The effect of social comparison tendencies on EFL teachers’ experience of burnout and instructional self-efficacy

Abstract Factors influencing teachers’ level of burnout have been studied by different researchers, however very few focused on the relationship among three subscales of: burnout, social comparison tendencies (upward and downward), and teachers’ instructional self-efficacy. Hypothesizing that a knowledge of the relationship between these factors may promote teachers’ mental and emotional state, and hence improve their performance in classrooms, this study explored the possible relationships between the above mentioned variables. To this end, a sample of 279 Iranian teachers (209 female, 70 male) from across the country agreed to act as the participants of this study, and provide the answers to a 46-item scale (BSCSE) which addressed their degrees of burnout, social comparison tendencies (upward and downward), and instructional self-efficacy. The analyses of the data based on the Spearman Rank Order Correlation tests indicated that there were a positive significant relationship between downward comparison and the first two subscales of burnout: exhaustion and depersonalization. In addition, teachers’ upward comparison tendency showed a positive effect on teachers’ sense of instructional self-efficacy. The findings of this study provide useful information for the EFL teachers and teacher supervisors who seek specific pedagogies that decrease the level of burnout in the educational context.


ABOUT THE AUTHORS
Niloufar Saber Gigasari is a graduate student and a language teacher. She is interested in language teaching, classroom management, and the issues in language learning. This article that focuses on the relationship between teachers' social comparison tendencies, self-efficacy and burnout in English language teaching profession is based on part of her MA thesis.
Jaleh Hassaskhah is currently a full time associate professor at the English Department of the University of Guilan. Her areas of interest include: teacher development and assessment, as well as exploring new horizons in language teaching. She has published different articles and books and also served as the reviewer and lecturer on these topics both nationally and internationally.

PUBLIC INTEREST STATEMENT
How teachers feel at work is affected by several social and psychological internal or external factors. One of these factors is burnout which is characterized by teachers' feeling of emotional drain, showing negative attitudes toward their students, and losing their feelings of accomplishment, which in turn may determine the quality of teachers' personal and social lives. This study explored the question of how teachers' social comparison tendencies may affect their experience of burnout, as well as their sense of instructional selfefficacy. The findings of the study suggest that when teachers compare themselves with colleagues whose performance is slightly weaker, they feel more tired and lonely, while their comparison with those doing slightly better boosts their sense of instructional self-efficacy which is the belief of a teacher in his or her teaching skills or abilities. The study implies that awareness of such tendency and of the factors which may promote teachers' mental and emotional state, may improve teachers' performance in their classrooms, and hence lead to students' learning.

Introduction
Teachers' main duty is to educate and guide the members of the society. However, teachers, like every other human beings, are susceptible to various threatening social and psychological factors which may affect their personal and social lives. One of those factors that may influence teachers and their attitude toward teaching is "burnout" which is defined as a psychological syndrome of emotional exhaustion, depersonalization, and sense of low personal achievement (Brenninkmeijer, 2002;Maslach & Jackson, 1981).
Burnout may develop when teachers through social comparison evaluate their abilities against others', especially when objective, nonsocial standards are not available (Taylor & Lobel, 1989;Wood, 1989). Such influencing mechanism of comparison between the self and others can affect teachers' judgment, experience, and behavior (Festinger, 1954), and although not mentioned directly by Festinger, it triggers the processes of coping with stress (Taylor, Buunk, & Aspinwall, 1990). Research shows that the direction and the amount of social comparison (upward or downward) can be influenced by the desire or need for self-enhancement (Festinger, 1954;Wills, 1981).
The upward direction of social comparison is interpreted differently by different researchers at different times. While some interpret it as people's interest in comparing themselves to those whose performance or abilities are slightly better (Taylor & Lobel, 1989), others (Wood, 1989) see it as an interest in people's evaluating their abilities and trying to improve these abilities. Grabowski and Broemer (2014), also suggested that self-evaluations are increased by the existence of social comparison with superior in-group members.
The downward direction of social comparison (Wills, 1981) focuses on the feasibility of increasing the subjective well-being through comparison with less fortunate ones. As a matter of fact, individuals attempt to compare themselves with others who are worse off, as a way to ameliorate their well-being. Wills (1981), introduced threatening situations as the cause of people's involvement in downward comparison and mentioned that it helps them to feel better about themselves, their situations and so forth. Of course, the ways people respond to evaluative comparison information is determined by the comparative context (e.g. Brewer & Weber, 1994;Schmitt, Silvia, & Branscombe, 2000). Bandura (1977) relates teacher-efficacy to self-efficacy, and argues that the latter is a cognitive process in which individuals construct beliefs about their own capabilities to organize and execute courses of action (Tschannen-Moran, Hoy, & Hoy, 1998). He also suggests outcome expectancy beliefs as the second component of self-efficacy. The second component was already defined by Ramey-Gassert and Shroyer (1992) as the belief that teachers' effective teaching can influence students' learning. Therefore, it can be inferred that perceived self-efficacy belief is distinguished from outcome expectancies belief since one might believe that a particular action will produce a specific outcome but the outcome might not be achieved as long as the person does not believe in his/her abilities to accomplish the action and in addition this belief might stop the ongoing action as soon as an obstacle appears. Schunk, MSchunk, and Meece (2006) and Meece (2006), also introduced personal efficacy as the outcome of the interactions among personal factors, behaviors, social and environmental conditions. As changes on teachers' attitude toward teaching can lead to emotional and even physical consequences, Cordes and Dougherty (1993), as well as Alavinia and Ahmadzadeh (2012) argue that it is of great importance to identify the variables and their correlates that affect the amount of job burnout among teachers.
The current study, therefore, was planned to both deepen our understanding of the variables (burnout, social comparison, and self-efficacy) and identify their possible relationships to show how the direction of these relationships may affect teachers' performance. In particular, burnout and its three dimensions of exhaustion, cynicism, and sense of inefficacy are going to be studied in relation to teachers' social comparison tendencies and their sense of instructional self-efficacy.

