Relationship between Emotional Labor and Burnout among Sports Coaches in South Korea: Moderating Role of Social Support

: The purpose of this study was to examine the moderating effect of social support in the relationship between three strategies of emotional labor (surface acting, deep acting, and genuine expression) and burnout among sports coaches in South Korea. Data were collected from 259 athletic coaches who were working at secondary schools at Gyeongsangnam-do in South Korea. Results indicated that three emotional labor strategies explained a signiﬁcant amount of variance in each of the three dimensions of burnout (emotional exhaustion, depersonalization, and reduced personal accomplishment). In regard to the moderating effect, social support signiﬁcantly moderated the relationship between surface acting and overall burnout. Speciﬁcally, the positive effect of surface acting on burnout was weaker for coaches with a high level of social support than those with a low level of social support. Some ways to engage in appropriate emotional labor strategies and to create environments where social support can be easily accessed were discussed.


Introduction
It is not surprising that sports coaching is a very stressful job with numerous emotional demands [1,2], as it involves high levels of direct and indirect interaction with athletes, as well as administrators, peers, referees, and parents. Sports coaches also need to perform multiple roles such as those of an instructor, mentor, friend, educator, counsellor, and even parental substitute while interacting with others. However, the high volume of interaction in performing multiple roles evokes considerable emotions in coaches, which they need to control in order to express proper responses to enable their organizational performance. Despite the fact that sports coaching is likely to be exposed to emotional labor due to the high demand of emotion, previous research on emotional labor has mostly been performed in the context of service industry employees. This is because such employees are required by organizations to control their emotions to meet the demands of customers in order to provide better service quality [3]. However, it is important to note that the nature of service provided by sports coaches is quite different from that of conventional service employees who encounter consumers [1]. Lee et al. (2015) [1] suggested that there are several differences in the nature of services between sports coaches and conventional service employees. First, sports coaches provide human services that change one's behavior, attitude, attributes, and social status for one's well-being [4]. In other words, while the ultimate goal of conventional service providers is to generate profit-based output through the provision of services to customers, that of the coach is to change and develop human beings (i.e., athletes). For instance, coaches may need to advise and help athletes with non-athletic issues (e.g., loss of friends, role conflicts, and depression) in addition to working on their athletic performance and skills. In addition, employees in public service occupations require a higher level of emotional labor than those in a profit-oriented context due to the very nature of the job [5,6]. As a public servant, thus, sports coaches are more likely to be exposed to emotional labor.
Second, sports coaches must interact with several other groups, including athletes, parents of athletes, peer coaches, referees, fans, and even the media [1]. This means that the coaches may need to express varying emotions to different groups in order to pursue and balance their main aim of developing the athlete. By contrast, conventional service employees focus primarily on providing a friendly and efficient facilitation service. Such differences between coaches and conventional service employees make coaches more likely to be exposed to emotional labor than their counterparts in conventional service industries.
Third, a longer period of interaction with people leads to a high level of emotional labor [7]. Sports coaches usually interact with key stakeholders (e.g., athletes, parents, administrators, and fans) for relatively long periods of time compared with conventional service workers. For instance, the coaches must deal with key stakeholders both in-season and off-season. Typically, the combination of in-season with off-season is usually two thirds of the year. Thus, coaches may need to spend more time and energy in emotional labor than their counterpart.
Because of the unique nature of the coach's working environment, coaching is well known for generating numerous emotional demands. These emotional demands lead coaches to engage in emotional labor, which is defined as a person's effort to manage unpleasant emotions and to express appropriate emotions for the job [8]. However, emotional labor is potentially harmful to individuals' well-being [9] resulting in job burnout, job dissatisfaction, and turnover intention [10,11].
According to the literature review, burnout among sports coaches has been well documented and has long been a topic of interest in a sports context. Work overload, role conflict, long working hours, perceived lower support, high expectations from others, lack of rewards, and excessive investment of mental and emotional energy are all causes of burnout among coaches [12]. Of the various causes of burnout among coaches, emotional labor is one of the most cited influencing factors [13][14][15].
With regard to the relationship between emotional labor and burnout, many scholars have found that emotional labor causes burnout in employees [3,[16][17][18]. Likewise, there is a growing body of research on the relationship between the two in a sports-coaching context [1,[19][20][21]. In particular, Lee et al. (2015) [1] proposed a comprehensive conceptual framework in which the psychological costs of three types of emotional labor strategies (surface acting, deep acting, and genuine expression) affect job burnout among sports coaches. Although a number of research studies on the relationship between emotional labor and burnout have been carried out, most have focused on the direct relationship between the two. Little work has been conducted to examine moderators on the effect of emotional labor on burnout in the sports-coaching context. In other words, not all coaches who experience emotional labor would burnout; therefore, there appears to be a need to explore the conditions under which sports coaches' emotional labor increases or decreases job burnout.
Kim, Hur, Moon, and Jun (2017) [20] suggested that "the situation in which employees work can function as a moderator is the link between emotional labor and employee outcomes" (p. 125). In the emotional labor literature, one of the salient identified moderators in the link between the two is the perceived level of social support received by employees from their organizations. Workplace social support comes from coworkers, supervisors, and the organization itself [21]. It is clear that the effect of emotional labor on burnout among sports coaches is mitigated if they receive social support from one of the three sources in the workplace. Therefore, the purpose of this study was to investigate the moderating effect of social support in the relationship between emotional labor and burnout among sports coaches in South Korea.

