Introduction

Effective and healthy development of relationships between people that have been more intense due to advances in technology are among the most significant issues of 21st century. One of the most important results of the human relationship is emotional burnout. Known as one of the most important problems of today’s society, emotional burnout issue has been more popular in recent years. It is observed that the studies carried out in relation to burnout are about healthy workers and teachers. Engin (1992), Ustun (1995), Demir (1995) pointed out that emotional burnout is seen among doctors and nurses; Girgin (1995), Torun (1995) and Tumkaya (1996) noted that it is also observed among teachers and educational leaders. It is also seen as a factor resulting in changing jobs, redundancy and early retirement, loss of skilled laborers, a reduction in service quality and other economic losses (Tanriverdi and Sarihan, 2013).

For a better service and to achieve better organizational goals, employers request employees to exert more efforts and make sacrifices emotionally. Employees who cannot respond to these demands become more subjected to emotional burnout and depression. Just as these syndromes decrease efficiency in workplace, it causes risks in health of employees and quality of work. Problems employees face at work where they spend most of their time negatively influence employees’ personal and work life and typically causes long-term discomfort (Ashforth and Humphrey, 1993). Since these discomforts that are named as burnout, emotional labour and depression which are experienced at individual level lead to undesired consequences both for employees and organizations, they have become issues researchers focus on more intensively.

Emotional burnout is experienced by employees who work in education, service and health sectors, by leaders and by university students who have not even started working. However, emotional labor that is specifically displayed during interaction with people is thought to be related to emotional burnout in different ways than other factors (Ergin, 1993). In studies conducted in Turkey and other countries, it is shown that emotional burnout is a problem faced by employees working in education sector.

Today the increase in the service sector and its importance causes high levels of competition in this field. Therefore, the quality of the service, customer satisfaction and fulfilling service organization’s goals are possible through the efforts of workers who serve and are often in face-to-face communication with other people. Thus, feelings and patience displayed by these employees are important in this process and therefore employees are requested to keep the customer satisfaction at the highest level using the state of emotional labor very well. Although feelings are abstract and exclusive to that specific person under normal circumstances, it has become almost a concrete product, which has an economic value in today’s world (Ozgen, 2010). In recent years, with the rapid growth of the service sector, ‘emotional labor’ has become an indispensable element of many service sectors and professions. Starting from the concept of emotion and labor, this concept which is qualified as ‘the conscious process of human impression evoked in his inner world’ continues to be focused in organizational behavior and its sub-disciplines by researchers and it continues to be a popular subject of research (Nejat, 2012).

Emotional labor is defined as presentation of workers’ feelings in case of predetermined patterns by controlling their emotions to evoke a positive impression on the people they have to deal when they work (Hochschild, 1983). Today, efforts and experiences the workers put forth while they are editing or projecting the feeling of the workers in accordance with the job requirements laid down, come to the fore and are also referred to as a sense of emotional labor (Cukur, 2009). The concept of emotional labor in mutual interpersonal relationships is defined as expression of workers’ feelings required by the organizations. (Robbins, 2005).Hochschild (1979), who brought the concept of emotional labor to the field of organizational behavior science, suggested that feelings are bought and sold in labor market, defined this concept as ‘the management of emotions for a fee’. (Hochschild, 1983, cited in Grandey, 2000). As can be understood from the definition, the concept is evaluated in the context of social exchange process and the studies of the concept are observed to focus on the emotions of individuals in the process of social interaction and how they manage the behaviors raised with these feelings.

