4.0 industrialized era (Ministry of labour Thailand, 2017a, 2017bWorld Bank Group, 2018). Hotels and restaurants together have accounted for 5.8% of the GDP in Thailand with an expansion of 8.5% in 2017. Nevertheless, pressing issues remain that need to be addressed, such as a skilled personnel shortage and the desired support and assistance from the government (National Statistical Office, 2013). Moreover, the shortage of hotel labour has resulted from a high turnover regarding new hires (Ministry of labour Thailand, 2012, 2013a, 2013b, 2013c. It would be challenging and critical to embark on a study of this topic in order to determine and elucidate the problem of burnout and turnover intentions in hospitality industry, and it is particularly worth investigating the moderating role of the differences of age, as well as how the manager's emotional intelligence support affects job outcomes. The objectives of this study were to examine the relationships among the emotional labour, burnout, job satisfaction, turnover intention, and organizational commitment of hotel frontline employees. Moreover, the study aimed to compare the differences of an individual's age and the manager's emotional intelligence as moderating roles influencing the burnout of hotel frontline employees in Thailand. In this study, 509 hotel frontline employees were investigated by selfadministered questionnaires with the structural equation modelling (SEM) and multiple group analysis technique. The results from the study's model could render several direct benefits to hotels, and related sectors by providing suggested managerial implications for human resource management in order to enhance beneficial strategies for emotional labour, specifically regarding differences in the age of employees and the manager's emotional intelligence support for hotel frontline employees.

Literature review and hypothesis development
The concept and strategy of emotional labour The concept of emotional labour concerns how frontline employees manage their emotions as part of their job demands (Hochschild, 2012). This involves a process of regulating both one's feelings and expressions according to the organizational goals required by organizational display rules irrespective of felt emotions; hence, the use of emotional regulation strategies at work by employees Hur et al., 2015). This scenario mostly happens to hotel frontline employees as they are service employees that have direct interaction with customers, both face-to-face and voice-to-voice, to fulfill their needs (Hochschild, 2012). They must make a truthful effort to experience and display appropriate emotions and at the same time obey the organization's emotional display rules (Chen et al., 2019). As frontline employees act as a contact point between customers and the organization, they are viewed as a source of service differentiation or competitive advantage for companies (Tsaur and Tang, 2013).
Employees tend to regulate their emotional displays as hotel frontline service providers, depending on the efforts and methods used to modify their emotion-feeling rules and display rules, and most studies have focused on two strategies of emotional labour generally used by employees to manipulate their emotional displays at work (İplik et al., 2014;Lee and Ok, 2014;Liu, 2017;Shani et al., 2014). First is surface acting or "acting in bad faith," which refers to suppressing a person's true felt emotions and manipulate one's feelings in the desired direction. The way in which an individual manipulates his or her expressions is shown to the society in response to the expectations or rules of performance (display rules) in each situation. Frontline employees may feel cautious about the rules of the service that are not in line with his or her genuine feelings (adjusting observable expressions to mask one's true feelings and pretending to feel the desired emotion). Last, deep acting or "acting in good faith" refers to having the truly desired emotions and making an effort to experience the desired emotions that lead to ordinary displays of one's emotions. In this process an individual manipulates his or her inner emotions so that feelings can be reflected in a way that is visible in a manner consistent with the display rules in each situation. In this way, one's personal adjusted inner sense follows the external expression. The service provider endeavours to maintain a general attitude while holding off his or her emotions, dissatisfaction with customers, or trying to change his or her mood to appear happy or bright while actually being bored or tired (modifying one's true feelings in order to express the desired emotion). Employees need to regulate their feelings and make an effort to adjust their emotions to appropriate expression when engaging in surface acting (Krannitz et al., 2015;Lee and Ok, 2014). Individuals that express emotions that contradict their real feelings and emotions can face conflicts and dissatisfaction (Grandey, 2000;Hochschild, 2012). The increase in acting on the surface can eventually result in energy depletion (Hur et al., 2015;Krannitz et al., 2015).

The effects of emotional labour on burnout and job satisfaction
Surface acting is one element of emotional labour's strategies seen as a self-overwhelming strategy aimed at controlling emotional displays (Chau et al., 2009). It has been linked to substance abuse, headaches, absenteeism, and sexual dysfunction (Hochschild, 2012), and it has also been linked to negative impacts on the organization, such as lower job satisfaction , and job commitment and job performance (Diefendorff and Richard, 2003), and could lead to a reduced sense of personal accomplishment (Lee and Ok, 2014). Surface acting has also been negatively related to job satisfaction  and to emotional exhaustion beyond deep acting and dissonance (Brotheridge and Lee, 2003) and has been positively related to burnout . Surface acting, additionally, has been correlated with negative work-related outcomes, decreased work performance, and higher levels of burnout (Krannitz et al., 2015;Liu, 2017).
With deep acting, another strategy of emotional labour, the employee truly experiences the expressed positive emotions, and consequently the correspondence between one's feelings and emotional displays decreases the experience of negative emotions (Humphrey et al., 2015;Rathi, 2014;Shani et al., 2014). Many studies have found that deep acting subsidizes a greater sense of an individual's effectiveness at work. When employees make an effort to feel the organization's required display emotions, their level of job satisfaction and personal accomplishment will be increased . The only honest expressions can have positive outcomes for employees. Furthermore, both the frequency of interactions and display rules for showing positive emotions have been shown to be related to a heightened sense of personal accomplishment (Lee and Ok, 2014), and this may generate a sense of satisfaction through the quality of provided services (Hochschild, 2012) and heightened work performance .