Emotional Labor
Emotional labor ensues when a service worker is expected to present pre-specified and appropriate emotions in a job situation in compliance with the rules of the organization [22]. The act of displaying socially desired emotions is a form of "emotional labor" because it demands an effort on the part of employees to manage their emotional expression or change their inner feelings in order to facilitate task and service effectiveness [22]. Even though there are many definitions of emotional labor, the common understanding of the concept is that it is the regulation of feelings and expressions at work [7,16,22].
In order to comply with emotional requirements at work, employees need to use emotional regulation strategies [7,16,23]. There are three different types of emotional regulation strategies: surface acting, deep acting, and genuine expression [23,24]. Surface acting is the management of one's outer emotional expressions, such as smiling, without any changes in inner feelings during client encounters. For example, sports coaches suppress their anger toward athletes' wrong behaviors and feign the emotions required by the organization. Thus, surface acting is a response-focused emotional regulation in which employees only alter the outward expression of their emotions through suppressing or faking emotional displays [25]. Deep acting requires the modification of an employee's inner feelings to express organizationally desired emotions. Deep acting appears authentic to the employee since it requires putting themselves in another's place [24]. For instance, sports coaches who try to understand and express empathy by thinking that a student-athlete's bad behavior may result from his or her deprived family environment is an example of deep acting. It is referred to as an antecedent-focused emotional regulation strategy, in which employees attempt to modify how they perceive or interpret a situation in order to adjust their emotional response [16,26]. The last emotional regulation strategy is genuine expression. Genuine expression occurs when employees experience appropriate emotions and express them spontaneously and automatically [24]. For instance, a sports coach may sincerely feel sympathetic toward a student-athlete who has a deprived family environment and express the emotion. It is important to note that genuine expression is a strategy in which the naturally expressed emotions correspond to the emotional requirements levied by organizations.