Review of literature
Teacher burnout has attracted the attention of many researchers all over the world (e.g. Buunk, Ybema, Gibbons, & Ipenburg, 2001;DeFrank & Stroup, 1989). While Maslach and Jackson (1981) defined burnout as "a syndrome of emotional exhaustion and cynicism that occurs frequently among individuals who do 'people work' of some kind" (p. 99), Jennett, Harris, and Mesibov (2003) associated it with a long-term occupational stress.
Burnout was also defined as a multidimensional construct. Maslach (2003) categorized it into three dimensions of emotional exhaustion, cynicism, and sense of inefficacy (Maslach, 2003). These three dimensions are defined differently by different researchers. Schwab (1983) defined emotional exhaustion as a prominent feature of burnout syndrome and argued that intensive interaction on a prolonged basis will lead to the emotional drain of teachers and teachers cannot act as they did earlier in their jobs. Jennett et al. (2003) referred to emotional exhaustion as a condition in which teachers' emotional resources are low and they experience the feeling that they are no longer able to give psychologically of themselves, and the latter was also defined as a condition in which teachers experience negative, cynical, indifferent attitudes, and feelings about their students.
The second dimension of burnout syndrome, cynicism, also called depersonalization, points to a condition in which teachers seclude themselves from the students and different aspects of the job (Maslach, Schaufeli, & Leiter, 2001). Flourishing negative cynical attitudes toward the students and imparting these attitudes in ways such as using derogatory labels was called depersonalization by Schwab (1983). This is the case while Skaalvik and Skaalvik (2007), express that, these feelings of cynicism might also occur in relation to colleagues.
The third aspect of burnout, the sense of inefficacy, also called "reduced personal accomplishment," was also defined as the inclination to judge oneself negatively in one's interactions with clients and feeling infelicitous and disgruntled with oneself and one's job (Maslach & Jackson, 1981), and as a condition in which teachers lose their feelings of accomplishment and recognize themselves as no longer making a meaningful contribution through their work and making negative evaluation of themselves (Schwab,1983;Skaalvik & Skaalvik, 2007). It also refers to a decrease in one's feelings of competence and successful achievement in one's job (Schaufeli & Greenglass, 2001). Festinger (1954) introduced social comparison as a crucial source of knowledge about oneself, and as an influencing mechanism of comparison between the self and others that can affect people's judgment, experience, and behavior. It was explained that in the process of comparison individuals usually try to compare themselves against objective, nonsocial standards, but when these standards are absent, individuals will attempt to make social comparison (e.g. Taylor & Lobel, 1989;Wood, 1989) against similar or dissimilar others (Festinger, 1954). Festinger (1954) used the term "a unidirectional drive upward" to suggest the very fact that people attempt to be more capable of their current level of performance and also individuals with whom they actually compare themselves. Taylor and Lobel (1989) reinterprets unidirectional drive upward, and suggest that what Festinger actually meant was that people are more interested in comparing themselves to those whose performance or abilities are slightly better. Wood (1989) expressed that in western cultures; individuals evaluate their abilities and continually try to improve these abilities. When combined with the desire to compare with similar others, individuals seek a point slightly better than that of comparison, others due to the drive upward and the unidirectional drive upward competition bring about.
The basic principal of downward comparison was proposed by Wills (1981), and the core idea of that was the feasibility of increasing the subjective well-being through comparison with less fortunate ones. Wills also emphasized that individuals learn to compare themselves with others who are worse off as a way to ameliorate their well-being. This occurs in situations where there is a decrease in well-being.
Upward comparisons which occur in relation to those acting better than the self, result in negative effect, and those downward comparisons that occur in relation to those acting worse than the self, cause positive effect (Aspinwall & Taylor, 1993). Wills (1981) mentioned that people involve in downward comparison in threatening situations because in this case it helps them to feel better about themselves, their situations and so forth. In addition, Wills focused on two versions of the downward comparison process. In the first version the individual enhances subjective well-being by comparing themselves with a person who is lower or worse than the self and in the second version the wellbeing may be increased in the process of comparing oneself with a person experiencing the same level of problems. Such arguments bring another construct into play.
The construct of personal efficacy was developed by Bandura (1977) and was claimed to have its roots in social learning theory (Ramey-Gassert & Shroyer, 1992). According to the work of Rotter (1966) teacher self-efficacy was expounded as the extent to which teachers believe that they can control the reinforcement of their actions; that is, the reinforcement can be controlled within themselves or it lays in the environment (Brouwers & Tomic, 1998). The second conceptual strand of theory and research grew out of the work of Bandura (1977). Bandura mentioned teacher efficacy as a type of self-efficacy and stated that self-efficacy is a cognitive process in which individuals construct beliefs about their capacity to perform at a given level of attainment (Tschannen-Moran et al., 1998). Teachers' belief in their capacity to affect student performance is called the "sense of efficacy" (Ashton, 1984).This belief of the teacher in his or her own teaching skills and abilities is manifested in his or her teaching behaviors (Ramey-Gassert & Shroyer, 1992). In fact these beliefs refer to what someone believes he or she is able to do under certain conditions and are different from one's capabilities or skills according to Evers, Brouwers, and Tomic (2002). Outcome expectancy beliefs were suggested as the second component of self-efficacy by Bandura. This component is defined in teaching as learning outcomes that are possible to be achieved through teaching or a teacher's belief that effective teaching can influence student learning (Ramey-Gassert & Shroyer, 1992). Friesen and Sarros (1989) scrutinized the sources of burnout among 128 school based administrators and 635 teachers using questionnaires. The findings of the study indicated a positive correlation between work stress and burnout and also work load and both stress and burnout. Friedman and Farber (1992) studied 641 teachers and showed a strong negative correlation between professional satisfaction and burnout. It also indicated that the stronger correlations to burnout were viewed in how teachers perceived themselves rather than how they felt that others perceive them. Additionally, it was suggested that if teachers give themselves credit for even partial educational success, they can avoid burnout. In fact, the study proved the fact that internal factors are important in experiencing burnout and clarified the fact that teachers' feeling about themselves and how they see themselves in society is a crucial factor in burnout studies.