Burnout
Burnout constitutes the signs of an individual or employee that is unreliable and is tired of a longterm job, leading to fatigue from excessive use of energy and thought. Symptoms of burnout have been discussed, first in terms of emotional exhaustion or exhaustion, frustration, and stress from work, which when experienced over a long period of time would lead to bitterness and unsuccessful work; second, depersonalization or an impairment of relationships with a bad attitude toward others, indifferent and uncaring behaviour of the person concerned, and paranoid and pessimistic views towards others; and third, a lack of personal accomplishment or feeling of unsuccessfulness and incompetence at work. It can involve the inability to motivate the people involved to achieve their expectations, and the inability to meet the expectations of the service recipients (Maslach and Jackson, 1981).
Burnout intensifies depression (Shani and Pizam, 2009). Cause of burnout is a development of loss of mind, lack of energy, and less work goals. These are consequences of working conditions such as lack of education or training, overwhelming amount of work, long working hours, low salaries and returns, and strict corporate policy.
From the above findings, hypotheses regarding the effects of emotional labour on burnout were developed as shown below.

Job satisfaction
Job satisfaction is determined by individuals' attitudes toward the job that encourages positive and pleasing emotional states when positive job values are promoted (Cheng and Yi, 2018). It comprises the condition of an individual's proper determination or attitude toward his or her work. Positive approaches of individuals to the job in terms of both physical and mental fulfilment will make them satisfied when they accomplish the objectives of the organization. Job satisfaction is important for employees, as a good working life responds to the needs of employees and positively affects their quality of life as well. This is also crucial for the business because it affects productivity, performance, profit, and many other outputs (Çelik et al., 2015). The hypotheses regarding the effects of emotional labour on job satisfaction were developed as shown below.
Hypothesis 2a: Surface acting negatively impacts employees' job satisfaction. Hypothesis 2b: Deep acting positively impacts employees' job satisfaction.

The moderating roles of age and the manager's emotional intelligence affect the relationship between emotional labour and burnout Age
Age-related motivational shifts lead individuals to change to active interaction with their environment. For instance, the optimization of emotional experience is prioritized in their future life. When elders' concern for their future is less relevant, attention to current feeling states will be heightened. The emotion regulation of individuals was improved across adulthood and into their older age. The age-related changes response to focused emotion regulation in coping with adversarial proceedings. Older people appear to better remember positive data than negative data. They have the ability to decrease certain negative emotional experiences in the interest of feeling more consistently positive, which makes them increasingly motivated to maximize their experience of positive emotions and to minimize the experience of negative emotions as they age (Charles and Carstensen, 2007).
People of different ages have different emotional management practices, and strategies to interact and cope with situations in their life. When an elderly individual's concern for the future is less relevant, attention to current feeling states will be heightened. This age-related motivational shift leads individuals to interact more with their current attitudes (Carstensen et al., 2003). Older people appear to, and better, remember a quantity of positive data than negative data. They have both the motive and the ability to decrease certain negative emotional experiences in the interest of feeling more consistently positive, which helps them to become increasingly motivated to maximize the experience of positive emotions and to minimize the experience of negative emotions (Charles and Carstensen, 2007). The motivational changes initiated by perceived restrictions concerning the rest of their lifetime make individuals more effortful and add instinctive strategies that support their emotion regulation.
Age relates positively to managing one's emotions (Hochschild, 2012). The elderly actually engage more in deep acting (Cheung and Tang, 2010) because they can draw from broader emotional memories than younger people, and therefore they are more capable of performing deep acting (Kruml and Geddes, 2000). They can capitalize on their positive emotions and diminish their negative emotional experiences (Dahling and Perez, 2010). In addition, older employees can control their emotions and external displays better than their younger peers (Kim, 2008). With greater work experience, more job demands, and fewer resources, older employees might be less able to use surface acting (Bakker et al., 2003). It could be said that older employees might adjust their predilection for displaying authentic emotions when they must deploy inauthentic emotions. The hypothesis below was consequently developed.
Hypothesis 3: The moderating role of age reduces the impact of deep acting on burnout.

Manager's emotional intelligence
Emotional intelligence has a direct impact on job performance, job satisfaction, interpersonal relationships, negotiation, organizational commitment, employee engagement, leadership, and organization citizenship behaviour (Mathew and Gupta, 2015). Emotional intelligence comprises the ability to recognize and express one's emotions appropriately, to use one's emotions in thinking and decision-making, and to regulate one's emotions in one's self and in others (Cummings and Worley, 2014). Emotional intelligence is significantly and positively related to job satisfaction and organizational commitment. Additionally, a positive relationship has been identified between job satisfaction and organizational commitment (Güleryüz et al., 2008). The manager that use effectively emotional strategies can control his/her emotions well to achieve active states, make correct decisions in demanding circumstances, and apply necessary skills for dynamic communal behaviour. He or she can exhibit emotional intelligence in monitoring his or her emotional expression is aware of what emotions are expected and (un)acceptable in a given interaction with his/her employees. This characteristic reflects the manager's ability to perceive and understand the emotions of his/her employees. The employees' perceptions of their managers' emotionality and sociability have been associated with increased employee satisfaction with their manager (Webb, 2009), and the emotional intelligence and support from a manager can have a significant impact on employees' job outcomes (Han et al., 2017) as it can reduce stress and burnout. Hence the following hypothesis was developed.
Hypothesis 4: The moderating role of the manager's emotional intelligence reduces the impact of deep acting on burnout.