Burnout as a Consequence of Emotional Labor
Even though burnout has long been a topic of interest in the field of sports science and sports coaches are exposed to various stressors such as long-lasting working hours, role conflict, work overload, and excessive expenditure of mental and emotional energy, few studies have examined the effect of emotional labor on job burnout in a sports context [1].
One of the most widely known outcome variables of emotional labor is job burnout. Burnout is defined as a negative response to stress and is viewed as a psychological syndrome occurring among individuals who do 'people work' [27]. Burnout has three components: emotional exhaustion, depersonalization, and a reduced sense of personal accomplishment [28]. Emotional exhaustion refers to the perceived depletion of energy at work [27,28]. Depersonalization is characterized by negative, insensible, or emotionally detached responses to other people working in the organization. Lastly, a reduced sense of personal accomplishment is defined as discounting successful work achievements and increased feelings of incompetence.
According to the meta-analysis research conducted by Hülsheger and Shewe (2011) [29], different dimensions of emotional labor strategies influence burnout in different ways. Recently, Lee and his colleagues (2015) [1] proposed a conceptual framework of emotional labor in sports coaching, where the three types of emotional labor strategies (surface acting, deep acting, and genuine expression) would differently influence job burnout. Surface acting generally creates emotional dissonance between inner and outward feelings that drain one's emotional resources over time [20]. Thus, when coaches engage in an emotional regulatory process that involves faking and suppressing true emotions such as surface acting, they are more likely to experience burnout. Some recent studies have empirically found that surface acting was positively related to job burnout for football coaches [19] and physical education teachers [13].
Deep acting refers to the process of actually trying to alter one's internally felt emotions in order to meet the required emotional expressions of the organization [22]. Although deep acting generates a lower level of emotional dissonance between inner and outer feelings than surface acting, it still involves some expenditure of effort to alter natural feelings, thoughts, and expressions and requires a certain amount of deliberate cognitive struggle in order to control the emotional experience [16]. These efforts cause a depletion of emotional resources. Thus, Lee et al. (2015) [1] proposed that coaches' deep acting positively influences job burnout.
As genuine expression is an unconscious and natural way of expressing emotions, it requires relatively low levels of mental and psychological effort compared to surface acting and deep acting. Genuine expression barely generates emotional dissonance between expressed emotions and felt emotions. Moreover, empirical evidence has shown that genuine expression is negatively associated with job burnout in a sports context [13,19]. Taken together, this study proposes the following hypotheses: Hypothesis 1. Surface acting will be positively related to burnout. Hypothesis 3. Genuine expression will be negatively related to burnout. Hypothesis 3.1. Genuine expression will be negatively related to emotional exhaustion.

Hypothesis 3.2. Genuine expression will be negatively related to depersonalization.
Hypothesis 3.3. Genuine expression will be negatively related to a reduced sense of personal accomplishment.

Moderating Role of Social Support
Social support, from sources such as supervisors, coworkers, and the organization, is generally defined and measured as "the availability of helping relationships and the quality of those relationships" [30] (p. 5). Two models have been suggested to explain the beneficial effects of social support [31,32]. The first is that social support benefits people by decreasing strain regardless of the stress experienced [32]. The other model views social support as an interaction effect; it interacts with a stressor to affect felt strain. This interaction effect is generally referred to as a buffering effect, where "buffering" means protection from the harmful effect of stressful events [31].
It is well documented that social support is a factor that enables people to cope with stress at work [33,34]. People receiving social support from their organizations, coworkers, or supervisors are more likely to access help such as instrumental assistance and esteem support [31]. In a supportive environment, employees suffering a stressful experience may perceive it as less stressful, which may then attenuate the stressful experience of strain [18]. By contrast, employees who do not have social support at work can suffer more severely from the harmful effects of the stressful experience.
Since emotional labor has been identified as a critical stressor for employees, some scholars have begun to examine the moderating effect of perceived social support in the context of emotional labor [20,[35][36][37][38]. For instance, Chen et al. (2012) [35] found that social support diminishes the positive relationship between surface acting and burnout. Likewise, because deep acting entails a certain amount of deliberate cognitive effort in order to control emotional experience [16], it is also possible that social support weakens the positive relationship between deep acting and burnout. As mentioned earlier, unlike the other two forms of emotional strategies, genuine expression barely generates emotional dissonance between expressed emotions and felt emotions; therefore, genuine expression was negatively associated with job burnout [13,19]. Thus, social support strengthens the negative relationship between genuine expression and burnout.
Despite evidence on the moderating effect of social support in the relationship between emotional labor and burnout, little academic attention has been paid to the sport context. Considering the moderating effect of social support, which mitigates the negative consequences of emotional labor in a sports context, suggested in the literature review, the following additional hypotheses were suggested: Hypothesis 4. The relationship between surface acting and overall burnout varies based on the extent of perceived social support, such that the positive effect of surface acting on burnout is weaker for coaches with a higher level of perceived social support than a low level of perceived social support.
Hypothesis 5. The relationship between deep acting and burnout varies based on the extent of perceived social support, such that the positive effect of deep acting on burnout is weaker for coaches with a higher level of perceived social support than a low level of perceived social support. Hypothesis 6. The relationship between genuine expression and overall burnout varies based on the extent of perceived social support, such that the negative effect of genuine expression on burnout is stronger for coaches with a high level of perceived social support than a low level of perceived social support.

