THE RELATIONSHIP BETWEEN THE EMOTIONAL LABOR AND BURNOUT LEVELS OF EFL TEACHERS AT UNIVERSITY PREPARATORY SCHOOLS

: This study investigated the relationship between the emotional labor and burnout levels of English as a Foreign Language (EFL) teachers in language preparatory schools at Turkish universities. Data were collected from 179 EFL teachers from both state and foundation universities, using Turkish adaptations of the Emotional Labor Scale covering surface acting, deep acting and naturally felt emotions, and the Maslach Burnout Inventory Educators Survey including emotional exhaustion, depersonalization, and reduced personal accomplishment. The findings revealed that the most commonly experienced dimensions of emotional labor and burnout were naturally felt emotions and emotional exhaustion, respectively. Surface acting showed a positive correlation with emotional exhaustion and depersonalization, whereas displaying genuine emotions had a negative correlation with burnout. Furthermore, surface acting and naturally felt emotions emerged as significant predictors of emotional exhaustion and depersonalization, while displaying natural emotions was a significant predictor of reduced personal accomplishment. These findings suggest that masking emotions fosters exhaustion and alienation from students, while the display of internalized and genuine emotions increases the sense of personal accomplishment and alleviates burnout. Therefore, we suggest that institutions and administrators should support the emotional well-being of EFL teachers by encouraging their display of preferred feelings instead of imposing display rules on them.