The effects of burnout and job satisfaction on turnover intention and organizational commitment Turnover intention
A high turnover rate of employees can destroy the quality of service at a hotel (Brien et al., 2015). If staff with long work experience leave their positions, the organization will lose substantially in terms of cost, time, and effort. Understanding the characteristics of turnover intention is imperative for hotel managers because it is connected to adverse behaviour, such as absenteeism, complaints, imputations of responsibility to others, lack of commitment, and job dissatisfaction . Turnover intention is the withdrawal behaviour as part of organizational well-being within the emotional labour model and constitutes the likelihood of an employee to leave an organization (Cho et al., 2017). When employees start to mention their job negatively and have less involvement in the organization, the behaviour of turnover intention may appear (Karatepe, 2013). One of the distinguishing features of the hotel and hospitality industry is the employees' turnover intention (Carbery et al., 2003;Hwang et al., 2014). It is an unpleasant thing if employees remain dissatisfied with their current positions or organizations . Thus, the following hypotheses were developed.

Organizational commitment
Organizational commitment can be a virtuous forecaster of organizational effectiveness (Steers, 1977) as organizational commitment levels affect employee perceptions and satisfaction. It is one of the main factors in preventing employees' turnover intention. Organizational commitment is the employees' internalization of the organization's goals, the high effort for the benefit of the organization, and the willingness to continue working and wishing to stay in the organization. It can enhance the productivity, effectiveness, and decrease the personal turnover rate of the organization (Bayram, 2005). Organizational commitment is an employee's intention to make the best effort for the organization based on his or her role as a member (Johnson and Spector, 2007) and loyalty. It embraces the goals and values of the organization and involvement both mentally and actively in corporate activities. Allen and Meyer (1990) proposed that organizational commitment consists of three different aspects. The first involves affective commitment; that is, the desire (want) to be in the organization. The second comprises continuance commitment; the necessity (need) to be in the organization. Finally, normative commitment means the obligation (ought) to be in the organization.
Organizational commitment can be a good forecaster of organizational effectiveness (Patiar and Wang, 2016) as organizational commitment levels affect employee perceptions and satisfaction. Those that are committed to the organization will have positive attitudes toward it, and this can improve their level of job satisfaction. Organizational commitment can enhance a person's productivity and on-the-job effectiveness, and decrease the personnel turnover rate of the organization (Deepa et al., 2014).
When employees experience stress-related burnout, they absent themselves more frequently and their turnover intent increases (Kim, 2015). Job satisfaction has been seen to significantly contribute to greater affective and continued commitment to the organization and to lower employee turnover intentions (Yang, 2010). Many studies have found that the relationships among emotional labour, job satisfaction, and burnout affect turnover intention (Deconinck and Johnson, 2009;Seery and Corrigall, 2009), and a significant and negative relationship has been shown between job satisfaction and turnover intentions (El-Nahas et al., 2012). The lack of organizational commitment and a low level of job satisfaction can cause turnover intention (Wu, 2011). The following hypotheses were then developed.
The model study of emotional labour and the moderating roles of age and manager emotional intelligence regarding burnout, job satisfaction, turnover intention, and organization commitment has been developed as shown in Figure 1.

Figure 1. Model of emotional labour and moderating roles of age and manager emotional intelligence regarding the job and organizational outcomes Method
Sample and data collection The data were collected from 15 hotels in Phuket, Thailand using the multiple-step sampling technique. The random sampling technique was used to select the hotels, and purposive sampling was used to define the subjects comprising frontline hotel employees. With permission from the hotels, a total of 509 self-administered questionnaires were distributed.

Operational constructs and measurements
The original English questionnaires represented the study's seven research constructs and were translated into Thai and interpreted back into English in order to check for mistranslation. The first part of the questionnaire was developed to enable the researchers to understand the respondents' demographics and to separate the individuals according to age. The younger and older age groups and the low and high support of the manager's emotional intelligence were, however, separated according to the means. The second part pertained to the two elements of emotional labour strategy; surface acting was measured using eight items, and deep acting using five items. The scales derived from the study of Brotheridge and Lee (2003), and Chu and Murrmann (2006).The third part concerned burnout-emotional exhaustion, lack of personal accomplishment, and depersonalization -in total 22 items. The scale included items synthesized and created based on the Maslach Burnout Inventory (MBI) (Maslach and Jackson, 1986). In the fourth part, in order to evaluate job satisfaction, the researchers used scales derived from the MSQ (Minnesota Job Satisfaction Questionnaire) by Weiss et al. (1967) with 20 items. The fifth part for assessing turnover intentions totalled four items, synthesized based on Netemeyer et al. (1996) and . The sixth part for evaluating organizational commitment-affective commitment, continuance commitment, and normative commitment-totalled 24 items based on Allen and Meyer (1990). Lastly, the part including the manager's emotional intelligence totalled 16 items using questionnaires created based on Wong and Law (2002). This part was used only for analysis using the multiple-group technique for separating the samples into two groups and was not combined in the main structural model.