According to Hochschild who widely discussed and brought up the concept of emotional burnout and emotional labor in organizations and businesses for the first time; emotional labor is the regulation of feelings to be observed by other people and presentation facially and bodily (Hochschild, 1983). While Hochschild says that emotional labor is shown ad observed facially and bodily; in addition to this view, other researchers (Wharton and Erickson, 1993; Rafaeli and Sutton, 1987), who conducted research on this subject, pointed out that emotional labor involves every effort reflected in the tone of voice and behaviors during speech. In this context, there have been researchers who say that emotional labor is the attitude displayed in face to face interviews and conversations. (Steinberg and Figart, 1999) Scientists describing emotional labor in other words define emotional labor underlining the impact on others as the effort to understand others, empathize with others, feel others’ feelings as if they were their own feelings. (England and Farcas, 1986). Also emotional labor is emphasized to deal with others ‘feelings, to understand them, to value what they feel, to care them and that this attitude is a key factor in the regulation of emotions. (James, 1989) Asforth and Humprey (1993) deal with emotional labor as an observable behavior status instead of the management of feelings. In parallel with this, emotional labor can be regarded as efforts spent by employees who strive to exhibit the feelings and behaviors required by organizations during the interaction and communication with people and the planning and controlling requirements of these efforts (Morris and Feldman, 1996). In another sense, based on all sorts of clues and signs we receive from people’s behavior and verbal expression to understand them; emotional labor is an emotional work during which we ignore the importance of our true feelings and even suppress them (Kleinman and Coop, 1993).

Emotional burnout can be shortly defined as mental and physical aspects of energy depletion as a result of the high demand of the life. According to Dr. Maslach (psychologist) working on burnout and conducting studies, this term is defined in three dimensions: emotional and physical burnout, reduction in personal accomplishment and depersonalization. According to Maslach’s definition, in addition to being subjected to physical and emotional distress, people experiencing burnout consumes his desire to work. He takes a negative, unfavorable, a cynical and skeptical attitude against work and himself. Also, emotional burnout is defined as the depletion of an individual’s internal resources because of fail, wear, loss of power and energy as a result of overload or unmet requests which are regarded as a threat in many professions (Freudenberger, 1974). Emotional burnout is a syndrome characterized by lower dimensions such as emotional burnout, depersonalization and reduced sense of personal accomplishment. (Maslach and Jackson, 1981) Emotional Burnout Syndrome is the feeling experienced when the person or workplace put big goals in front of the employees, then you can’t reach the objectives of the organization or what they want, you get tired of it and run out of energy by being disappointed (Ersoy, Yildirim and Edirne, 2001). Emotional burnout seen in people who are in an intense relationship with others can be defined as depersonalization and reduced personal accomplishment syndrome (Maslach and Zimbardo, 1982). The prolonged time spent with the people faced in the business environment, frequent interaction with the same people, increasing number of people served, having severe problems of the people he serves and face-to-face relationship with these people increase the level of burnout. Considering this situation in terms of teachers, teachers’ frequent face to face relationship with students, parents, managers and staff in workplace, working in crowded classes and schools can be said to influence and increase the level of burnout.

Different from emotional labor and burnout, depression is a state of illness. People in depression are emotionally unhappy, pessimistic and hopeless. One of the prominent signs of depression is decreasing interest in work, getting no sense of pleasure from life and work and lack of involvement in work. He usually feels bored, discomforted. Sometimes, he may feel like he lost all his feelings. His interest in the world decreases in depression or completely disappears, the capacity to invest in and connect other people disappears; the states of self-reproach, condemnation, reproach, seeing himself down appears and individuals lose their self-esteem substantially. (Cevik and Volkan, 1993; Mak and Cheung, 2011; CostaRequena et al., 2011).

The aim of this research

Emotional labor, emotional burnout and depression have become issues that have been receiving attention from researchers as they could cause problems in personal and organizational life. To identify the problems causing burnout and bring solutions to them has become very important today. This study, therefore, aims to bridge the gap in literature in terms of relationships between three important organizational variables.

Methodology

Since the study aimst to search the mediating effect of depression on the relationship between teachers’ emotional labor and burnout, the research design is specified as the relational scanning model. Relational scanning model, as a scanning model, is a research model which aims to define the presence or degree of a large number of variables between two or more variables. The relational scanning model has two types namely the correlation type and the comparison type. While in the correlation type, whether the variables change with the change or not in the other variable and how this change occurs; in the comparison type it is examined whether there are differences based on dependent variables by forming groups based on independent variables at least between two variables. (Karasar, 1995:81-82)

The population of this research consists of teachers working at schools in the city of Gaziantep during 2013-2014. The sample of research consists of 90 male, 73 female teachers and a total number of 163 teachers were chosen objectively with the stratified sampling technique.

