Effects of Intrinsic Motivation on Teacher Emotional Labor: Mediating Role of Affective Commitment

Emotions have an important place in the survival of humankind since its existence. Despite the importance of the emotions in human life and it is not taken on the agenda for a long time in working life. This study aims to examine the impacts of intrinsic motivation on teacher emotional labor with mediating effect of affective commitment. In the study, the relational screening model one of the quantitative research methods was used. The research was carried out 345 teachers who work in ten public schools in Istanbul. As the data collection tools three different scales were used. The intrinsic motivation scale was developed by Lawler and Hall (1970) and adapted to Turkish language by Yılmaz (2008). The emotional commitment scale consists of six items as a sub-dimension of the organizational commitment scale developed by Allen and Mayer (1990). Emotional labor scale, by Diefendorff et al. (2005), it was created, and it was adapted to Turkish in the teacher sample by Basım and Beğenirbaş (2012). Scale has three dimensions as surface acting, deep acting, and natural behaviors. Results of the study suggest that intrinsic motivation has a negative impact on surface acting and has a positive impact on natural behaviors (genius emotions). On the other hand, affective commitment is not mediated this impact. The findings were discussed, and recommendations were made to the practitioners and researchers.


INTRODUCTION
In today's market conditions, human factor takes place among the significant sources to be used in the process of obtaining long-term and sustainable growth. Human factor with its emotional and psychological aspects should be considered within the scope of individual and organizational performance (Biçkes et al., 2014). Individuals experience many emotions both in business and out of work. Emotions is a part of the human behavior and in many areas from the individual's inner world to mutual relations, it is decisive and has a fundamental role in human behavior (Erkuş and Günlü, 2008). Emotions have an important place in the survival of humankind since its existence. Despite the importance of the emotions in human life and it is not taken on the agenda for a long time in working life.
It is difficult to think independently of working life from emotions. Emotions are an integral part of life and management. Emotions are important for understanding the behaviors in the organization. Especially in the works where the human element is in the foreground, its intensity is felt even more. Controlling emotions in organizational life are expected and directed from employees (Çoruk, 2014). Especially in the works where the human element is in the foreground, such as education institutions, this expectation is even higher.
In today's conditions, employees working in the service sector are responsible the task of directing the feelings of the other party by using their own feelings. This understanding is also the starting point of emotional labor constitutes (Oral and Köse, 2011). Emotional labor can be defined as the act of expressing the emotions desired by organizations (Morris and Feldman, 1997). Emotional labor includes evaluating when to express feelings and when not to, assessing what are and what are not appropriate expression of emotion, and managing and regulating such expression of emotions. Grandey (2000) defined emotional labor as both feelings and expressions of emotions for organizational purposes. In organizational context, emotional labor involves displaying the expected feelings by employees during service encounters (Akın et al., 2014).
According to the definition of emotional labor in the context of education proposed by Hochschild (1983), students are similar to customers, and operating a school is similar to managing a company that must sell products and service customers. Teachers commonly interact with people and may address various emotional problems at work. Many studies have suggested that, because of external pressures, teachers who cannot address their emotions promptly experience work stress (Fu, 2015). However, teaching is also an emotional process, in which teachers can manage, monitor, and regulate their emotions to achieve teaching effectiveness and to create a positive learning environment. Ideal teachers are known to emote feelings of enthusiasm, happiness, confidence, self-assurance, and passion about and satisfaction toward teaching (Zhang and Zhu, 2008).
Three strategies of performing emotional labor are surface acting, deep acting and genuine emotions. Surface acting refers to an employee's efforts to pretend to experience certain emotions even if he/she does not really feel them (Ashfort and Humphrey 1993;Diefendorff et al. 2005). Changing expressions to change feelings can be a definition of surface acting (Tsang, 2011). In surface acting, employees modify their feelings as if experiencing the expected emotions without actually changing their feelings. Surface acting may appear as pretending to experience a positive emotion or suppressing a negative emotion (Grandey, 2003). A typical example of surface acting is employees faking a smile to display the organizationally required emotion (Cheung et al., 2011). Deep acting refers to employees' efforts to follow the expected display rules (Ashfort and Humphrey, 1993). Deep acting can be defined as changing feelings to change the expression of emotions (Tsang, 2011). In deep acting, employees try to modify their feelings to act in line with expectations (Grandey, 2003). Third dimension of emotional labor is naturally felt emotions or genuine emotions. Diefendorff et al. (2005) empirically studied genuine emotions to show the differentiation between deep and surface acting. Feeling really sad in case of a customer's problem is an example of genuine emotion. Diefendorff et al. (2005) showed that this emotional labor strategy was negatively correlated with surface acting, but positively correlated with deep acting among human service employees.
Teaching as a profession requires a high level of emotional labor and is different from other professions in certain ways. Teaching is an emotional practice and good teaching is charged with positive emotions. Positive teacher-student relationships are essential to create a sound classroom atmosphere (Akın et al., 2014). Akın and his colleagues (2014) found that more experienced teachers reported higher levels of emotional labor. Their findings highlight the need for teacher-training programs to raise awareness of the emotional demands of teaching and consider ways to enhance emotion regulation skills in experienced as well as recently qualified staff.
Akın and his colleagues found that emotional labor levels of teachers were 4.47 for deep acting, 3.99 for surface acting, and 4.53 for genuine emotions on a five-point Likert scale. Turkish primary school teachers are mostly engaged in genuine emotions followed by deep acting and surface acting. The female teachers used deep acting and surface acting strategies significantly more often than the male teachers do. On the other hand, there were no significant differences in genuine emotions scores (Akın et al., 2014).
Çoruk (2014) studied on emotional labor behaviors of administrative personnel working in universities. According to the results, it is seen that participants display genuine emotions more than other emotional labor strategies and they display surface acting behaviors partially. In terms of gender variable and deep acting aspect, there is a significant difference in favor of male participants and there is not any significant difference for other aspects. As compared to service period in the assigned department, there is a significant difference among the views of participants for deep acting and genuine emotion aspects. Besides, there are also significant differences for deep acting aspect in terms of service period in his/her profession and age variables. Şat and his colleagues (2015) found that there were significant differences of sub dimensions of emotional labor in terms of marital status, gender, educational background, types of institution and seniority of teachers.