Analysis methods
The structural equation modelling (SEM) technique was used in this study. In the first stage, the structural equation model was developed concerning hypothesis testing regarding the research objectives, and to investigate the relationships among emotional labour burnout, job satisfaction, turnover intention, and organizational commitment. In the second stage, this study applied the multiple-group analysis technique in order to test the differentiations among similar models estimated for different groups or respondents. The moderating roles regarding age and the manager's emotional intelligence that moderated the effects of emotional labour concerning job outcomes were tested.

Descriptive statistics
The 509 subjects had ages ranging from their 20s to 50s. After identification by mean, 319 of the respondents were separated into the younger group, while 190 were placed in the older group. Regarding the support of the manager's emotional intelligence, 249 were separated into the lower group while 260 were placed in the high-level support group.

First stage of the SEM in the study
In this stage, a four-step approach (Mulaik and Millsap, 2000) was applied. Each latent variable was investigated using exploratory factor analysis in order to select the essential variables for the measurement model. In this process, some indicators with low factor loading were eliminated. Confirmatory factor analysis was performed to test the measurement model (Hair et al., 2010) and to investigate all of the construct variables and to evaluate the composite reliability and average variance extract, reported together with the Cronbach alpha and KMO values. In Table 1, the reliability of the observed variables in each construct was reported using Cronbach alpha, which passed the criteria (0.73 to 0.91). KMO and Bartlett's test of sphericity were used as measures for accepting the sample adequacy. The KMO of all the constructs ranged from 0.50 to 0.85.
Further, Bartlett's test of sphericity indicated less than 0.05, demonstrating the validity and suitability of the tool in this study. The composite reliability score was high (0.72 to 0.91) compared with a standard of 0.6 (Hair et al., 2010). The results of an average variance extracted (AVE) of each construct are between 0.51 and 0.75 that were all passed the criteria of the convergent validity. Table 1 and Table 2 report the results of the model fit indices. The results were above the criteria of a model fit index, supporting the unidimensionality of the scales. The results for adaptability were chisquared/degree of freedom (df) <2, CFI >0.92, TLI >0.92, RMSEA <0.08 and SRMR <0.08. Note: β = Standardized factor loading, df = degree of freedom, *t-value is significant at p <0.05 Table 3 presents the discriminant validity assessed by comparing the square root of average variance extracted in the diagonal with the correlation coefficients (off-diagonal) for each construct in the relevant rows. The correlation coefficients for each construct exhibit both positive and negative values between -0.06 and 0.59. After comparing these values with the correlation of interest, the square roots of average variance extracted values were all larger than the correlation values of the respective constructs. Thus, the discriminant validity could be accepted for this measurement model and supported the discriminant validity among the constructs. Lastly, the structural model was analysed. Figure 2 shows that the results of the model proved to have an acceptable fit. The chi-squared = 1464.736, degree of freedom = 713, CFI = 0.932, TLI = 0.926, RMSEA = 0.046, and SRMR = 0.070. For the hypothesis testing, regarding hypothesis 1a, the results showed that surface acting had significantly positive effects on burnout (β = 0.56, p*=0.01), and deep acting displayed significantly negative effects on burnout in hypothesis 1b (β = -0.60, p*=0.00). In hypothesis 2a, surface acting exhibited significantly negative effects on job satisfaction (β = -0.25, p*=0.00). Regarding hypothesis 2b, the result demonstrated that deep acting had significantly positive effects on job satisfaction (β = 0.56, p*=0.00). In hypothesis 5a, burnout displayed significantly positive effects on turnover intention (β = 0.53, p*=0.00). Concerning hypothesis 5b, the result revealed a significantly negative effect of job satisfaction concerning turnover intention (β = -0.11, p*= 0.03), and regarding hypothesis 6a, burnout exhibited significantly negative effects on organizational commitment (β = -0.15, p* = 0.02). Concerning hypothesis 6b, the result also showed significantly positive effects of job satisfaction on organizational commitment (β = 0.65, p*=0.00). Therefore, hypotheses 1a, 1b, 2a, 2b, 5a, 5b, 6a, and 6b were all supported in this study.

Second stage of the SEM in the study
The second SEM stage presented the analysis of the moderating effects using the multiple-group analysis approach regarding the research objectives-to compare the age differences of the frontline employees and the manager's emotional intelligence support that moderated the negative effects of emotional labour regarding job satisfaction and burnout.