Relation of burnout with other variables
Anderson 1980 conducted a study on 459 classroom teachers in Connecticut. In this study, the relationship between aspects of burnout and perceived need deficiencies were examined. The difference between what the organization should do to foster the needs of individuals such as psychological, security, society, esteem, and self-actualization and what the organization is actually doing was defined as need deficiency. The results of the study indicated that feelings of burnout are more common among teachers whose organization did not increase self-actualization and esteem needs (Schwab, 1983). The findings revealed the need to know and focus on the other side of these factors which are the internal factors that a teacher might experience rather than factors that are controlled externally.
Mclntyre 1981 examined the relationship between teachers' locus of control and aspects of burnout among 469 special educators in Connecticut and Massachusetts. The teachers were divided into two groups of internally and externally oriented according to locus of control and burnout were more common among teachers who were externally oriented (Schwab, 1983).
Schawb and Iwanicki 1982b and Crane 1981 examined three aspects of burnout and their relation with role conflict and role ambiguity among 469 randomly selected teachers from Massachusetts. Crane 1981 also studied three aspects of burnout and their relation with role conflict and role ambiguity among 443 special education teachers who were from eight central cities in Connecticut. According to the findings of these studies the subcategories of role conflict and role ambiguity that are important factors in social life of people and their work place, explained a statistically significant amount of variance in feelings of emotional exhaustion and depersonalization (Schwab, 1983). Schwab and Iwanicki (1982a) classified teachers according to different variables. When the teachers were classified according to their marital status, the place they had worked-urban, suburban, rural-, their degree and the number of years they had taught, no significant difference in teachers' feelings of burnout were reported. On the contrary, there was a significant difference when the teachers were classified according to their gender, level taught, and age. The findings of the study indicated that gender and grade level taught were related to teachers' feelings of depersonalization and female teachers indicated lower negative attitudes toward their students compared to male teachers. High school and middle school/junior school teachers also showed more negative attitudes toward their students than elementary school teachers. The relation between grade level taught and feelings of personal accomplishment were also considered and the findings suggested more frequent feelings of accomplishment among Elementary teachers than high school teachers. The question whether in situations in which teachers share the same comparison tendency, the level taught is still an important factor or not still remains. Russell, Altmaier, and Van Velzen (1987) examined burnout among 600 public school teachers in Iowa and used a questionnaire to find the results. The aim of the study was to consider the effect of job related, stressful events, and social support. The findings of the study indicated that receiving positive feedback concerning skills and abilities and having supportive supervisors create a condition in which teachers are less vulnerable to burnout. Based on these findings the question whether social factors which are from the context and are received from the workplace and society are more important than those coming from the mind was raised. Mazur and Lynch (1989) conducted a study with the aim of considering differential impact of administrative, organizational, and personality factors on teachers' burnout. Two hundred public high school teachers participated in this study and the findings suggested that leadership style was not an important predicator of teacher burnout. It was also reported that organizational stress factors and personality characteristics were significant causes of burnout. Cheuk and Wong (1995) studied the relationship between job-related difficulties and burnout and the effect of social support on burnout. 80 in service teachers at the university of Macau participated in this study and answered a questionnaire. The result of the study indicated a strong relationship among difficulties with fellow teachers, supervisors, students, and parents with burnout and it was indicated that people with whom workers and specifically teachers are in relation play an important role in the experience of burnout. Brouwers and Tomic (1998) researched the direction and time frame of relationships between perceived self-efficacy in classroom management and three dimensions of burnout. 243 secondary school teachers participated in this study. The findings of the study recommended a longitudinal effect of self-efficacy on depersonalization and a synchronous effect on personal accomplishment. Considering the relationship between perceived self-efficacy and emotional exhaustion the time frame was synchronous.