Data collection tools

In this research, three scales were used to collect the necessary data. Burnout Scale was developed by Maslach and it was adapted to Turkish by Cam (1992). Emotional Labor Scale was developed for teachers by Cukur (2009) and it has 20 items. The reliability coefficient was found to be 0.187 as a result of the reliability analysis.Depression Scale was developed by Goldberg (1999) and adapted to Turkish by Karakus (2012). The reliability coefficient was found to be 0.874 as a result of reliability analysis.

Statistical Analysis

The analysis of the data was carried out in the SPSS 17.0 statistics package program. Data was analyzed using hierarchical multiple linear regression method. Before starting the analysis of the data, data was examined in terms of missing values, discrete values, normality and multiple changes; in other words, assumptions of the analysis were tested. In our research model, while analyzing the effect of independent variables such as emotional labor and depression which predict dependent variable, burnout level of teachers, hierarchical multiple regression in which gender, age and marital status changes were controlled was analyzed. The mediating effect which was first raised by Baron and Kenny (1986), can be defined as the state of being the third mediator variable in the relationship between the two variables. It depends on certain conditions to say that the third variable mediates the relationships between the two variables or it has an indirect effect (Savas, 2012).

As seen in figure 1, there are Predictor, Predicted and Mediator variables. Baron and Kenny point out that to prove the mediating effect by regression analysis, you need to fulfill the following four conditions. Predictor predicts Predicted significantly. Predictor predicts Mediator significantly. Mediator predicts Predicted by controlling the effect of Predictor. There has been a significant decrease in Predictor’s predicting the Predicted by controlling the effect of Mediator or this relationship becomes statistically insignificant.

Figure 1. The mediating effect in Baron and Kenny model

In the fourth stage, if it becomes insignificant in the case that Predictor predicts the Predicted, Mediator is called ‘whole mediating variable’; if there is a significant decrease in the case that Predictor predicts The Predicted, mediator is called ‘partial mediating variable’ (Frazier, Tix and Baron, 2004). In partial mediating effect, we can say that Predictor has both a direct effect and an indirect effect through mediator. In such a model, the effect of predictor on predicted variable which is calculated without considering (controlling) is the total of the direct and indirect effects (Savas, 2012).

There are fourteen different testing methods to test the significance of the mediation effect in literature. Sobel Test, which is frequently used in literature and defined to produce the most reliable results, was used (Simsek, 2007). MedGraph Program, which is developed by Jose (2013) to use Sobel Test and is available for use on web, was used.

Findings

Demographics of teachers who participated in this research, their frequency percentage were given on Table 1 according to gender, marital status and age variables

Table 1. The frequency and percentage of the participants according to demographic variables

When we look at Table 1 according to gender variable, our working group consists of 90 male teachers and 73 female teachers. Thus, there are more male teachers in the study group. When we look at the marital status variable, there are 129 married, 34 single teachers. We can see that the percentage of single teachers is quite less than the percentage of married teachers. According to age variable, 42 % of teachers are between 20-29, and 39.9 % of teachers are between 30-39, 18 % of teachers are over 40 years old. Here, we can see that most of the teachers are under 40, and few of them are over 40.

Findings

Table 2. Multiple Regression Analysis Results about the Mediating Effect of Depression on effect of Teachers’ Emotional Labor on Burnout

ΔR2=0,407***, (*p<.05, **p<.01, ***p<.001)

As seen on Table 2, in the regression analysis, after the depression level was added to the model, the effect of emotional labor on emotional burnout decreases from β=-0.181 to β =-0.168. That means that the effect of emotional labor decreases. That emotional labor effect decreases after the level of depression is added to the model, that emotional burnout increases and this effect is significant shows that depression level is mediating variable in this relationship. When depression level is ignored, the effect of emotional labor on emotional burnout may be high. This may cause us to make a mistake.

In order to implement Sobel Test about the significance of mediating effect, some data on Table 2 were added to MedGraph 1 program and following Table is created.