Intrinsic Motivation and Emotional Labor
Intrinsic motivation is described as "any motivation that arises from the individual's positive reaction to qualities of the task itself; this reaction can be experienced as interest, involvement, curiosity, satisfaction, or positive challenge" (Amabile, 1996). This concept refers to the state where organizational members are motivated and committed by the task itself and their own enthusiasm for the task (Amabile, 1988). In other words, they are not motivated so much by the external outcome, which refers to the desire to expend effort to gain outcomes external to the work itself (Amabile, 1993). Deci and Ryan (1985) note that the critical component of intrinsic motivation is selfdetermination. Individuals who are intrinsically motivated tend to be driven by inherent interest in the work itself and enjoyment, and thus they feel naturally drawn toward carrying out their work. Here, the decision to make the effort is self-determined (Grant, 2008).
Intrinsic motivation is marked by the interest, curiosity, continued learning, and a spirit of challenge experienced by an employee when stimulated by the work itself rather than external outcomes, such as rewards or the absence of punishment (Deci and Ryan, 1985). Intrinsic motivation is the motivation to perform an activity for itself, in order to experience the pleasure and satisfaction inherent in the activity (Deci et al., 1989).
Despite the general agreement on Deci and Ryan's (1985) self-determination model of motivation as the most pertinent related to emotional labor, there is little empirical evident to support the presumed relations between intrinsic motivation and deep acting, (Sisley & Smollan, 2012). Hochschild's assumption that emotional labor is performed only for pay suggests that this act is extrinsic motivated (and, therefore, not very self-determined), but in current study It could be also argued that employees performing with a smile for the pleasure of satisfying the other (the client) does so for intrinsic reasons.
Engagement in deep acting (attempt to modify the internal, subjective feeling) (Hochschild, 1983) is associated with perception of positive display rules, number of years in service (Grandey, 2000) or with the duration of client-employee interaction. Surface acting (attempt to modify the emotional expression by either suppression or simulation) is associated with perception of negative display rules (Diefendorff et al., 2005) and with negative affectivity (Truta, 2014).
Researchers have shown that intrinsic motivation has an effect on organizational behavior subjects such as emotional exhaustion, turnover intention, career expectation (Haukes et al., 2003), work engagement (Putra et al., 2017) organizational identitiy (Schutz and Lee, 2014), job performance (Grant, 2008;Karatepe and Tekinkus, 2006;Lawler and Hall, 1970;Tierney, Farmer andGraen, 1999, Collins, 2010;Dinger et al., 2015). However, there is not enough research about relationship between intrinsic motivation and emotional labor. Truta (2014) found that intrinsic motivation as predictors explains 35% of the total variation of deep acting use. Analysis of unstandardized coefficient shows that between intrinsic motivation and deep acting the relation is a positive one (B = .271), the increase of intrinsic motivation is association with a positive change of the deep acting.
Based on the literature discussed above; Hypothesis 1a: Intrinsic motivation negatively affects surface acting.