Teaching is a multifaceted profession loaded with emotions (Sutton & Wheatley, 2003).
Teachers' emotions are central to their identity and interwoven with their everyday teaching practices (Miller & Gkonou, 2018).The inevitable interactions with students, colleagues, administrators, and parents make schools a place where a huge volume of emotions is constantly exchanged.Within this context, teachers care about students' feelings to increase their motivation in learning to achieve their own academic objectives (Noddings, 1992).Although care in teaching fosters students' success and teachers' job satisfaction, it may lead to teachers' anxiety, emotional strain, disappointment and anger (Nias, 1996).This is due to the fact that caring professions requiring emotional sensitivity also demand emotional labor which obligates teachers to regulate their emotions for professional reasons (Hochschild, 1983).
This challenge becomes more complicated with the neo-liberal policies of today's world that require employees to deploy utmost efforts physically, intellectually, and emotionally to gain more profit (Kart, 2011).Since the late 2000s, there has been a shift in how universities are perceived, with increasing recognition of their roles as service providers.Thus, students and parents are now seen as valued customers, expecting the best service (Berry & Cassidy, 2013;Constanti & Gibbs, 2004).Consequently, universities compete to provide the best educational environment and opportunities to admit and retain more students.This competition was stimulated in Turkish higher education with the establishment of foundation universities after the constitutional changes in 1984.As most state universities governed centrally failed to meet the changing and increasing needs of the young generation with their limited funding, foundation universities attracted students with their growing number of programs and opportunities (Erguvan, 2013).Students select the programs in both state and foundation universities based on the scores they obtain in a competitive nationwide entrance examination in Türkiye.Although only 8.67% of total students at tertiary level (YÖK, n.d.) constitute the student population in foundation universities due to their high tuition fees, these universities raise the quality of education through competition while alleviating the burden on the state.In such a competitive context, teachers at tertiary level take on a big responsibility to better prepare their students for the labor market and to keep the rankings of their institutions high (Swart et al., 2013).This situation may lead to more emotional challenges, which makes emotional labor an important issue to explore as a possible source of teacher burnout within the context of universities, because teachers are vital for universities to achieve their institutional and educational goals.
Having stated that, we aim to investigate EFL teachers' emotional labor and burnout levels in language preparatory schools, both in state and foundation universities.Furthermore, we aim to explore the relationship between the dimensions of emotional labor and burnout to gain a deeper understanding of their correlation at the tertiary level.In Türkiye, the importance attached to the teaching and use of English at a higher education institution holds substantial influence over the university choices of numerous students.The medium of instruction is English in 443 programs offered in a variety of faculties at 40 out of 129 state universities, and in 234 programs at 36 out of 74 foundation universities (YÖK, n.d.).In addition to these, there are numerous programs in which 30% of courses are offered in English.This suggests that language preparatory schools are mandatory for a considerable number of university students.These students are required to be competent users of general and academic English to be able to survive in their departmental courses instructed in English.Therefore, it is crucial to understand English teachers' emotional well-being in university preparatory school settings in order to eliminate any source of discomfort and exhaustion for them so they can provide a high-quality language education to their students.
The term emotional labor was first coined by Hochschild (1983) as displaying or concealing emotions for the sake of organizational goals in service-oriented businesses.Emotional labor manifests itself in two forms.The first one is surface acting which is an emotion control mechanism in which employees suppress undesired emotions and exhibit desired ones so that the organizational goals are fulfilled.The other one is deep acting which is a situation whereby employees adjust their displayed emotions to the expected ones during an interaction with customers by putting themselves into the shoes of these service recipients.Hochschild conceptualized these terms considering jobs requiring a high amount of client-employee interaction during which employees are obliged to behave in a way defined by the rules of the organizations, termed as 'feeling rules'.The main purpose of these implicit or explicit rules is to constrain emotional expressions of employees by dictating them appropriate emotions they should exhibit.This is done with the aim of attaining the desired level of performance (Diefendorff & Gosserand, 2003;Diefendorff et al., 2006).
However, the rules imposed by the workplace may conflict with employees` genuine feelings and in that case, they feel 'estranged or alienated from an aspect of self' because of emotional dissonance (Hochschild, 1983, p. 7).Hochschild's principal argument is that some effort is required during the act of managing emotions.When displayed emotion is modified according to the rules of an organization, even if it contradicts with what the employee really feels, that emotion is exchanged like a product, and turns into a commodity.Such commoditization of personal emotions negatively affects the employee in the long run by eventually leading to job stress, turnover, and burnout.Unlike Hochschild (1983), Ashforth and Humphrey (1993) suggest that complying with the organization's norms or rules does not necessarily bring detrimental effects to employees.They believe these emotions displayed at work are not internalized and therefore not genuine; they are mandated just to facilitate service transaction and quality.For this reason, they suggested replacing Hochschild's term feeling rules with emotional display rules, which dictate employees to act, talk, and get dressed in accordance with organizations` expectations.They later added another strategy which is termed as naturally felt emotions, because service providers do not always act by masking or moderating their emotions through surface acting or deep acting; instead, they may also experience and display the expected emotions genuinely or spontaneously.Performing surface acting or deep acting may cost some amount of effort, usually high in surface acting and low in deep acting, yet naturally felt emotions may cost employees little or no emotional burden as the feeling is natural.Using emotional labor strategies can decrease the emotional stress experienced at work and increase the service quality for the organization.However, over exposure to surface acting can lead to emotional dissonance, which can in turn bring about burnout.
Burnout was first defined by Freudenberger (1974) as the state of exhaustion of an individual's internal resources stemming from loss of power and failure owing to overloading or unmet requests.According to Maslach et al., (2001, p. 397), 'burnout is a prolonged response to chronic emotional and interpersonal stressors on the job'.They proposed three dimensions of burnout: emotional exhaustion, depersonalization, and reduced personal accomplishment.Emotional exhaustion is the state of depleted emotional resources and being incompetent to provide service to others.Its symptoms include extreme and prolonged tiredness, energy loss, and feeling of being emotionally worn-out (e.g., Maslach, 1982;Sweeney & Summers, 2002).The second one is depersonalization, wherein an individual starts harboring negative attitudes towards others, leading to distancing, indifference and a perception of them as objects rather than individuals.The third dimension is reduced personal accomplishment through which individuals feel that they lack competency to carry out their job, and they end up in a state of negative evaluation of themselves (Maslach & Jackson, 1981;Maslach et al., 2001).