Participants and Procedures
The target population in this study was athletic coaches who were currently working at secondary schools in Gyeongsangnam-do province located in the southern area of South Korea. According to the Gyeongsangnam-do Office of Education, there were 505 athletic coaches at secondary schools in 2019. The authors obtained all email addresses of 505 coaches with permission from the Office. However, 61 emails automatically bounced back, resulting in 444 athletic coaches who were eligible to participate in the survey. Of these, 278 coaches completed the questionnaires for a response rate of 59.7%. However, 19 incomplete questionnaires were excluded due to the failure to answer several key survey items. Therefore, a total of 259 questionnaires were included in the final data analysis.
Out of the 259 completed questionnaires, the sample was comprised of 81.5% males (n = 211) and 18.5% females (n = 48) with an average age of 34.37 years (SD = 5.95). At the time of the study, the total number of athletic coaches in secondary schools at Gyeongsangnam-do was 505. The ratio of male to female coaches in the province was about 84.3% (n = 426) to 15.7% (n = 79); hence, the study sample successfully reflected the study population. A majority of respondents had earned a bachelor's degree (n = 208; 80.3%). For types of sports, 54.4% (n = 141) worked at a team-based sport, while 46.6% (n = 118) worked at an individual-based sport.

Instruments
To measure all 42 survey items in the current study, the Teacher Emotional Labour Strategy Scale (TELSS: 11 items), the Maslach Burnout Inventory-Educators Survey (MBI-ES: 22 items), and nine items for social support were employed. It is important to note that all measurement items selected for this study were originally written in English. Thus, the current study used the back-translation method recommended by Brislin (1970) [39] to ensure the quality of translation in Korean.
Emotional Labor Strategies. To measure sports coaches' adoption of varying emotional labor strategies, the current study adopted 11 items from the TELSS developed by Yin (2012) [40]. Specifically, the instrument was comprised of three subscales: surface acting (5 items), deep acting (3 items), and genuine expression (3 items). Scores were recorded on a seven-point Likert scale from one (very strongly disagree) to seven (very strongly agree).
Burnout. There is a burnout scale which is specifically designed for coaches. The scale, named coaching burnout questionnaire (CBQ), is grounded in the athlete burnout questionnaire (ABQ) that was initially tested for validity and reliability in an adolescent athlete population [41]. Some previous studies have used the ABQ to measure coach's burnout based on the modification of the scale [42,43]. However, some reliability issue has been reported [42] so caution should be given in the use of CBQ. Thus, this study has employed the MBI-ES [27] which has been the most widely used instrument to measure burnout levels of individuals working in the school environment including athletics [12]. This instrument is comprised of 22 items divided into three subscales: emotional exhaustion (9 items), depersonalization (5 items), and reduced personal accomplishment (8 items: reverse-coded items). Participants rated each item on a seven-point Likert scale one (never) to seven (every day).
Social Support. To assess perceived social support, nine items were adapted from previous studies [44][45][46]. Specifically, three items were adapted from Rhoades and Eisenberger's [45] work to assess organizational support (e.g., My organization is willing to help me if I need a special favor), three items were adapted from Shanock and Eisenberger's [46] work to measure supervisor support (e.g., My supervisor really cares about my well-being), and three items were adapted from Ducharme et al.'s [44] work to measure coworker support (e.g., My coworkers are very helpful when I encounter difficulties with my work). Scores were recorded on a seven-point Likert scale from one (very strongly disagree) to seven (very strongly agree).