Intrinsic Motivation and Affective Commitment
As one dimension of organizational commitment, affective commitment is the emotional affinity of the employee and his identification with in organization (Allen and Meyer, 1990). Affective commitment is described respect for organizational goals and rules, love and respect for managers, emotional intimacy, sharing organizational vision and mission (Yılmazer, 2010). Employees with high affective commitment adopts the values, goals and objectives of their organizations; they want to remain part of the organization. Affective commitment has three basic elements. These are the belief in organizational purpose and values and acceptance of them, volunteering to make more efforts for the benefit of the organization and a strong desire to continue of organizational membership (Iverson and Buttigieg, 1999). Affective commitment is not an individual's passive beyond obedience but it includes active sacrifice for the organization to be better. High affective commitment affects many positive performance outcomes such as teamwork, participation and empowerment (Thakreve and Mayekar, 2016). In fact, Pinder (1998), defined motivation as a body of energizing forces, while Meyer and Herscovitch (2001) defined commitment as a force connecting the individual to a course of action. Meyer et al. (2004) in accordance with Becker et al. (1996) indicated, "The primary bases for the development of affective commitment are personal involvement, identification with the relevant target, and value congruence".
Prolific research in the area of relationships between work motivation and organizational commitment has been conducted over the past few years (Warsi et al., 2009;Ingram et al., 1989). Bono and Judge (2003) highlighted a positive relationship between autonomous motives for accomplishing work objectives and affective commitment. Galletta, Portoghese and Battistelli (2011) found that intrinsic motivation is positively related to affective commitment. Moreover, Lam and Gurland (2008) found that self-determined work motivation predicted identification commitment. Millette and Gagné (2008) showed that autonomous motivation of volunteer workers was positively related with their engagement in the volunteer work. Finally, Choong, Wong and Lau (2011) found that intrinsic motivation has significantly predicted the organizational commitment. Based on these results, it will be likely that when people feel to have to accomplish work tasks, they will become attached to, and feel obliged toward their own organization.