Emotional exhaustion, depersonalization and low personal accomplishment are frequently observed in the teaching profession, which demands "a high sense of ideals and a high degree of interaction with other people," and they can impact teachers' capacity to effectively fulfil their roles (Evers et al., 2005, p. 425).Research on teachers' emotions over the decades has well documented the role of some internal and external factors leading to burnout in educational settings.These factors included a range of personal variables, such as teachers' age, teaching experience, level of education, gender, and several others, along with their negative self-oriented cognitive style, involvement in professional development activities, and self-efficacy beliefs.(e.g., Evers et al, 2005;Güneş & Uysal, 2019;Kulavuz-Önal & Tatar, 2017;Lackritz, 2004;Mede, 2009;Sadeghi & Khezrlou, 2014;Sünbül, 2003).Research has also provided evidence of the adverse effects of factors related to students, such as inattentiveness and disrespect (Friedman, 1995), low motivation, and misbehaviour (e.g., Güneş & Uysal, 2019;Shunying, 2015).Additionally, organizational factors like teachers' heavy workload, working conditions, the number of students taught, time invested in various activities, limited resources, inadequate administrative and collegial support, a lack of in-service training, insufficient appreciation, delayed rewards, and low salary have been shown to contribute to teachers' burnout (e.g.Güneş & Uysal, 2019;Lackritz, 2004;Shunying, 2015;Yastıbaş, 2021).While existing literature has identified various factors influencing teachers' emotional well-being, its relationship to emotional labor remains to be discovered in many educational settings.
The previous literature provides evidence that emotional labor can have both positive and negative effects on burnout depending on the dimension which is more prominent (Hochschild, 1983;Zhang & Zhu, 2008;Kinman et al., 2011).Displaying surface acting is usually associated with burnout, especially emotional exhaustion, because of the emotional dissonance caused by suppressed feelings (Mackenzie, 2012).Deep acting and naturally felt emotions, however, are usually negatively correlated with the dimensions of burnout and can mitigate the negative effects of burnout (e.g., Basım et al., 2013;Yao et al., 2015;Yılmaz et al., 2015;Youngmi, 2016).
Investigating discipline-specific emotional labor that confronts teachers in different contexts is necessary.This exploration is essential to understand their practices and develop policies that can offer professional and personal support to them (Loh & Liew, 2016).Studies that explored the challenges specific to language teaching have revealed that English language teachers are subject to stressors that aggravate their feelings of exhaustion.These stressors include teaching in an international context, constant interaction with students as required by certain methods, uncertain working conditions, cultural differences between teachers and students, and feeling insecure about their jobs, especially at private institutions (King, 2016;Mercer et al., 2016).Besides, not all non-native English language teachers are communicatively competent enough to use English in class, which increases teachers' anxiety (Horwitz, 1996;Talbot & Mercer, 2018).Furthermore, certain stress-inducing factors are specific to the context.Loh and Liew (2016) demonstrated how, in the exam-oriented environment of Singapore, the conflict between English language teachers' pedagogical beliefs and the academic expectations of institutions and parents could give rise to feelings of dissatisfaction, disappointment, and guilt, when their teaching methods did not produce the expected test scores.Value-laden and subjective nature of grading students' essays and creative writings, coupled with the weight of providing suitable feedback, made teachers emotionally vulnerable in this setting as well.Similarly, Benesch (2020) showed that the time dedicated to reading and offering feedback on student writings, along with determining the appropriate type of feedback to provide, are factors that contribute to emotional labor for language teachers in a US university setting.
Despite the growing body of literature addressing the challenges of language teaching, the exploration of how language teachers handle their emotions in response to these challenges remains an underexplored issue across all school levels.Kim and Kim (2018) investigated how three Korean teachers displayed their emotions while teaching EFL to young learners in private and public elementary schools.These teachers reported their tendency to suppress negative emotions caused by young learners' distracted behaviours.They felt exhausted especially when there is a conflict between their actual feelings and the institutional expectations.Taylor (2020) found that twelve non-Japanese EFL teachers in a language school in Japan were consciously engaged in surface acting on a daily basis as they believed surface acting was necessary to facilitate teaching and meet student expectations.In a study conducted with 164 English language teachers working at 10 universities in China, Zhang and Zhu (2008) found that Chinese teachers mostly displayed deep acting and genuine emotions.The study revealed a positive correlation between surface acting and all burnout dimensions; in contrast, naturally felt emotions and deep acting were negatively correlated with emotional exhaustion, depersonalization, and reduced personal accomplishment.The results supported the argument that while surface acting increases burnout, deep acting and naturally felt emotions strategies can decrease the level of burnout.
In Türkiye, the relationship between burnout and emotional labor was investigated at a variety of school levels, but not in the field of English language teaching.At primary school level, Akın et al. (2014) explored the emotional labor levels of 370 primary school teachers working in private and state schools located in the capital city.The findings revealed that the participants mostly displayed their genuine emotions.Teachers working at private primary schools displayed deep acting and genuine emotions more frequently than those working at state schools, and emotional labor was found to be a significant predictor of the participants' burnout levels.In a study conducted with 160 primary school teachers, 217 secondary school teachers, and 162 high school teachers, Cingöz (2018) found that surface acting was a significant predictor of depersonalization and personal accomplishment; deep acting was a significant predictor of personal accomplishment; and displaying genuine feelings was a significant predictor of emotional exhaustion, depersonalization and personal achievement.One of the few studies conducted at the higher education level was undertaken by Özbingöl (2013) with 310 academics working at foundation and state universities.The study revealed that surface acting was positively correlated with burnout, while deep acting was negatively correlated.Özbingöl also showed that the harmful effects of surface acting on burnout diminished when academics who masked their true feelings were provided with supervisory support.
Although emotional labor was investigated within the field of education and other sectors to some extent, it is still a new conceptual direction in language teaching as pointed out by King (2016), and its relation to English language teachers' burnout has been an issue that needs to be explored thoroughly.Furthermore, to the best of our knowledge, no study had been conducted with EFL teachers at the preparatory schools of Turkish universities despite the importance and challenges of preparing university students to study their own major in a foreign language.Therefore, the current study fills a significant gap by investigating the emotional labor and burnout levels of an underexplored group of participants at an underexplored setting by addressing the following research questions: 1 What are the emotional labor levels of EFL teachers working at the preparatory schools of Turkish universities?
2 What are the burnout levels of EFL teachers working at the preparatory schools of Turkish universities?
3 a) Is there a correlation between the dimensions of emotional labor and the burnout levels of teachers?b) How do the dimensions of emotional labor affect the dimensions of burnout?