In the study of Evers et al. (2002), 490 randomly selected teachers participated and 3 different questionnaires measuring burnout, self-efficacy and teachers attitudes regarding the effectiveness of the study-home as an educational innovation were used. In this study, the term study-home is defined as a student centered approach in which the aim is to train independent and responsible students who can take the responsibility of their own academic achievement. The analysis of the results recommended that self-efficacy beliefs for each of the three domains were related to depersonalization and exhaustion in a significantly negative way and the three domains were significantly positively related to the personal accomplishment and having more negative attitudes toward the study home system results in suffering more from depersonalization and emotional exhaustion.
Michinov's two cross-sectional studies (Michinov, 2005) on two groups of 72 and 100 police officers aimed to investigate the relationship among social comparison directions, occupational burnout, health complaints, and job burnout. The findings suggested that, those who compared themselves to worse-off employers experienced a lower level of perceived control and those who had frequent upward comparison experienced little emotional exhaustion and high levels of perceived control.
Greenberg, Ashton-James, and Ashkanasy (2007) considered six areas of organization inquiry and studied the role of social comparison processes. The researchers argued that the extent of involvement in social comparisons is influenced by the degree to which assessment of resources can occur objectively due to the absence of comparative information. Further, comparison with an identified individual causes a reduction in stress level as the person relates himself to better ones. While relating oneself to worse-off individuals, the level of stress increases due to a feeling of inadequacy and feeling less qualified. Upward comparison might also result in stress if the decision is based on the perceived capacities. Additionally, under evaluation of one's capabilities as a result of downward comparison, causes stress at the outset, but as the task is accomplished it leads into a reduction in stress level. Lastly, the focal point of the study is the momentary nature of stress-relieving or stressindicating effects of engaging in social comparisons. Skaalvik and Skaalvik's (2007) focused on the relationship among teacher self-efficacy, perceived collective teacher efficacy, external control, strain factors and teacher burnout. The participants of the study were 244 elementary and middle school teachers. The findings of the study suggested teacher efficacy as a multidimensional construct including six separate but correlated dimensions of self-efficacy, including instruction, adapting education to individual students "needs," motivating students, keeping discipline, cooperating with colleagues and parents, and coping with changes and challenges. Besides, the findings suggested a strong relationship between teacher self-efficacy, collective teacher efficacy and teacher burnout. Skaalvik and Skaalvik (2010) examined the relationship between teachers' perception of the school context, teacher self-efficacy, collective teacher efficacy, teacher burnout, teacher job satisfaction, and teachers' beliefs that factors external to teaching puts limitations to what they can accomplish. About 2249 Norwegian Elementary and middle school teachers participated in this study. According to the findings of the study, the variables of teacher self-efficacy, collective efficacy, and two dimensions of burnout were differently related, both to school context variables and to job satisfaction.
In conclusion, as the studies reported were mainly concerned with aspects of burnout, the current study, therefore, was planned to raise consciousness about other social and psychological constructs which might have direct correlation with it. The constructs which are related to both teachers and the people with whom they have interaction. To this end, the researchers studied experienced teachers from different organizations as a vulnerable group to burnout in order to find out the relationship between the directions of their social comparison tendencies in relation to their colleagues with their sense of instructional self-efficacy, and identify the impact of these internal feelings on their experience of burnout.