Affective Commitment and Emotional Labor
Commitment in the workplace is a force that binds an individual to a course of action that is of relevance to one or more targets, accompanied by a specific mind-set (affective in this case) reflecting the emotional attachment and identification to the target, which plays a role in shaping and motivating workplace behaviors (Meyer and Allen, 1991;Meyer and Herscovitch, 2001). Affective commitment represents an indicator of person-environment fit in the workplace (Cohen, 2003;Morrow, 1993) Commitment to the rules might be intimately related to their reliance on particular emotional labor strategies, and may reflect the ease with which employees will adopt these rules (Ashforth and Humphrey, 1993;Gosserand and Diefendorff, 2005;Hochschild, 1983). Indeed, previous research showed that the relationships between attitudes (i.e. commitment) and behaviors (i.e. emotional labor displays) tend to be stronger and more significant when they refer to similar contexts, actions, timelines, and targets (Lavelle et al., 2007). Gülova and her colleagues (2013) found that a positive significant relationship between three emotional labor behavior and organizational commitment on student affairs office employees. Finally researchers found that expression of naturally felt emotions significantly the related between affective commitment (Lapointe et al., 2012).
Hypothesis 3a: Affective commitment negatively effects surface acting.
Hypothesis 3b: Affective commitment negatively effects deep acting.
Hypothesis 3c: Affective commitment positively effects genuine emotions. In summary, the conceptual framework in this study links intrinsic motivation and emotional labor with a partially mediating role of affective commitment. Figure 1 illustrates the conceptual model of this study.

METHOD
With this in mind, related literature was reviewed within the context of the study; such as the impacts of intrinsic motivation on emotional labor with the mediating effect of affective commitment were examined in ten public schools in Istanbul.

Intrinsic Motivation
Emotional Labor

Participants
Participants were 345 teachers and managers working in eight schools located in Turkey. The response rate was 100% for teachers, 99% of the teachers asked to participate were included in analyses. As shown in Table 1, 64.9% of the participants were female; the majority were between 25-34 years old (54.3%); 57.9% of them were branch teachers; 88.7% of them have a bachelor degree, 31.0% of them work in primary school.

Data Collection Tools
Three different scales were used to collect data. The intrinsic motivation scale was developed by Lawler and Hall (1970) and adapted to Turkish language by Yılmaz (2008). It included 4 dimensions named as interest, sufficient vision, preference, respectability and 20 items. Items rated on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Reliability of the scale in the current study was found to be .90.
The emotional commitment scale consists of six items as a sub-dimension of the organizational commitment scale developed by Allen and Mayer (1990). In the teacher sample of the scale, the Turkish adaptation was made by Han et al. (2018). The scale items are answered as 5-point likert, ranging from "strongly disagree" to "strongly agree". The internal consistency coefficient of the scale was found as .91 in this study.
Emotional labor scale, by Diefendorff et al. (2005), it was created, and it was adapted to Turkish in the teacher sample by Basım and Beğenirbaş (2012). Scale has three dimensions as surface acting, deep acting, and natural behaviors. The participants evaluated with the help of the five likert scale (1 = Never, 5 = Each Time). Surface acting in scale 6, deep acting 4 and natural behaviors is measured by 3 items. Internal consistency of the scale has been calculated as 80 (Basım ve Beğenirbaş, 2012). In this study, the internal consistency coefficient has been found .85 for surface acting, .88 for deep acting, .86 for and natural behaviors. The internal consistency coefficient of the overall scale was found as .77 in this study.

Data Collection Process
All participants were full-time schoolteachers residing in the Ministry of Education. Invitation letters were sent to 10 schools in Küçükçekmece area, a major district located in the İstanbul city. All the invited schools, two school from each school type (kindergarten, primary school secondary school, Anatolian high school, vocational high school) agreed to participate in this study. Each participant received a questionnaire form, in which the study objectives, the use of information, and psychological measures were included. Confidentiality was ensured and no individual response was reported to the school. To enhance the response rate, a summary of findings were provided to individual participants as a token of appreciation.

The Statistical Analysis of the Data
Descriptive statistics (Mean, SD, correlation) of major variables were first computed. The role of affective commitment as a moderator between intrinsic motivation and emotional labor, were examined with a series of moderated regression analyses. Using the procedure suggested by Baron and Kenny (1986), the main effects of the predictor and hypothesized moderator were controlled before determining the moderating effect of the hypothesized moderator on the predictor-outcome association. Following the recommendation of Aiken andWest (1991) andFraizer, Tix, andBarron (2004), all main effects of predictors and interaction terms were centered before entering the regression analyses.