METHOD
This study adopts a quantitative approach to investigate the emotional labor and burnout levels of EFL teachers.It uses a correlational survey design to explore the relations between the dimensions of emotional labor and those of burnout (Creswell, 2012).

Participants and Setting
The study was conducted with EFL teachers who responded positively to the call for participation to this study sent via-email to universities with a preparatory school in Istanbul.One hundred seventy-nine (179) full-time EFL teachers from five universities that allowed research within their settings gave consent to become the participants of the study.Sixty-seven percent (n=120) of these participants worked at the preparatory schools of two foundation universities, whereas 33% (n=59) were employed at the preparatory schools of three state universities located in Istanbul, Türkiye.Participants of the study worked 28 hours a week at foundation universities and an average of 21 hours a week at state universities.
With regard to the features of the participants, 78.8% (n =141) of the teachers were female and 21% (n = 38) were male.The teachers ranged in age from 24 and 67 years old, and in teaching experience from 1 to 26 years.Their teaching experience within the current institution ranged from 1 to 20 years.Besides, while the majority of the teachers, 92% (n=165), had a BA degree in English language teaching and English/American literature, 69.3% (n =124) of teachers held an MA and Ph.D. degrees within the same fields of studies.
As for the setting, the medium of instruction was English in all programs at the foundation universities and one of the state universities where the participants were employed.The other two state universities, however, had some programs in which either all or 30% of the courses were offered in English.Therefore, the preparatory school was mandatory for the students of these universities and programs unless they received a passing score on the proficiency exam set by the institution to determine their language proficiency level when they first arrived at the university.While those with a passing score could directly get enrolled in their departmental courses, the other students were placed in either one of beginner, intermediate, or advance level programs at the preparatory schools depending on their proficiency exam scores.The students were obliged to attend language classes in these levels for two semesters, with the option of taking the proficiency test again after the first semester to move on to an upper level or to their own programs.Within the setting of these universities, the students took courses on language skills including reading, writing, speaking, listening, and grammar for 24 to 25 hours a week at foundation universities, and 20 to 24 hours a week at state universities.

Data Collection
Data were collected during the Spring semester of 2019, before the COVID-19 pandemic struck, through two instruments: The Emotional Labor Scale (ELS)-Turkish version by Basım and Beğenirbaş (2012), and the Maslach Burnout Inventory Educators Survey (MBI-ES)-Turkish version by Girgin (1995).The former was originally developed by Diefendorff et al. (2005), and the latter by Maslach et al. (1986).Before the data collection, the participants were asked to sign the consent form that informed them of the purpose and possible benefits of the study.They also filled in a background questionnaire that included questions to gather demographic information on their age, gender, educational background, the type of institution in which they work, and their years of teaching experience in total and at the current institution.Two instruments, namely ELS and MBI-ES, are described below.