Methodology
The following section presents the participants, instruments, data collection, and data analysis procedures of the study.

Participants
The researcher employed non-probability convenience sampling and allowed as many language teachers who were accessible to participate in the study. This was achieved through online participation of English language teachers from different cities across the country, as well as offline cooperation of those from the language centers particularly located in this research context-Guilan (N = 170, 109 respectively). The participants varied in their level of education, age, teaching experience, and work place (Tables 1-5). Regarding the level of education, MA holders (N = 151) were the largest group of participants, 119 of the participants held BA degree (N = 119) and the lowest number of the participants were for PhD holders (N = 9). Considering the age of the participants (3.6%) of the total sample that was (N = 10) were lower than 20 years old. The largest group of participants (N = 172) were within the range of 20-30 years old that was about (61.6%) of the total sample. There were also (N = 80) participants within the range of 30-40 years old. That was equal to (29.4%) of the whole participants. Participants older than 40 years old made the last age group (N = 15) which equaled to (5.4%) of the total participants. The total participants of the study were classified into  four groups according to their years of teaching experience. Teachers who had teaching experience lower than two years comprised to (20.4%) of the total sample that was (N = 57). 31.5% of the whole participants had teaching experience between two to five years that was (N = 88). Additionally, greater number of the participants (N = 96) had taught English for 5 to 10 years that was about (34.4%) of the total sample. The last group of participants comprising (13.6%) of the total participants had higher than 10 years experience of teaching English (N = 38). As the last group, teachers were classified into different groups based on their work place. The work place included language centers, universities, or both. The largest group of participants (N = 235) had the experience of teaching English just at language centers and comprised (84.2%) of the total sample. Besides, the smallest group of participants (N = 4) reported that they had the experience of teaching English at universities (1.4%), and the last group were participants who had the experience of teaching English at both language centers and universities (N = 40).