Figure 2. Structural model of the effect of emotional labour on the job and organizational outcomes of hotel frontline employees in Thailand
In the first step, the multi-sample confirmatory factor analysis technique was used in order to test the measurement invariance in the steps described below. Configural invariance, metric invariance, and scalar invariance were tested across different population groups following the steps of the measurement model. The criteria for the invariance testing of the measurement model comprised a change in the chi-squared and a change in the degree of freedom with non-significance when comparing the nested models (Hair et al., 2010); that is, the configural model against metric model, and the metric model against the scalar model. In addition, the change in the comparative fit index (Cheung and Rensvold, 2002) and the root mean square error of approximation was not over the threshold value of 0.01 without a significant difference between the two nested models regarding model fit, indicating a strong invariance (Li et al., 2015).
In Table 4, testing the invariance in the measurement model began with testing for configural invariance, assessing whether the four constructs, namely surface acting, deep acting, burnout, and job satisfaction, were best described in the tool of this study across groups. Concerning the age of the individuals and the manager's emotional intelligence, the first stage of the configural invariance showed that all of the separated models exhibited an acceptable model fit, and all factor loadings were significant (p <0.05). Next, the metric invariance model was tested with factor loading constrained to be equal across each group. When compared with the configural invariance, the metric invariance testing stage exhibited a change in the chi-squared and a change in the degree of freedom, indicating no significant change, while the change in the CFI and the change in the RMSEA were below 0.01. This result indicated that factor loadings were invariant across each group. In order to establish scalar invariance, intercepts and factor loadings were constrained to be equal across groups. When compared with the metric invariance, the scalar invariance testing stage exhibited a change in the chisquared and a change in the degree of freedom, indicated no significant change, while the change in the CFI and the change in the RMSEA were below 0.01. These results suggested that configural, metric, and scalar invariance held across the age and the manager's emotional intelligence groups. As a result, the tool for this study could be reliably applied to the individuals in all groups.
In order to test hypothesis 3 and 4, this study compared the path coefficients across the groups of age and the manager's emotional intelligence. Multiple-group analysis in the SEM approach was used to test the moderating roles of the different employee groups. In order evaluate the differences across groups that could identify the moderating negative effects of deep acting on burnout, this study followed the criterion of a significant change in the chi-squared (p <0.05) (Hair et al., 2010;Schumacker and Lomax, 2004) between two nested structural models, i.e., the unconstrained model and the constrained model. Note: p-values were significant at p < 0.05, CFI = comparative fit index; RMSEA = root mean squared error of approximation; SRMR = standardized root mean square residual, MEI = manager's emotional intelligence Table 5 demonstrates the standardized path coefficients compared among groups. After constraining the parameters, it specified a change in the chi-squared, and indicated the significant role of the moderator. From the hypothesis testing, the results demonstrated that moderating roles were found. The older group and the high support group from the manager's emotional intelligence showed stronger negative effects of deep acting concerning burnout. The results strongly differed from those statistically significant in the constrained model regarding H3 and H4. Therefore, age and the manager's emotional intelligence were seen to moderate the effects of emotional labour on job outcomes. Note: β= Standardized factor loading, χ2=chi-squared, df=degree of freedom, *p<0.05, Low MEI= group of low support from manager's emotional intelligence, High MEI= group of high support from manager's emotional intelligence

Conclusion and recommendation
This study demonstrated that surface acting has a positive effect on burnout and a negative effect on job satisfaction. Hotel frontline employees, who often use a surface acting strategy while dealing with customers, will easily exhibit signs of burnout. Despite the value placed on one's emotions, the employees still could not express their emotions in order to prevent conflict with work expectations. Emotional indications, that are inconsistent with actual emotions and the general loss of appreciations between real feelings and required emotional displays, cause emotional dissonance by experiencing a psychological contradiction between genuine internal and emotional feelings (Grandey, 2000;Grandey et al., 2015). The negative effects of emotion at work may come from long or drawn-out conflict between emotional expression and emotion. When service employees use the strategy of becoming emotionless, they require appropriate emotional displays that are very difficult (Hochschild, 2012). In addition, inconsistency between external and internal emotional states is related to a higher level of burnout. The results revealed that surface acting is especially related to employees' emotional exhaustion, one of the elements of burnout (Brotheridge and Lee, 2003). Moreover, when employees feel that meeting emotional demands at work requires a strong effort and a feeling of detachment from customers, they may feel a lowered sense of personal accomplishment (Grandey, 2000). Thus, surface acting causes negative work-related outcomes, such as burnout, decreased work performance, and harmful well-being outcomes, and these can be linked to lower job satisfaction Lee and Ok, 2012).
In contrast, deep acting has a negative effect on burnout and a positive effect on job satisfaction. Deep acting is the strategy of trying to modify one's internal emotions to match the requirements of the circumstances. For example, hotel frontline employees may try to cheer themselves up before starting work or having interactions with their customers. When employees engage in deep acting to adapt to the organization's displayed rules, it prevents them from contrasting emotions; the conflicts between true emotion and appropriate emotion that need to be performed (Humphrey et al., 2015;Hur et al., 2015). Honest expressions could have positive outcomes for employees, and result in more effective emotional skills. When employees make an effort to feel the required emotions, they feel emotional congruence between their true feelings and their emotional display, hence increasing their sense of personal accomplishment. This feature will prevent them from symptoms of burnout. Additionally, deep acting reduces the sense of emotional exhaustion and decreases the discrepancy between one's internal feelings and external expressions, thus increasing the positive interaction of employees. Deep acting can also restore positive emotions and develop job satisfaction and employees' sense of wellbeing Humphrey et al., 2015). Moreover, it prevents burnout linked to work stress as a psychological fatigue condition resulting in decreased physical energy, lowered resistance, more work dissatisfaction, frequent work absence, and work ineffectiveness among hotel frontline employees (Jung and Yoon, 2014;Lee and Ok, 2014).
This study also found supporting evidence that burnout has a positive effect on turnover intention, and negative effects on organizational commitment. Hotel frontline employees encounter complicated interactions with demanding customers, a substantial number of work hours, job pressure, physical demands of the job, and inadequate supervision. Such experiences lead to burnout and emotional exhaustion. The result of a lowered sense of accomplishment will be occupational stress and job dissatisfaction. Turnover intention can be caused by a low level of job satisfaction (Balouch and Hassan, 2014;Hwang et al., 2014) and a lack of organizational commitment (Zopiatis et al., 2014). Employees that show the signs of burnout will tend to display behavioural turnover intentions and exhibit a low level of organizational commitment (Cho et al., 2017).
In contrast, job satisfaction has seen to be negatively associated with absenteeism and turnover intentions, and has been linked to decreased turnover intentions. Frontline hotel employees that show higher commitment will perform more effectively and have a lower level of desire to quite the organization. Similar to the support described in the study of Cho et al. (2017), both Lee and Ok (2014). Further, the study of Yang (2010) indicated that job satisfaction contributes to lower employee turnover intention and subsidizes higher organizational commitment. Additionally, El-Nahas et al.
(2012) found a negative relationship between job satisfaction and turnover intentions.
However, the moderating roles of age, concerning the negative effects of deep acting on burnout, were stronger among older frontline employees, meaning that older employees displayed a higher level of deep acting concerning their employment services and a weaker level of burnout, resulting in a stronger negative effect on burnout than that of the younger personnel. The older hotel frontline employees, who engage in higher levels of deep acting, will have a lower level of burnout than that of the younger personnel. Age was viewed as a moderator influencing different effects of emotional labour concerning burnout and job satisfaction (Cheung and Tang, 2010;Hur et al., 2014). Because the older employees can draw more easily from broader emotional memories than the young ones (Cheung and Tang, 2010), they can exhibit better performance of deep acting and this can help them prevent burnout.
The moderating role of high support from the manager's emotional intelligence effects was found to produce a negative relationship between deep acting and burnout. The results showed that the group of hotel frontline employees that received a high level of support from the emotional intelligence of their managers had a lower level of burnout than the group that received a weak level of support from the emotional intelligence of their managers. This result from the study is consistent with the study of Han et al. (2017), who indicated that the emotional intelligence and support from managers have a significant impact on employees' job satisfaction. Additionally, the support from managers' emotional intelligence could help employees make changes concerning the impact of emotional labour regarding burnout, as a higher level of job satisfaction was found to be associated with less burnout (Weng et al., 2011). Thus, the emotional intelligence of the manager can help hotel frontline employees develop a strategy of deep acting concerning their work, and this can prevent them from experiencing burnout.