The Emotional Labor Scale (ELS)
The original emotional labor scale (Diefendorff et al., 2005) consists of three subscales with 14 items: Surface acting (7 items; e.g.item 2-I fake a good mood when interacting with students.),deep acting (4 items; e.g.item 7-I try to actually experience the emotions that I must show to students), and naturally felt emotions (3 items; e.g.item 11-The emotions I express to students are genuine).These statements are placed on a Likert scale with 5 points ranging from 'never' (point 1) to 'always' (point 5).The scores of each subscale are calculated and three separate scores are found for each participant instead of a total score.Higher scores for each subscale indicate higher emotional labor level for that subscale.For the construct validation of the Turkish version of the scale used to collect the data in this study, Basım and Beğenirbaş (2012) carried out explanatory and confirmatory factor analyses by administering the instrument to two different sample groups of teachers working at secondary schools.Following the explanatory factor analysis, they removed one item from the surface acting dimension, and grouped the remaining thirteen items under the constructs of surface acting, deep acting and naturally felt emotions in accordance with the original instrument.Hence the lowest and the highest possible scores that can be obtained on the surface acting, deep acting and naturally felt emotions subscales of the instrument are 6-30, 4-20, and 3-15 respectively.Factor loading values they found were between.53and.81 for surface acting, .72 and .88 for deep acting, and .82and .89for naturally felt emotions.For the current study, confirmatory factor analysis we conducted revealed factor loadings between .68 and .78for surface acting, .72 and .73 for deep acting, and .63 and .67 for naturally felt emotions.
Regarding the internal consistency of the scale, Cronbach's alpha values of the subscales varied between .83 and .88 for the first sample, and .84 and .86 for the second sample.Cronbach's alpha value for the total scale was found to be .80for both samples.Internal consistency reliability of the scale was also tested for the current study, and the Cronbach alpha values were found to be .92for surface acting, .91 for deep acting and .88 for naturally felt emotions.

The Maslach Burnout Inventory -Educators Survey (MBI-ES)
The MBI-ES developed by Maslach et al. (1986) based on the original Maslach Burnout Inventory (MBI) (Maslach & Jackson, 1981) consists of 22 items with three subscales: Emotional exhaustion (9 items; e.g.item 8-I feel my job wears me out), depersonalization (5 items; e.g.item 5-I feel I treat some of my students as if they were inhumane objects), and personal accomplishment (8 items; e.g.item 19-I have done many valuable things in my job).The MBI-ES is not different from the MBI except the replacement of the word 'recipients' with 'students' in some statements since students are considered service recipients of teachers.The Turkish version of MBI-ES used in the current study by Girgin (1995) also includes 22 items representing three subscales placed on a 6-point Likert scale, ranging from never (point 0) to always (point 6).Higher scores in the subscales of emotional exhaustion and depersonalization denote higher level of burnout while higher scores in the subscale of personal accomplishment reveal lower degree of burnout.The items of this latter dimension are positively worded; therefore, in the current study, the items were reversely coded, and the term reduced personal accomplishment was preferred for the convenience in interpretation and reporting.The scores of each subscale are calculated and three separate burnout scores are found for each participant instead of a total score (Maslach & Jackson, 1981).The lowest and the highest possible scores that can be obtained on the emotional exhaustion, depersonalization, and reduced personal accomplishment subscales are 0-54, 0-30, and 0-48, respectively with cut-off points denoting low, moderate, and high levels of burnout (i.e., for emotional exhaustion, scores between 0-16 shows low, 17-26 moderate, and 27-54 high; for depersonalization, scores between 0-8 shows low, 9-13 moderate, and 14-30 high; and for reduced personal accomplishment, the scores between 0-30 shows low, 31-36 moderate, and 37-48 high level of burnout).The internal consistency of the scale was checked by Girgin (1995), who reported Cronbach's alpha values of .87,.63,and .74for the emotional exhaustion, depersonalization, and reduced personal accomplishment subscales, respectively, in the adapted version of MBI-ES.In the current study, the reliability of the scale was tested for the data collected and Cronbach's alpha values of .91,.78,and .80 were obtained for emotional exhaustion, depersonalization and reduced personal accomplishment subscales, respectively.
The confirmatory factor analysis of MBI-ES was also conducted in the current study.Factor loadings we found were between .63 and .76 for the emotional exhaustion subscale, .60 and .71for depersonalization subscale, and .61and .74for the reduced personal accomplishment subscale.