Instrument
The instrument used for this study is a questionnaire which merged three scales, i.e. EFL teachers' burnout, social comparison tendencies, and instructional self-efficacy into one and for this reason is called the Burnout, Social Comparison, Instructional Self-efficacy Scale called by the acronym BSCSE hereafter. The BSCSE-Appendix 1 consisted of three parts and 46 items. Each part of the BSCSE focused on one of the three variables of the study. The first part adopted from the Maslach Burnout Inventory (MBI) measured the burnout level of the teachers. The MBI consisted of 22 items described in the literature as "the most widely used operationalization to burnout" (Maslach & Jackson, 1981). The MBI includes three subscales of Emotional Exhaustion with nine items, Depersonalization with five  items, and Personal Accomplishment with eight items. According to MBI burnout syndrome is defined as (1) high levels of Emotional Exhaustion, (2) high levels of Depersonalization, (3) reduced Personal Accomplishment.
The second part-Items 23-37 of the BSCSE included 11 questions from Iowa-Netherlands Comparison Orientation Measure (INCOM) created by Gibbons and Buunk (1999). However, in the BSCSE and for this research purpose, six of its items were modified on the basis of experts' comments and revision to measure the tendencies of social comparison called upward or downward.
Finally, the third part measured instructional self-efficacy on the basis of Bandura's instrument of teacher self-efficacy scale (1977).

Figure 5. The distributions of the participants in terms of centers they taught English.
addition to the 46 items, the scale required the participants to provide the following factual information: gender, level of education, age, years of teaching experience, and centers in which they taught English.
The instrument was piloted on 30 teachers (25 female, 5 male) in order to get reliability check of the BSCSE and get feedback on the clarity, understanding, and perceived relevance of the items. Cronbach's Alpha (1951) as a popular method to measure internal consistency of the items of a total test or scales of a test (Jung & Goldenson, 2008) was employed to check for the reliability and internal consistency of items. The value of reliability was explained according to the reliability standards suggested by Barker, Pistrang, and Elliott (1994), and according to the value of Cronbach's Alpha for the questionnaire (0.807) the instrument could be considered as the reliable tool for the main study.

Data collection procedure
279 teachers who participated in the study answered to BSCSE that was developed in two versions, hard and soft copies. A cover letter that explained the purpose of the study, confidentiality, and the rights of research participants accompanied the online/soft version. Participants were also clarified about the procedures and the option they had, i.e. they could withdraw from this research at any time they wished. The researcher herself administered the offline/ hard-copy of the questionnaire to illustrate the instructions for teachers to complete the questionnaire and they were ensured of anonymity of their responses. Teachers were also cleared up to respond the BSCSE items according to feelings toward their job and their working experiences.

Data analysis procedure
To answer the research question "Is there any relationship between social comparison tendenciesupward and downward-, the three subscales of burnout, -emotional exhaustion, depersonalization, and personal achievement-and instructional self-efficacy?" Spearman Rank Order Correlation tests were performed using SPSS 22.

Reliability analyses of the BSCSE questionnaire
Items reliability of the questionnaire was measured through a pilot study of the 46 items of the questionnaire on 30 EFL teachers and explanation of the value of the reliability was based on the reliability standards suggested by Barker et al. (1994) (Table 6).
The results of the reliability analysis are presented in Table 7.
The value of Cronbach's Alpha for the questionnaire was (0.807) and it was a good indication of the reliability of the instrument to be used for the main study.

Table 6. Suggested standards for the reliability index
Source: Barker et al. (1994).

Interpretation
Reliability indices

Descriptive statistics for the items of the "exhaustion" section of the questionnaire
The first section of the burnout questionnaire examined the teachers' viewpoints in terms of exhaustion (Table 8). In the first place, the highest mean rank was reported for item (2) that inspected their opinion toward "working with people all day long" (X = 3.98). In comparison to other items of this section the respondents were somehow homogenous in their ratings of this item (SD = 0.90). Secondly, the participants informed that they "felt they worked too hard at their job" (X = 3.24). On the contrary, the lowest mean rank was reported for item (7) that evaluated their prospect toward "feeling like they were at the end of their rope" (X = 1.78). Besides, the highest degree of inconsistency was observed among the responses provided for item (1) that scrutinized their outlook towards "feeling emotionally drained by their work" (SD = 1.21).

Descriptive statistics for the items of the "depersonalization" section of the questionnaire
The second section of the questionnaire included seven items related to "depersonalization" ( Table  9). The highest mean rank was found for item (10) that expressed their perspective toward "having the impression that their students made them responsible for some of their problems" (X = 2.98). In addition, this item had the highest standard deviation among the items of this category (SD = 1.20) that implied greater heterogeneity among the answers provided for this category. In contrast, the lowest mean rank (X = 1.82) and the smallest variation (SD = 1.010) was discovered for item (14) that assessed their ratings of "being afraid that their job was making them uncaring."