Data Analysis
A confirmatory factor analysis (CFA) was first employed as a methodological tool to test the psychometric properties of the scales. Then, three separate multiple regression analyses were conducted to examine relationships between three dimensions of emotional labor strategies and three dimensions of burnout. Next, the moderating effect of perceived social support on the relationship between emotional labor strategies and the overall burnout was tested with three separate hierarchical regression analyses. With regards to hierarchical regression analysis, we first entered each of the three emotional labor strategies as independent variables into three separate regression analyses. Then, social support as a moderator was entered in the second step at each of the three regression analyses. Lastly, the cross-product term (interaction term) which was created by multiplying the independent variables and the moderator (three dimensions of emotional labor strategies x social support) was entered in the final step of three regression analyses. In the hierarchical regression analyses, independent (three emotional labor strategies) and moderating variables (social support) were centered in order to minimize multicollinearity problems.

Reliability, Validity, and Descriptive Statistics
Reliability and validity tests were conducted to assess the quality of seven measures filled by the study participants (surface acting, deep acting, genuine expression, emotional exhaustion, depersonalization, reduced personal accomplishment, and social support). First, the reliability of the measures was evaluated with Cronbach's alpha coefficients. All Cronbach's alpha coefficients were greater than cut-off value of 0.70, ranging from 0.71 to 0.88 [47].
Confirmatory factor analysis (CFA) was then conducted to evaluate the convergent and discriminant validity of the measures with AMOS 25.0. The results of CFA indicated that the measurement model showed a good fit to the data, as indicated by several fit indices: x 2 /df (2676.828/924) = 2.897, RMSEA = 0.07, CFI = 0.94, TLI = 0.93, and SRMR = 0.05. Given the acceptable model, all factor loadings of the measurement model were significant in the predicted direction with greater than 0.50 (p < 0.01), providing evidence of convergent validity. In addition, all average variance extraction (AVE) values were greater than the cut-off of 0.50 [48]. Discriminant validity among the seven measures was also checked with the method suggested by Fornell and Larcker (1981) [48]. The AVE for each construct was greater than the squared correlations between the focal construct and any others, providing the evidence of discriminant validity. The descriptive statistics, correlations, Cronbach's alpha, squared correlations, and AVEs of the study variables are reported in Table 1.

Relationship between Emotional Labor and Burnout
Multiple regression analyses were performed to test Hypotheses 1, 2, and 3. Table 2 provides a summary of the results for three dimensions of emotional labor strategies predicting three dimensions of burnout. The first regression model was significant (F = 23.861, p < 0.001, R 2 = 0.219), indicating that a significant amount of variance in emotional exhaustion was explained by the three dimensions of emotional labor strategies. More specifically, surface acting (β = 0.191, p < 0.01: H1-1 support) and deep acting (β = 0.307, p < 0.001: H1-2 support) had significant positive impacts on emotional exhaustion, while genuine expression (β = −0.184, p < 0.01: H1-3 support) had a significant negative impact on emotional exhaustion. In regard to depersonalization, the second regression model was also significant (F = 10.370, p < 0.001, R 2 = 0.109), indicating that a significant amount of variance in depersonalization was explained by the three dimensions of emotional labor strategies. Specifically, surface acting had a significant positive impact on depersonalization (β = 0.329, p < 0.001: H2-1 support), while deep acting had rather a significant negative impact on depersonalization (β = −0.139, p < 0.05: H2-2 reject). However, genuine expression was not statistically significant (β = −0.102, p > 0.05: H2-3 reject).
Lastly, as to reduced personal accomplishment, the third regression model was also significant (F = 65.205, p < 0.001, R 2 = 0.434), indicating that a significant amount of variance in reduced personal accomplishment was explained by the three dimensions of emotional labor strategies. Specifically, surface acting (β = 0.329, p < 0.001: H3-1 support) and deep acting (β = 0.262, p < 0.001: H3-2 support) had significant positive impacts on reduced personal accomplishment, while genuine expression (β = −0.364, p < 0.001: H3-3 support) had a significant negative impact on reduced personal accomplishment.