Data Analysis
Before the data were statistically analyzed using SPSS version 22 on Windows, they were checked for normality.As the kurtosis and skewness values for each subscale were between +2.0 / -2.0, parametric tests were used to analyze the normally distributed data of the study (Tabachnick & Fidell, 2013).To measure the emotional labor and burnout levels of participants, descriptive statistics were performed.To explore the relationship between emotional labor and burnout along with their dimensions, Pearson correlation analysis was conducted.The multiple regression analyses were also carried out to see if emotional labor and its dimensions could be considered as predictors of burnout.

FINDINGS AND DISCUSSION
This section presents the findings of the study in relation to EFL teachers' emotional labor and burnout levels, and how the subscales of emotional labor and burnout correlate.

The Emotional Labor Levels of EFL Teachers
In order to answer the first research question concerning the emotional labor levels of EFL teachers working at the preparatory schools of Turkish universities, the means and the standard deviations of the scores obtained through the ELS were calculated for each dimension as shown in Table 1 below Table 1 shows that the participants experienced naturally felt emotions (M = 12.05, SD = 2.56) more frequently than deep acting (M = 11.14, SD = 4.70) and surface acting (M = 15.35,SD = 6.29).In other words, EFL teachers working in university preparatory schools mostly displayed genuine emotions, and they resorted to deep acting and surface acting strategies less often.

The Burnout Levels of EFL Teachers
To answer the second research question related to the burnout levels of EFL teachers working in the preparatory schools of Turkish universities, the scores obtained through MBI-ES were calculated by considering the cut off points given in the method section.Tables 2 and 3 show the descriptive statistics of the burnout scores, and the number and percentages of participants who obtained low, moderate and high scores on each dimension of burnout.As shown in Tables 2 and 3, out of 179 participants, 65 (36.3%) and 61 (34.1%) received a score revealing a moderate and high level of emotional exhaustion (M = 23.28,SD = 11.37)respectively.105 (59%) received a score showing a low level of depersonalization, while 46 (26%) and 28 (16%) received a score showing their moderate and high levels of depersonalization (M = 6.67,SD = 5.55) respectively.Almost all of the participants (n = 178) obtained a score showing their low level of reduced personal accomplishment (M = 12.63, SD = 5.97).Among these, emotional exhaustion was the most prevalent dimension of burnout experienced at both moderate and high levels.

The Relation Between Emotional Labor and Burnout
To answer the third research question of the study, which explores the possible correlation between the dimensions of emotional labor and the burnout levels of teachers, we used the Pearson correlation test.The results were shown in Table 4 below.As shown in Table 4, all the dimensions of burnout were positively correlated with surface acting [emotional exhaustion, r(177) =.463, p <.01; depersonalization, r(177) =.435, p <.01; reduced personal accomplishment, r(177) =.267, p <.01].On the other hand, there was a negative correlation between naturally felt emotions and all the dimensions of burnout [emotional exhaustion, r(177) = -.469,p <.01; depersonalization, r(177) = -.424,p <.01; reduced personal accomplishment, r(177) = -.454,p <.01].In other words, as surface acting level increased, the level of emotional exhaustion, depersonalization, and reduced personal accomplishment increased as well, yet a decrease in naturally felt emotions caused an increase in the levels of the three dimensions of burnout.However, there was no statistically significant relationship between the dimensions of burnout and deep acting (p > 0.05).
Finally, to be able to answer the last research question regarding how the dimensions of emotional labor affect the dimensions of burnout, multiple regression analyses were conducted.The values pertaining to the results of the analyses are displayed in Tables 5, 6, and 7.According to the values presented in Table 5, surface acting [B =.30, t(175) = 3.02, p <.001], and naturally felt emotions [B = -.28,t(175) = -2.98,p <.001] were found to have an effect on emotional exhaustion.In other words, there was a positive relationship between surface acting and emotional exhaustion; yet the relationship between naturally felt emotions and emotional exhaustion was negative.On the other hand, deep acting did not have an effect on emotional exhaustion.Emotional labor explains a significant proportion of variance in emotional exhaustion [R 2 =.25, F(3, 175) = 20.39,p <.001].The results indicated that EFL teachers' surface acting and naturally felt emotions predicted the level of emotional exhaustion, whereas deep acting did not.The results presented in Table 6 indicate that surface acting [B =.28, t(175) = 2.78, p <.001], and naturally felt emotions [B = -.23,t(175) = -2.38,p <.001] affected depersonalization.Surface acting exhibited a positive correlation with depersonalization, whereas naturally felt emotions demonstrated a negative association with depersonalization.Deep acting, on the other hand, was observed to have no impact on depersonalization.Emotional labor also explained a significant proportion of variance in depersonalization [R 2 =.20, F(3, 175) = 16.03,p <.001].While EFL teachers' surface acting and naturally felt emotions affected the level of depersonalization, deep acting was found to have no effect on it.As the results indicate, naturally felt emotions [B = -.54,t(175) = -5.60,p <.001] affected reduced personal accomplishment negatively.However, deep acting and surface acting did not have an effect on reduced personal accomplishment.Emotional labor also explained a significant proportion of variance in reduced personal accomplishment [R 2 =.20, F(3, 175) = 16.15,p <.001].Naturally felt emotions levels of EFL teachers affected the reduced personal accomplishment; however, deep acting and surface acting were found to have no effect on it.