Table 3. Results of Sobel Test Analysis about The Significance of Mediating Effects of Depression Levels on The Effect of Teachers’ Emotional Labor on Burnout

As seen on Table 3, Sobel Test, which is implemented, by adding some data on Table 2 into MedGraph 1 program, shows that the mediating effect of emotional labor seems to be significant in the level of p<,005. When results of analysis made to identify mediating effects are analyzed, it is seen that depression is partial mediator on the effect of emotional labor on the burnout.

Figure 2.The mediating effect of depression level on levels of teachers’ emotional labor predicting burnout and identified beta coefficients.

As seen in figure 2, as expressed in the model above about the mediating effect of depression levels on levels of teachers’ emotional labor predicting burnout, depression levels seem to be a partial mediator. This means that ignoring depression levels in the model causes the effect of emotional labor on burnout to be calculated higher than it is. While emotional labor predicting about emotional burnout is β=-0.168 namely significant, it seems that emotional labor is a good predictor of depression.

Result and Discussion

This study has shown that there is a mediating role of depression on the effect of emotional labor on burnout. In the regression analysis made about the effect of emotional effect on burnout, after the depression level is added, the effect of emotional intelligence decreases from -0,181 to -0,168. The fact that emotional labor effect decreases after the depression level is added to the model and the significance of this effect shows that depression level is a mediator variable in this relationship. As a result of these findings, it is concluded that the features of teachers’ emotional labor affect their emotional burnout and that there is a mediating role of depression status while performing their professions. It is identified that depression relationship decreases emotional burnout. In the light of these findings, it is thought that displaying anxious, stressed and tense behaviours and feeling displeased and needy (Carver and Scheier, 1990) increase their emotional burnout levels. However, it is also observed that people who are in an effort to communicate with other social, self-confident, energic, friendly and cheerful people around exhibit a decrease in the emotional burnout.

As also shown by Iorga, Stanescu and Illiescu (2012), emotional laborers experience negative affective outcomes at work place and this study has illustrated that burnout is one of these negative affective outcome directly being influenced by emotional labor levels of teachers. Yilmaz, Altinlurt, Guner & Sen (2012) and Jim & Guy (2009) have also revealed the interaction between emotional labor and emotional burnout. Yilmaz et al. (2012) have also pointed out that both surface acting and naturally-felt emotions are important predictors of emotional exhaustion and depersonalization of teachers. it seems that emotional labor is a good predictor of depression.

Looking at the models tested and verified in the current study, it could be stated that the more teachers try to hide their emotions and try to sacrifice themselves by showing a good attitude towards others even in hard conditions, the more they become inclined to emotional burn-out. Furthermore, teachers’ high emotional labor levels make them more subjected to depression which then lead to emotional burn-out experiences. A related finding was reached in service sector by Bayram, Aytac, Dursun (2012) who also emphazise that when employees hide their emotions their emotional exhaustion and depersonalization levels rise up. This takes us to Iorga, Stanescu and Illiescu (2012)’s statement that there is a story behind service with a smile.

This study cast lights on the importance of emotional labor phenomenon and emotional burnout in many organizations and how they could negatively working life, lead to syndromes like depression how they could disrupt the quality of life (Diefendorff et al., 2005). Teaching is an intellectual and moral practice fraught with contradictions, impediments, and challenges both quotidian and extraordinary (Santoro, 2011). Working as a teacher means being exposed to depression and more importantly to burn-out as it necessitates a lot of emotional labor work. Having said that, policy makers and school administrators should be made aware of these emotional and psychological risks that teachers could experience in their daily practices and act more responsibly towards teachers.

Suggestions for policy makers

Administrators should be informed about how to reflect their feelings to the employees.

In order to improve the education quality of schools, school managers should take a step to decrease the teachers’ burnout levels.

In this context, one of the things to be done is to try to reduce the depression levels of the staff.

School managers should determine the sources of depression and try to help teachers get rid of it.

Suggestions for researchers

The researchers may carry out different studies about what the other variables that cause burnout and depression are.

This study was conducted as a quantitative sample of Gaziantep. Similar research in different countries in different samples can be carried out. Also, qualitative studies could be made to better explore these concepts.