Moderating Effect of Social Support
To examine the moderating effect of social support on the relationship between three dimensions of emotional labor strategies and overall burnout, hierarchical multiple regression analyses were performed. We created three interaction terms using three dimensions of emotional labor strategies and overall burnout and entered each interaction term into Step 3 in Table 3.  First, the interaction effect of social support on the relationship between surface acting and burnout was significant (β = −0.183, p < 0.001). Specifically, beyond the explained variance of surface acting and social support, the surface-acting-social-support interaction (R 2 = 0.322, p < 0.001; ∆R 2 = 0.031, p < 0.01) explained significant incremental portion of variance in burnout. To understand the nature of the significant surface-actingsocial-support interaction, we plotted burnout scores at combinations of the mean ± 1 SD (high and low levels) for surface-acting measures, as recommended by Hayes and Matthes (2009) [49]. The interaction plot (Figure 1) demonstrated that the positive effect of surface acting on burnout was weaker for coaches with a high level of social support than those with a low level of social support (H4 support). However, neither the deep-acting-socialsupport interaction (β = 0.061, p > 0.05; ∆R 2 = 0.004, p > 0.05) and the genuine-expressionsocial-support interaction (β = 0.037, p > 0.05; ∆R 2 = 0.001, p > 0.05) were found significant (H5 and H6 reject). The plots for the two non-significant interactions were revealed in Figure 1.