Discussion
In the present study, we explore the emotional labor and burnout experienced by EFL teachers working within preparatory school settings at state and foundation universities in Türkiye.One noteworthy finding of our study is that the preparatory school teachers primarily express genuine emotions in their interactions with their students, with a preference for deep and surface acting being less prominent.This suggests that Turkish EFL teachers are not merely playing a role to mask their authentic feelings towards their students at the tertiary level.This finding concurs with the findings of previously conducted studies with teachers across different educational levels in Türkiye, such as those by Akın et al. (2014), Erken (2018), Hoşgörür & Yorulmaz (2015), and Yılmaz et al. (2015), which revealed that Turkish teachers tend to predominantly display genuine emotions in their relations with students, while surface-level emotions are exhibited to a lesser degree.
While the pattern of displaying true emotions could suggest that the educational environment in our context promotes sincere teacher-student relationships, potentially leading to positive learning experiences for students (King, 2016), further research is required across various educational levels, particularly within EFL contexts, to draw such conclusions definitively.Similarly, the finding that our participants engage less in surface acting also aligns with the findings of studies conducted by Li and Liu (2021) and Yin et al. (2017).While Li and Liu (2021) focused on EFL teachers in Chinese secondary schools and Yin et al. (2017) in primary schools in Hong Kong, Li and Liu (2021)'s explanation that the extensive time spent with students allows teachers more room to authentically express their emotions may also explain why our participants, who teach 21-28 hours weekly, do not resort to surface acting.The substantial amount of time they spend with their students can make it difficult to fake their true emotions, whether positive or negative.
Another finding of the study is that the majority of EFL teachers felt emotional exhaustion at moderate and high levels.This study lacks a comparative perspective on burnout in terms of personal and organizational variables.However, the working conditions in participating universities might have offered teachers a variety of reasons to feel emotionally distressed.In the current study, EFL teachers' extended working hours in language preparatory schools in foundation universities could have left them feeling drained within their institutions.Besides, given that the contracts are renewed annually based on performance, job security might be a potential source of distress for teachers working in these universities (Erguvan, 2013).The instructors in state universities, however, are centrally assigned and enjoy a lifetime job security, a guarantee that remains in effect regardless of their performance or any problematic relations with students (Akın et al., 2014).
On the other hand, one of the public universities involved in the present study admits students with top nationwide scores and provides a competitive environment for both students and teaching staff.In the remaining two universities, teachers deal with larger class sizes compared to their colleagues in the other participating universities.Besides, research has shown that factors such as student behaviours, teachers' participation in professional development activities, administrative and social support, work-related stress, salary and job satisfaction contribute to teachers' burnout in the Turkish EFL language preparatory school context (Kılavuz-Öner & Tatar, 2017;Yastıbas, 2021).However, since the study does not incorporate qualitative data, the specific reasons for their exhaustion remain speculative.
The study also provides valuable insights into the relationship between the emotional labor and burnout levels of EFL teachers in language preparatory schools at the tertiary level.The results show that surface acting is positively correlated to two dimensions of burnout, namely emotional exhaustion and depersonalization, corroborating the findings of previous studies (Akın et al., 2014;Brotheridge & Grandey, 2002;Grandey, 2003;Näring et al., 2006;Zhang & Zhu, 2008).This implies that an increase in pretense corresponds to higher levels of emotional exhaustion and depersonalization among EFL teachers, because surface acting triggers the "emotive dissonance and alienation of emotions from the self" (Tsang et al., 2021, p.13), ultimately contributing to burnout.
On the other hand, the study identifies a negative correlation between naturally felt emotions and all dimensions of burnout.Both surface acting and naturally felt emotions are significant predictors of emotional exhaustion and depersonalization, confirming the findings of Yılmaz et al. (2015).Furthermore, displaying natural emotions emerges as a significant predictor of reduced personal accomplishment.When teachers authentically express their true feelings rather than strictly adhering to imposed displayed rules, their feelings of personal accomplishment improve.This aligns with the findings of Akın et al. (2014) and Zhang and Zhu (2008).These findings suggest that fostering a work environment where EFL teachers can openly convey their genuine emotions, without being suppressed by the rules or norms in state and foundation universities, can contribute to decreased sense of burnout and an increased feeling of personal accomplishment.In such an environment, teachers will be more likely to maintain their authentic identities, making burnout more manageable for them.
Finally, our study suggests that deep acting does not have a significant impact on burnout.This finding contrasts with the results of Zhang and Zhu (2008), who identified deep acting as a predictor of emotional exhaustion among Chinese teachers who primarily engaged in this type of emotional labor.Similarly, our findings do not align with Akın et al. (2014) who identified deep acting as the most significant predictor of emotional exhaustion and depersonalization among primary school teachers.