Managerial implications for hotel frontline employees
The study contributes constructive outcomes that are expected to assist with further development of the selection, training, and special activities of frontline workers in this industry, as well as their fairer evaluation and welfare provision. This would lead to real quality service that would promote organizational performance, organizational commitment, better retention of employees, reduced turnover rates of employees, and cost of training, as well as improve service quality and business efficiency.
The negative results of emotional labour lead to hotel frontline employees' emotional exhaustion and burnout, and also reduce the employees' job satisfaction. Higher degrees of using emotion regulation on the job are related to higher levels of employees' emotional exhaustion and lower levels of employees' job satisfaction. Hospitality and service organizations, such as the hotel business, should heighten their employees' satisfaction with their job in order to lower turnover intention and to increase the meaningfulness of their work by setting clear goals consistent with the employee's values, beliefs, and personal goals. The progressions of surface acting and deep acting may be linked to employee stress and health, as well as organizational well-being, resulting in positive or negative effects on the individuals and their performance. In surface acting, individuals generally do not like to feel "fake." However, in the long term, the suppression of real emotions and expressing false feelings by using this strategy will result in stress outcomes. Surface acting might evoke the feeling from customers that the employees' service expressions are manipulated and look unnatural. When the customers feel insincerity in the service, they will tend to assess the quality of service at a low level. Frontline employees need to make greater effort to solve these challenges. In contrast, deep acting is another strategy of emotional labour that also requires adjusting one's emotions, but differently. Hotel frontline employees, engaging in deep acting through a good faith type of emotional labour, have reported lower emotional exhaustion and burnout. When the customer can feel real service emotion and realize that this type of employee is performing his or her service from his or her heart and is putting him or herself in the customer's shoes, this is when the employee feels that the problems of the customer are similar to their his or her own. The customers will make the service scenario easier and smoother. The organizations must strengthen the deep acting skills of employees. "Acting in good faith" involves truly desired emotions and effective efforts to experience desired emotions leading to an ordinary, believable and natural display. Employees manipulate their inner emotions so that their feelings can be reflected in a way that is visible in a manner consistent with the display rules in each situation. When their inner feelings are adjusted to follow their external expression, the inner feelings are consistent with the external appearance, which would not cause emotional conflicts and would not cause burnout or turnover intention. Therefore, deep acting involves trying to feel internally positive, yielding an authentic and positive emotional display involving resource supplements.
The core service concept of hospitality providers, such as hotel businesses, aims to provide satisfying experiences for customers. One of the essential responsibilities of frontline staff is to engage in positive emotions. Staff members should continuously be reminded that favourable display rules are efficient. Therefore, the organization should recognize, understand, and provide practical managerial implementation using the human resource management process requiring desirable emotional regulation strategies that must include means to support job engagement and improved methods to retain valuable employees. Increased involvement and employee loyalty within the organization (Miarkolaei, 2014) to contribute to job performance and organizational commitment (Goodwin et al., 2011) can be achieved following the processes described below.
First, proper and practical consideration should be included in the selection process for each type of employee. For instance, the selection procedures ought to include the assessment of prescribed job attributes (Mann, 2007) and endeavour to recruit and hire employees that are a good fit with frontline jobs requiring intense social interaction and direct actions towards effective hiring decisions , and candidates should be hired that focus on customer satisfaction and service quality excellence (El-Nahas et al., 2012;Lam and Chen, 2012;Lee and Ok, 2012). Hotel managers should conduct careful screening procedures regarding their customer service behaviours when hiring new employees.
Organizations may consider focusing on developing methods to identify and hire employees that can manage their emotions well (Cho et al., 2017). Employees may be involved in the initial evaluation to measure the emotion regulation practices in the selection process that could be used to identify which applicants constitute suitable matches for frontline hotel positions. Hotels should develop indicators that measure the emotion labour strategy of applicants in the process of recruiting and selecting them. Applicants that have more constructive attitudes would constitute desired hotel service employees. For instance, in the event of new employee hiring, the management might create some service scenarios for the applicants to visualize their natural reaction. Additionally, managers could provide problem-solving cases in order to allow qualified staff to show how they would solve problems effectively and best satisfy the customer. Creating these customer satisfaction situations should provide the candidate with the opportunity to demonstrate their confidence when facing front desk challenges. Organizations should use corporate actors to test the candidate's reaction they might be unaware. The mentioned scenarios might evaluate the candidates regarding their deep acting and emotional intelligence.
In the recruitment process, emotional intelligence assessment may help select those employees with high emotional intelligence. Moreover, the study's result found an indifference concerning gender, age of the hotel frontline employees. Therefore, older employees with lengthy work experience and high emotional intelligence would be more desirable and should be considered when hiring (Hur et al., 2014;Jung and Yoon, 2014). The organization may re-decide to adjust the requirements to recruit employees regarding age. The range of age requirements may be extended, for the reason that older and more experienced employees could perform better strategies involving emotional labour than their younger counterparts.
Second, plans and organized training programs should be developed to help with targeted training frontline employees to improve their emotional ability. Hotels should provide programs that manage emotions and motivate frontline employees to internalize their emotions rules at work in order to reduce the impact of harmful emotions. The training process could improve and maintain high quality services while protecting frontline employees from the adverse effects of emotional labour. Hospitality management should train employees in the proper control of emotional labour. A standard tutorial manual that details proper emotional expression in service encounters should be recorded and provided. Recently, most hospitality organizations have paid attention to the concept of customercentred behaviours. Thus, the hotel should also develop more training programs to treat and develop their employees' feelings and service strategies. An essential component of orientation programs is to match appropriate attitudes and display rules to hotel frontline employees by centring on customers' feelings and service quality as a focal point. Three techniques (Chu, 2002) help frontline employees clarify their roles and prevent irritated feelings among the service providers. First, managers could inform employees that the hotel is aware of and recognizes the emotional role that employees place in service provision. It could encourage employees to increase their service quality by using their emotional strategies and be more satisfied and committed to their work and to the organization. Second, managers could provide occasions to discuss employees' negative emotions caused by their jobs. Frontline employees can talk about and share their experiences, including the strategies to deal with and face challenges effectively. Finally, the hotel manager can develop programs to improve the ability of frontline employees to overcome annoyance or to avoid frustration. It could support employees in experiencing positive emotions through understanding and commiserating with customers, thus helping employees to change negative emotions appropriately as necessary, so that frontline employees may study quality customer service standards and adopt deep acting more simply. As indicated, deep acting has potential benefits for both employee and customer outcomes (Lee and Ok, 2014). In addition, the training process should be also strengthened through case studies. The service standards and procedures, including the service recovery process, should be adjusted and revised to make them current, including a discussion of cases and circumstances.