Discussion
Research on emotional labor has primarily focused on consumer services rather than human services such as those provided by sporting coaches. Sporting coaches tend to interact with numerous stakeholders (e.g., athletes, parents, peers, and referees) and are required to perform multiple roles (e.g., instructors, mentors, educators, and counselors) [1]. These require a high level of emotional labor from the coaches. Thus, emotional labor in sports coaching can negatively affect coaches' personal outcomes with conditions such as job burnout. While there is a growing body of research on emotional labor and its outcomes in a sports context, relatively little attention has been given to sporting coaches [14,15]. Furthermore, existing studies on sporting coaches have only focused on direct relationships involving emotional labor and its outcomes. Little work has been conducted to investigate moderating variables (social support in this study) between coaches' emotional labor and their work outcomes [50]. To fill this research gap, this study attempted to examine the moderating effect of social support on the relationship between emotional labor strategies and job burnout.
The results indicated that surface acting had a positive impact on all three dimensions of burnout among sports coaches. According to the conservation of resources (COR) theory, surface acting is more likely to cause symptoms of burnout such as emotional exhaustion due to the psychological effort required to fake or suppress negative emotions [51,52]. In other words, surface acting is likely to deplete psychological energy, as it involves long-lasting internal tension between one's displayed (suppressed) and true feelings, which, in turn, generates emotional dissonance [53]. This increases the likelihood of burnout. Additionally, surface acting incurs psychological cost by generating feelings of inauthenticity, which is another precursor to burnout [1]. Thus, the continuous experience of emotional dissonance and inauthenticity would result in burnout among sports coaches, which is consistent with previous studies in a sports context [13][14][15]. Based on the findings, we can conclude that surface acting is detrimental to coaches' psychological well-being.
Regarding deep acting, this study found that deep acting was positively associated with emotional exhaustion and a reduced sense of personal accomplishment among coaches, while it had a negative association with depersonalization. According to previous studies, while deep acting does generate less emotional dissonance than surface acting, it still involves some degree of effort to change spontaneous feelings, thoughts, and expressions [1,16], as it is not an automatic process [54]. In other words, deep acting also entails a certain amount of deliberate psychological effort in order to regulate emotional experience. Thus, this emotional effort may lead to job burnout, specifically related to emotional exhaustion and reduced personal accomplishment.
However, some previous studies indicated that deep acting barely generates any emotional dissonance or feelings of inauthenticity, since the inner feelings have been altered before they are expressed [15,16]. In line with this, this study found a negative relationship between deep acting and depersonalization among sports coaches. However, when examining the association between deep acting and overall burnout (a combination of three dimensions), the current study found that deep acting had a positive impact on burnout (β = 0.384, p < 0.001). Consequently, deep acting is as detrimental overall to coaches' psychological well-being as surface acting.
With regard to genuine expression, this study found that genuine expression in sports coaches was negatively associated with the two facets of burnout: emotional exhaustion and reduced personal accomplishment. The results are aligned with those of prior studies [14,15]. Additionally, numerous studies outside of the sports context indicated that as genuine expression involves the spontaneous generation of appropriate feelings and expressions, which means that there is a natural congruence between felt and expressed emotions, it rarely generates emotional dissonance, psychological efforts, or a feeling of inauthenticity [9,16,52], which in turn can generate positive outcomes for psychological well-being (e.g., reduced burnout). Therefore, our findings extended this notion to the sports-coaching context in that genuine expression was negatively associated with coach burnout.
With regard to the moderating effect of perceived social support on the relationships between the three dimensions of emotional labor and overall burnout, perceived social support buffered the positive relationship between surface acting and job burnout.
Specifically, the findings indicated that surface acting had a weaker positive relationship with overall burnout among sports coaches with high levels of social support than their counterparts with lower levels of social support. The findings were consistent with previous work [8,33] that showed perceived social support attenuates (buffers) the harmful effect of surface acting on psychological health outcomes (e.g., burnout).
Unlike surface acting, the current study failed to confirm social support as a significant moderator in the positive link between deep acting and burnout, and in the negative link between genuine expression and burnout. Specifically, the levels of burnout of coaches using the deep-acting strategy could not be reduced, although they receive social support in their workplace. In a similar manner, the level of burnout of coaches using the genuine expression strategy would not be increased although they do not receive social support.