CONCLUSIONS
This study provides evidence that Turkish EFL teachers working in university preparatory schools engage in emotional labor.Teaching is inherently an emotional process influenced by various factors, including contextual and administrative ones, which can either positively or adversely affect it.Therefore, it is imperative to closely examine the university contexts in which these teachers work in order to identify and eliminate any source of strain that may threaten their emotional well-being.In this regard, the administration should take the responsibility to enhance working conditions and provide appropriate emotional and professional support to mitigate the effects of burnout.Schools can improve their social climate for teachers by offering emotional mentoring, which can motivate teachers to cope with the feelings of exhaustion and alienation (Kinman et al., 2011).
Considering that feigning emotions fosters exhaustion and sense of detachment from students, whereas expressing genuine feelings enhances the perceptions of personal accomplishment and alleviates burnout, EFL teachers should be encouraged to act authentically.In this regard, institutions should avoid imposing rigid display rules that compel teachers to conceal or fake their emotions, as such mandates are likely to result in emotional labor and discouragement among teachers.Therefore, the institutions should consider redefining display rules to encourage the authentic expression of emotions and deep acting, rather than surface acting in order to foster the emotional well-being of EFL teachers in preparatory schools.
The study has some limitations that should be taken into account while interpreting its findings.Firstly, the data came from EFL teachers working at five universities in Istanbul.It is important to acknowledge that the emotional labor and burnout levels of teachers in different foundation and state universities located in other parts of Türkiye may yield different results due to diverse institutional and contextual factors.Therefore, future research should aim to involve a more extensive sample of EFL teachers from a wider range of universities.These studies can adopt a comparative view to investigate the emotional well-being of teachers at foundation and public universities.Secondly, the present study adopted a quantitative research design exclusively.This choice was made to investigate correlations between emotional labor and burnout, necessitating statistical analysis of quantitative data.However, incorporating qualitative data could have provided further insights into teachers' levels of burnout and emotional labor, moving beyond mere correlations and shedding light on the complex nature and underlying reasons of these two phenomena.Therefore, we suggest future research should involve qualitative data to gain more comprehensive understanding of burnout and emotional labor, including teachers' coping strategies for dealing with the adverse effects of emotional labor and burnout.
Lastly, our data were collected prior to the Covid-19 pandemic, reflecting experiences in a face-to-face education environment.The emergence of new challenges and factors associated with the sudden shift to online teaching during the pandemic may have influenced teachers' burnout differently.Consequently, further research may investigate in-service teachers' burnout during and after these difficult times of mandatory online education to take necessary measures to protect their psychological well-being.

Table 1 . Descriptive Statistics for ELS
.

Table 4 . Correlation Between the Dimensions of Emotional Labor and Burnout
** shows significance at 99% confidence level