Descriptive statistics for the items of the "personal achievement" section of the questionnaire
The third section of the burnout questionnaire comprised of (8) items appraised the respondents' view with respect to their "personal achievement" (Table 10). The respondents rated item (21) higher than the other items of this category (X = 4.20). This item investigated their views towards "being easily able to create a relaxed atmosphere with their students." In comparison, item (16) that measured their attitudes with respect to their "feeling of full of energy" had the lowest mean rank in this section (X = 3.74). Besides, the respondents were highly varied in their ratings of this item (SD = 1.02). However, the highest degree of consistency was reported for item (20) in which the respondents expressed that "Through their work, they felt that they had a positive influence on people." (SD = 0.80).

Descriptive statistics for items of the "social comparison upwards and downwards" section of the questionnaire
Items 23-37 were related to social comparison upwards and downwards (Table 11). In this section, item (32) was found to have the highest mean rank (X = 3.87). The respondents reported "They often liked to talk with others who were higher than them about mutual opinions and experiences." In contrast, they expressed their lowest attitude with respect to item (37) that measured EFL teachers' perceptions of "comparing how they were doing socially with other people who were less important than them." (X = 2.41). Concerning the degree variation with which the respondents provided the answers, item (30) had the highest variation (SD = 1.13). This item evaluated their responses towards "comparing what they had done with how those who were much worse than them had done to find out how well they had done something." On the contrary, they were relatively consistent in their ratings of item (31) that evaluated their viewpoints with respect to "trying to find out what others think, those who face similar problems as I face."

Descriptive statistics for items of the "teacher instructional self-efficacy" section of the questionnaire
Items (38-46) evaluated "teacher instructional self-efficacy" (Table 12). The majority of the participants supported that "they can get students to work together" (X = 4.06). On the other hand, item (38) that inspected the participants' viewpoints toward "influencing the class sizes in their school." received the lowest mean rank in this category (X = 3.30).
While the participants were highly divergent in their perceptions of item (39) that asked about their "ability to get through to the most difficult students" (SD = 0.90), they appeared to be consistent in their ratings of item (46) that measured their ability "to get students to do their homework" (SD = 0.70).

The research question and objective of the study
The current study focuses on burnout in EFL context and studies the effect of the two variables of social comparison tendencies and self-efficacy of teachers on burnout. This study is the first study in Table 10. Item statistics for the items of the "personal achievement" section of the burnout questionnaire Note: The significance of bold values shows the highest and lowest SD and mean of the items.  Iran that considers the effect of one social and one cognitive variable on burnout which is mostly known as a purely social phenomenon.
Researchers have argued that burnout as a social phenomenon is widely spread and increasing number of individuals experience it every year (e.g. Burke, Greenglass, & Schwarzer, 1996;Gold, 2001). As mentioned earlier, existence of burnout among the teachers has its own consequences. According to Brenninkmeijer (2002) the consequences might lead to changes on the individual or organizational level. Brenninkmeijer also introduced the organizational consequences of burnout as turnover and absenteeism. On the other hand, Cordes and Dougherty (1993) specified individual changes as leading to emotional and physical consequences, having effects on health behavior, and relationships of teachers.
Therefore, burnout needs to be studied due to the consequences it has for both individuals and the whole society. In this regard, finding and knowing about the variables influencing the level of burnout is of great importance to the students, teachers, and the whole society.
As Gold (2001) stated the quality of education is seriously affected by the effects of burnout, and the teachers are either leaving their jobs, or they are not satisfied and feel that they are not prepared to handle stress in their job. Consequently, according to Alavinia and Ahmadzadeh (2012) knowing the causes of burnout and all the factors which influence the amount of job burnout will be of significant importance to teacher educators and school administrators in order to provide the teachers with conditions in which they'll feel less stressed.