Theoretical Contributions
Although sports coaches are widely exposed to emotional labor due to the high volume of interaction with various persons in performing their multiple roles, there is a lack of research on the examination of emotional labor in the context of sports coaching. Thus, this study makes a significant contribution to the emotional labor literature to fill this gap in a sports-coaching context by examining the relationships between emotional labor, social support, and burnout. First, this study adds to the knowledge base by empirically demonstrating that three types of emotional labor (regulation) strategies had different effects on burnout for sports coaches. Overall, while surface-acting and deepacting strategies were positively associated with burnout, the genuine-expression strategy was negatively associated with burnout.
The most noteworthy theoretical contribution of this study lies in the moderating effect of workplace social support on the relationship between emotional labor and burnout. Although this study did not find the moderating effects of social support in all three facets of emotional labor strategies, social support was found to play a significant role in attenuating the harmful effect of surface acting on burnout. Compared to the other two facets of emotional labor strategies, surface acting, which involves the process of suppressing unpleasant emotions and displaying different emotions, typically generates higher levels of psychological costs (e.g., emotional dissonance and a feeling of inauthenticity) for coaches [55]. Given the detrimental consequences of surface acting, it is important for coaches, particularly those who engage in surface acting, to receive social support at work.
It has been argued that workplace social support is more likely to attenuate the harmful effects of a stressor if the type of support received closely corresponds to the type of stressor [31]. Considering this argument, an examination of the relative impact of different types of workplace social support (e.g., informational, practical, emotional, and appreciative) received from different sources (e.g., colleagues, supervisors, managers, and administrators) in future research would help coaches manage the emotional labor process more effectively.
Next, the sample of the current study is sports coaches at Korean secondary schools. While some research has recently begun to study emotional labor for sports coaches [1,14,15], none of the available studies on emotional labor dealt with coaches in South Korea at the time of writing this paper. Due to the fact that South Korea' s athletic performance in various international sports competitions, including the Olympics, has been outstanding during the last 30 years, South Korea today is internationally regarded as a sporting powerhouse at the level of elite sports [56]. Such outstanding performance could not be possible without athletic coaches at secondary schools who played critical roles in recruiting, managing, and developing athletes and led them to become successful athletes [57]. However, the coaches might have experienced emotional struggles due to numerous inevitable stressors such as excessive interaction with others, long working hours, high expectations from others, emphasis on winning, tight scheduling, lack of rewards, excessive expenditure of mental energy, and among others. These stressors may entail a considerable amount of emotional labor. Therefore, this study contributes to a growing body of knowledge on emotional labor by empirically examining the moderating effect of social support on the relationship between emotional labor strategies and job burnout for sports coaches in South Korea.
As mentioned previously, the current study would make a theoretical contribution to the emotional labor literature by examining relationships among emotional labor, burnout, and social support of sports coaches. However, the current study has not investigated the possible influence of other important factors in the relationships among the three variables. Fox example, the important factors could be the level of coaching, working hours, gender, years of experience, ethnicity, and so on. Therefore, it will be necessary for future research to investigate the difference of relationships among emotional labor, burnout, and social support according to various other factors of coaches.

Practical Implications
The results of this study indicate that while surface-and deep-acting strategies had a mostly positive relationship with coach burnout, genuine expression had a mostly negative relationship with burnout. Given this fact, genuine expression is a better strategy to protect the coach's well-being than the other two strategies. This leads to the conclusion that it is critical for sports coaches to learn how to engage in genuine expression rather than surface acting and deep acting.
Interventions clearly mitigate the negative impact of emotional labor in the sportscoaching profession and help coaches enhance their emotional management skills. Little is known, however, about the type of interventions that might be successful in achieving these aims in the context of sports coaching, but some suggestions can be made. According to the extant literature, a structured intervention program (i.e., training session) is useful for the unconscious emotional regulation process [58]. For instance, after participants in the reappraisal condition were exposed to various stimuli that induce negative emotions, the participants were then encouraged to reinterpret the contents of the stimuli to change their emotional experience. Christou-Champi et al. [58] found that participants in the reappraisal condition had naturally regulated their emotions such as genuine expression after two weeks. This suggests that a structured intervention program can encourage coaches to automatically control for negative emotions and might be beneficial in helping coaches manage emotional labor. Additionally, school administrators/managers should provide regular training programs designed to assist coaches in understanding what emotional labor is, the consequences of emotional labor strategies, and how to engage in genuine expression in a sports-coaching context.
The findings of this study also indicate that social support attenuates the harmful effect of surface acting on coach burnout. Thus, it is important for school administrators to provide a work environment where coaches can easily access social support from coworkers or supervisors. For instance, enhancing coach support networks to increase group cohesion and the provision of emotional mentoring might be useful. Specifically, reciprocal peer coaching, where pairs of sports coaches work together to reflect on practice and provide mutual support, can help coaches develop their knowledge, skills, attitudes, and value. Therefore, peer coaching programs might be appropriate in helping coaches to develop more adaptive emotion management skills, which is beneficial in minimizing the harmful impact of emotional labor on health well-being (i.e., burnout). Particularly, such a peer coaching program is more effective for coaches with the surface-acting strategy than others (i.e., deep acting and genuine expression) as the moderating effect of social support was only found in the former.