Third, hotels should create and clarify the roles of organizational support for frontline service employees. Especially regarding the concept of focusing on customers as a centre of attention, the organization should provide programs to develop customer service quality (Lee and Ok, 2012) that provides practical information and useful behaviours in performing during emotionally-challenging service encounters, especially in dealing with privileged customers (Chu, 2002). Training programs could develop customer service quality, especially privileged customers, in order to reduce the adverse effects of emotional labour, and also to personalize the employee-customer relationship (Lee and Hwang, 2016), for example by creating and supporting a customer-oriented culture, updating customer background information whenever possible, and ensuring that employees have a clear conception of their customers' behaviours (Chu and Murrmann, 2006). The high-ranking customers are known as customers that generate high income for the hotel. At the same time, they aim to receive high-quality services as well. Therefore, the learning policies, personal details, preferences, and basic needs of these customers mentioned above will enable hotel frontline employees to more easily provide quality services without having too many negative side effects of their emotional strategy (surface acting). It will help reduce the pressure on employees when they perform services concerning a delicate situation or a difficult customer. Furthermore, organizations should help employees recharge valuable emotional energy (Lee and Ok, 2012). Emotional breaks should be provided such as inserting a 10-to 15-minute rest time for every 2-hour shift rotation.
Fourth, developing cooperative teams and boosting morale to improve service quality as well as building a sense of teamwork including encouraging support from supervisors may increase positive perceptions and decrease employees' negative emotions . Knowing how other employees feel and learning how to provide emotional support could support cooperative teams in constructive ways. In addition, it could also accomplish positive emotional management (Chu, 2002). Thus, hotel management needs to instruct frontline employees on how to help each other by sharing their feelings and suggesting in how to understand each other. When someone is having a bad day on the job, and whenever someone can perform excellent service quality, they should be encouraged to share their technique. This strategy could not only make colleagues and organizations recognize those employees, but also encourage those employees and others to develop service quality by employing appropriate emotional labour strategies, both personally and in teams.
Fifth, management should maximize the benefits of emotional capital by considering the compensation of employees' resources, and adopt a reasonable wage and welfare structure to help employees compensate for the loss of emotional resources. Compensation and organizational support via effective organizational communication could decrease role stress and job burnout (Yang, 2010). Employee satisfaction does not depend only on the establishment of physical rewards and salaries. The organization should implement tools that could create better communication channels between employees and employers, such as implementing an ongoing comprehensive evaluation feedback system and providing appropriate education and training for employees (Wu, 2011). In order to ensure that service employees continue to act successfully in providing quality services, they need to be not only recognized but also rewarded. Further, supervisors may increase positive perceptions and focus on reactive steps to help employees cope with their negative emotions (Cho et al., 2017;Lam and Chen, 2012). These will help boost the morale of the employees. The organization must always keep in mind that employees that are not happy in their work will lack self-esteem and will ignore the responsibility of providing customer service. In addition, high-value (the ability to provide excellent service) employees also experience a high turnover rate (Chu, 2002). Thus, the organization should develop strategies to retain this kind of employee.
Sixth, the organization could take advantage of the multiple emotional strategy skills of older employees (Hur et al., 2014) by allocating appropriate types of employees to the most suitable work in order to minimize negative and maximize positive emotion (Cho et al., 2013). Further, managers can tailor training programs to suit the employees' unique needs (Cheung and Tang, 2010). Older or senior staff should be encouraged to serve as role models for younger colleagues and give them more opportunities to suggest techniques for service tasks. This strategy will make them realize their value, experience, and efficiency. At the same time, it could make employees realize the value of their work experience and efficiency, promote feelings of self-worth, upgrade the quality of the services, and enhance job satisfaction for them as well.
Lastly, if employees perceive high levels of supervisor support such as the manager's emotional intelligence that acts as a buffer against work stress, they may report lower levels of burnout (Grandey, 2000). The managers' roles involve monitoring, controlling, and directing the emotional labour of their frontline employees in their performance on the job. As a result, management may observe the importance that emotional intelligence plays in the workplace, especially for the managers performing their roles as a part of the social complexity of their organization. Hotels should focus on observing how emotional intelligence, especially that of the manager, forms a crucial part of the social complexity of the organization. Training to develop and support the manager's and employees' emotional intelligence would therefore be an appropriate management initiative. Implementing formal or informal reward programs in order to recognize those managers that exhibit high behavioural emotional intelligence would help to motivate and encourage employees' job satisfaction (Han et al., 2017) and could potentially help achieve a competitive advantage for the organization. In order to increase the emotional intelligence and managerial skills of the managers, organizations should have discussions and conferences (Wolfe and Kim, 2013) that could stimulate using their emotional intelligence. Moreover, organizations should encourage employees to incorporate emotional intelligence into their regular work routines. This could help them demonstrate higher job satisfaction, reduce their burnout, and increase their organizational commitment behaviour . Training programs, e.g., programs to develop emotional support, to handle employee complaints and to provide feedback and constructive opinions should be integrated in managers' training programs (Chu, 2002).
Human resource management should give importance to the emotions and emotional intelligence of their managers and employees to increase job satisfaction (Çekmecelioğlu et al., 2012). Employees may obtain performance feedback through supervision, co-workers or in performing the job itself and realize that the ability to perceive and manage emotions is strongly associated with the effective execution of their tasks. Consequently, workers may incorporate emotional intelligence in their regular work routines.