FINDINGS
Means and standard deviation of variables are shown in Table 2. Table 2 shows that the highest mean is genuine emotions with 4.09. The second highest mean is intrinsic motivation with 4.02, followed by affective commitment with 3.86 and deep acting with 3.13. The lowest mean is surface acting with 2.07. Linear regression analysis was conducted for Hypothesis 1a, 2a and 3a, using the same control and criterion variables with Hypothesis 4a. The results (see Step 3 in Table 3) showed that the relationship between intrinsic motivation and surface acting was significant and negative ( = .368, p < .01). Hence, Hypothesis 3a was supported. Hypothesis 4a was examined using the procedure proposed by Baron and Kenny (1986): (a) The predictor (intrinsic motivation) must be related to the criterion (surface acting), as supported by Hypothesis 1a; (b) The predictor (intrinsic motivation) must be related to the mediator (affective commitment), as supported by Hypothesis 2a; (c) The mediator (affective commitment) must be related to the criterion (surface acting), as supported by Hypothesis 3a; (d) The effect between the predictor (intrinsic motivation) and the criterion (surface acting) must decrease when the mediator is controlled. When affective commitment was entered in the regression model (see in Table 4), the beta coefficient for job performance was statistically significant and reduced from  = -.288 (p < .001) to  = -.264 (p < .01), but the beta coefficient for affective commitment was not still significant ( = -.063, p > .05). Thus, affective commitment does not mediated the relationship between intrinsic motivation and surface acting, and Hypothesis 4a was not supported. Results of the study suggest that intrinsic motivation has a negative impact on surface acting but affective commitment does not mediated this impact.  Linear regression analysis was conducted for Hypothesis 1b, 2b and 3b, using the same control and criterion variables with Hypothesis 4b. The results showed that the relationship between intrinsic motivation and deep acting was not significant. Hence, Hypothesis 3b was not supported and Hypothesis 4b was not examined.
Linear regression analysis was conducted for Hypothesis 1c, 2c and 3c, using the same control and criterion variables with Hypothesis 4c. The results (see Step 3 in Table 5) showed that the relationship between intrinsic motivation and genuine emotions was significant and positive ( = .368, p < .01). Hence, Hypothesis 3c was supported. Hypothesis 4c was examined using the procedure proposed by Baron and Kenny (1986): (a) The predictor (intrinsic motivation) must be related to the criterion (genuine emotions), as supported by Hypothesis 1c; (b) The predictor (intrinsic motivation) must be related to the mediator (affective commitment), as supported by Hypothesis 2c; (c) The mediator (affective commitment) must be related to the criterion (genuine emotions), as supported by Hypothesis 3c; (d) The effect between the predictor (intrinsic motivation) and the criterion (genuine emotions) must decrease when the mediator is controlled. When affective commitment was entered in the regression model (see in Table 6), the beta coefficient for genuine emotions was statistically significant but increased from  = .344 (p < .001) to  = .348 (p < .01), and the beta coefficient for affective commitment was not significant ( = .033, p > .05). Thus, affective commitment does not mediated the relationship between intrinsic motivation and genuine emotions and Hypothesis 4c was not supported. Results of the study suggest that intrinsic motivation has a positive impact on genuine emotions but affective commitment does not mediated this impact.