RQ:
Is there any relationship between social comparison tendencies-upward and downward-, the three subscales of burnout, -Emotional Exhaustion, Depersonalization, and Personal Achievement-, and instructional self-efficacy?
The following null hypothesis was suggested: H0: There is no relationship between social comparison tendencies-upward and downward-, the three subscales of burnout, -Emotional Exhaustion, Depersonalization, and Personal Achievement-and instructional self-efficacy.
The research question dealt with examining the possible relationship between subcategories of social comparison tendencies (i.e. upward and downward), three subscales of burnout (i.e. emotional exhaustion, depersonalization, and personal achievement), and instructional self-efficacy. The main assumption of using Spearman Rank Order Correlation Test was that the data obtained from the Likert-scale questionnaire was ordinal and the association between series of rank order data was going to be computed.
As it is depicted in Table 13, a positive correlation was reported for the relationship between "upward tendencies and personal achievement" (ρ = 0.185, N = 279, p ≤ 0.05). It was found that upward tendencies could help explain (3.42%) of the total variance of personal achievement {(0.185*0.185)*100}.
Correspondingly, negative significant relationship was reported for the connection between "depersonalization and teacher instructional self-efficacy" (ρ = −0.207, N = 279, p ≤ 0.05). This implied Finally, positive significant relationship was found for the relationship between "personal achievement and teacher instructional self-efficacy" (ρ = 0.495, N = 279, p ≤ 0.05). This showed that there was a significant overlap between these two variables and personal achievement could explain (24.50%) of the total variance of teacher instructional self-efficacy {(0.495*0.495)*100}.
The following Table 13 presents the results of Spearman Rank-Order Correlation test: The findings reflected that the highest relationship was found between the degree of exhaustion and depersonalization (ρ = 0.603) closely followed by the relationship between personal achievement and teacher instructional self-efficacy (ρ = 0.495). However, the lowest relationship was found for the relationship between downward tendencies and exhaustion (ρ = 0.123).
The interpretation of the findings in the light of the existing literature will be discussed in the following section.

Discussion and conclusion
According to the research question of the study, the relationship between the three subscales of burnout, instructional self-efficacy, and the tendencies of social comparison was investigated. Social comparison, consisting of upward or downward tendencies might be an influencing factor on the level of experienced burnout among EFL teachers. Exhaustion as the first subcategory of burnout indicated a positive and significant correlation with downward comparison, which means that teachers with more downward comparison experience higher levels of exhaustion in their job. These are the teachers who constantly compare themselves with less fortunate ones or those experiencing similar problems (Wills, 1981), therefore, these teachers feel that they are no longer capable of giving psychologically of themselves (Jennett et al., 2003).
Moreover, the second subcategory of burnout called depersonalization also suggested a significant positive relationship with downward comparison. Teachers with frequent downward comparison isolate themselves from the students and different aspects of their job (Maslach et al., 2001). Contrary to the first and second subcategory, the third subscale of burnout named personal achievement indicted a positive correlation with upward comparison. It can be concluded that teachers who compare themselves to those whose performance or abilities are slightly better (Festinger, 1954) experience higher levels of personal achievement in their job and are less vulnerable to experience burnout. Further, teacher instructional self-efficacy that has a positive significant relationship with personal achievement suggested no significant relationship with upward or downward comparison. It can be concluded that teachers experiencing higher levels of instructional self-efficacy are less vulnerable to experience exhaustion and depersonalization, however these teacher experience higher levels of personal achievement. This group of teachers' assessment of one in relation to others cannot affect their views about controlling a class-instructional self-efficacy.
Taken together, the results of the study suggest that burnout as a psychological and social phenomenon which exists in many professions and is affected by many different factors is also influenced by social comparison tendencies and instructional self-efficacy of teachers in different ways.
These findings are in line with the findings of Russell et al. (1987) that focused on job related social events and social support and revealed the fact that receiving positive feedback on skills and abilities and also receiving support in workplace reduce the level of burnout among teachers. This is the case while, Cheuk and Wong (1995) stressed the same issue by suggesting that difficulties with colleagues, supervisors, students and parents can increase the level of burnout among teachers. The results of the study also suggest that while teachers' views on their ability to influence their students and class is not affected by their upward or downward comparison, an increase in teachers' instructional self-efficacy can guarantee lower level of burnout among EFL teachers as it has a positive and significant relationship with the feeling of personal achievement.
The results of this study highlight policy considerations and possible implications for EFL teachers for preventing burnout among EFL teachers.
Supervisors need to set a safe challenging atmosphere in which the teachers are motivated to compare themselves with those whose abilities or performances are slightly better. This can happen when teachers with better performance or abilities are supported by supervisors. Further, encouraging and reassuring the teachers of their abilities to control and influence their students and their class can also be a part of supervisors' responsibilities which can lead to experiencing lower levels of burnout.

Funding
The authors received no direct funding for this research.