Limitations and future research suggestions
This study has some limitations related to the sampling method, such as the restraint of the data collected from the subjects working in the hotel. The research needed to request permission from the hotel before approaching employees that were asked by general managers to fill out the questionnaires. Some hotel employees found it inconvenient to access and cooperate with the researchers. The inconvenient sampling method in this study made it very difficult to maintain a consistent manner when collecting the data as should have been.
In addition, the subjects were working in various types of hotels with different emphases on service quality. Hotels with different ratings have different emphases on service quality and emotional labour, and these different demands will affect employees' perceptions of emotional labour consequences. In future research study, the emotional labour research model used this study should be applied to groups of hotels specified by star ratings, e.g., luxury hotels or four-and five-star hotels. This study's model also could be applied to be used in the other service industries, including hospitality educational institutions.
The model used in the study was analysed using the basic and advanced methods of structural equation modelling. This study used the developed questionnaires as a quantitative method to collect the data from hotel frontline employees. The next study should endeavour to contribute more regarding the strategy of emotional labour. Focus groups and in-depth interviews involving a qualitative study could be used to collect more in-depth data to collect more specific details regarding the knowledge of factors on individuals, emotional labour strategies, and job outcomes.
Another limitation was that in this study, the outcomes of emotional labour of hotel frontline employees were rated by themselves and these ratings are subject to recall bias. Future research should also consider interviewing the hotel customers and managers of the service organization to evaluate them using this study's model.
Lastly, this study investigated the indirect effects of emotional labour (surface acting and deep acting) via burnout and job satisfaction concerning turnover intention and organizational commitment. Thus, one interesting suggestion for future research would be to explore the mediating roles of burnout and job satisfaction regarding the effects of emotional labour on job and organizational outcomes.