DISCUSSION AND CONCLUSION
In this section, the research findings are discussed with previous research findings also taken into consideration. The levels of teachers' intrinsic motivation were found high (between 3.41-4.20). In other words, teacher are not motivated so much by the external outcome, which refers to the desire to expend effort to gain outcomes external to the work itself (Amabile, 1993). Deci and Ryan (1985) note that the critical component of intrinsic motivation is self-determination. However, the emotional labor scores change according to the factors. The highest average scores were collected under the factor of "genuine emotions" ("high" between 3.41-4.20) and the lowest mean scores were collected in the "surface acting" factor ("low" between 2.01-2.61). Also, deep acting factor has average score ("middle" between 2.61-3.40). Akın and his colleagues results are not consistent with current study finding. They found that emotional labor levels of teachers were 4.47 for deep acting, 3.99 for surface acting, and 4.53 for genuine emotions on a five-point Likert scale. Only genuine emotions has highest score in both research. According to the Çoruk (2014) results, it is seen that participants display genuine emotions more than other emotional labor strategies and they display surface acting behaviors partially. In addition, these findings are consistent with current study findings.
Based on the results above, it was found that the negative effect of intrinsic motivation on surface acting is 8.30% and the positive effect of intrinsic motivation on genuine emotions is 11.80%. Previous results also support this finding and it was found that intrinsic motivation affects emotional labor. When individuals' performance in an organization is based on intrinsic motivation, they tend to be highly engaged in the task itself, and as a result, their performance improves. According to Grant (2008, 49), when people are intrinsically motivated, they tend to be process focused and thus, they view their task as "an end in and of itself". Researchers have shown that intrinsic motivation has an effect on organizational behavior subjects such as emotional exhaustion, turnover intention, career expectation (Haukes et al., 2003), work engagement (Putra et al., 2017) organizational identitiy (Schutz and Lee, 2014), job performance (Grant, 2008;Karatepe and Tekinkus, 2006;Lawler and Hall, 1970;Tierney, Farmer andGraen, 1999, Collins, 2010;Dinger et al., 2015). Surface acting (attempt to modify the emotional expression by either suppression or simulation) is associated with perception of negative display rules (Diefendorff et al., 2005) and with negative affectivity (Truta, 2014). These findings are consistent with current study findings.
In current study, the results showed that the relationship between intrinsic motivation and deep acting was not significant. However, Truta (2014) found that intrinsic motivation as predictors explains 35% of the total variation of deep acting use. Analysis of unstandardized coefficient shows that between intrinsic motivation and deep acting the relation is a positive one (B = .271), the increase of intrinsic motivation is association with a positive change of the deep acting. This finding is not are consistent with current study findings.
It was found that the effect of intrinsic motivation to affective commitment is 30%. Bono and Judge (2003) highlighted a positive relationship between autonomous motives for accomplishing work objectives and affective commitment. Galletta, Portoghese and Battistelli (2011) found that intrinsic motivation is positively related to affective commitment. Moreover, Lam and Gurland (2008) found that self-determined work motivation predicted identification commitment. Millette and Gagné (2008) showed that autonomous motivation of volunteer workers was positively related with their engagement in the volunteer work. Finally, Choong, Wong and Lau (2011) found that intrinsic motivation has significantly predicted the organizational commitment. Based on these results, it will be likely that when people feel to have to accomplish work tasks, they will become attached to, and feel obliged toward their own organization. ). All of these findings are consistent with current study findings.
It was found that the negative effect of affective commitment on surface acting is 4% and the positive effect of affective commitment on genuine emotions is 4% . Commitment to the rules might be intimately related to their reliance on particular emotional labor strategies, and may reflect the ease with which employees will adopt these rules (Ashforth and Humphrey, 1993;Gosserand and Diefendorff, 2005;Hochschild, 1983). Indeed, previous research showed that the relationships between attitudes (i.e. commitment) and behaviors (i.e. emotional labor displays) tend to be stronger and more significant when they refer to similar contexts, actions, timelines, and targets (Lavelle et al., 2007). Gülova and her colleagues (2013) found that a positive significant relationship between three emotional labor behavior and organizational commitment on student affairs office employees. Finally researchers found that expression of naturally felt emotions significantly the related between affective commitment (Lapointe et al., 2012).

SUGGESTIONS
In this section, recommendations based on the research findings are presented under two headings, namely to practitioners and researchers.

Recommendations for Practitioners
1. Intrinsic motivation has positive effect on genuine emotions. Increasing intrinsic motivation level of teachers will increase the genuine emotions.