Teaching and Teacher Education

Transcription

1 Teaching and Teacher Education 27 (2011) 1029e1038 Contents lists available at ScienceDirect Teaching and Teacher Education journal homepage: Teacher job satisfaction and motivation to leave the teaching profession: Relations with school context, feeling of belonging, and emotional exhaustion Einar M. Skaalvik *, Sidsel Skaalvik Norwegian University of Science and Technology, Department of Education, 7491 Trondheim, Norway article info abstract Article history: Received 13 October 2010 Received in revised form 30 March 2011 Accepted 1 April 2011 Keywords: Teacher job satisfaction Teacher motivation Belonging Emotional exhaustion School context Values This study examines the relations between school context variables and teachers feeling of belonging, emotional exhaustion, job satisfaction, and motivation to leave the teaching profession. Six aspects of the school context were measured: value consonance, supervisory support, relations with colleagues, relations with parents, time pressure, and discipline problems. The participants were 2569 Norwegian teachers in elementary school and middle school. The data were analyzed by means of SEM analyses. All six school context variables were related to job satisfaction and motivation to leave the teaching profession. These relations were primarily indirect, mediated through feelings of belonging and emotional exhaustion. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction and purposes The shortage of wellqualified teachers is a significant problem in many countries (Ingersoll, 2001; Loeb, DarlingHammond, & Luczak, 2005). This problem is increasing because of teacher attrition, i.e. the departure of teachers from their teaching jobs. For example, 25 percent of beginning teachers in the US leave teaching before their third year, and almost 40 percent leave the profession within the first five years (Chang, 2009). Additionally, the majority of teachers leave the profession before they reach retirement age (McDonald, 1999). The high rate of teacher attrition is reported not only in the U.S., but also in other countries around the world regardless of differences in their educational system, for instance in Australia, China, and England (Hong, 2010). In Norway, where we have conducted the present study, there are no official statistics showing either teacher attrition (teachers leaving the teaching profession) or teacher turnover (teachers moving between schools). However, a study by Schøne (1999) revealed that from November 1995 to November percent of the teachers left their positions, 4.6 percent left the teaching profession whereas 4.2 percent moved to another school. * Corresponding author. Tel.: þ /þ /þ ; fax: þ addresses: einar.skaalvik@svt.ntnu.no (E.M. Skaalvik), sidsel.skaalvik@ svt.ntnu.no (S. Skaalvik). More alarmingly, Köber, Risberg, and Texmon (2005) found that 84 percent of those with a teacher s degree who were under 67, which is the retirement age in Norway, were employed in the autumn of 2003, though only 57 percent were employed in the education sector. Moreover, they reported that 49 percent of all teachers leaving their teaching positions in 2003 went to jobs in noneducational sectors. The attrition rate seems to follow a Ushaped curve, with the highest attrition rate early and late in teachers careers (Borman & Dowling, 2008; Rinke, 2007). Teacher attrition has been studied in relation to working conditions as well as the personal characteristics of teachers, teachers life situation, and teachers roles, responsibilities, and status. Similarly, Day, Sammons, Stobard, Kington, and Gu (2007) discriminate between situated factors, personal factors, and professional factors. In this study, we were concerned with school context variables, which also may be termed working conditions or situated factors. Recent research reveals that working conditions are important predictors of teacher attrition (see a metaanalysis by Borman & Dowling, 2008). A study of teachers who left their teaching positions in England also showed that an excessive work load was the most important factor leading to teacher attrition (Smithers & Robinson, 2003; see also Buchanan, 2010), while a study by Weiss (1999) found that supportive environments were related to teachers motivation to stay in the profession. Although the present study was inspired by the problem of teacher attrition, this article does not examine teacher attrition directly. Instead, we examine assumed antecedents of teacher X/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi: /j.tate

2 1030 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e1038 attrition, namely teacher job satisfaction and motivation to leave the teaching profession. One purpose of the study was to examine relations between school context variables and teacher job satisfaction, as well as the motivation among Norwegian teachers to leave the teaching profession. Understanding the relations between school context variables and teacher job satisfaction is important because the school context or working conditions may be changed and improved. Another purpose was to analyze whether the relations between school context variables on the one hand, and teacher job satisfaction and motivation to leave the teaching profession on the other, were, at least partly mediated through teachers feeling of belonging and physical and emotional exhaustion. 2. Theoretical framework 2.1. Teacher job satisfaction and motivation to leave the profession In the research literature, job satisfaction is regarded as the positive or negative evaluative judgments people make about their jobs (Weiss, 2002). For instance, Locke (1976) defined job satisfaction as a pleasurable or positive emotional state resulting from the appraisal of one s job. In accordance with these definitions, we conceptualize teacher job satisfaction as teachers affective reactions to their work or to their teaching role (see Skaalvik & Skaalvik, 2010a; Zembylas & Papanastasiou, 2004). A problem with the research on teacher job satisfaction is that there is no agreement about how to measure the construct (Skaalvik & Skaalvik, 2009, 2010a), which has been studied as both: (a) a facetspecific job satisfaction measuring the extent to which teachers are satisfied with specific aspects of their job, and (b) an overall sense of satisfaction with the job (Moe, Pazzaglia, & Ronconi, 2010; Sargent & Hannum, 2005). Skaalvik & Skaalvik (2010a) point out that a problem with the facetspecific approach is that different circumstances may be important to different teachers. As a result, such measures overlook the fact that the impact of different circumstances on overall job satisfaction is dependent on how important each of the circumstances is to the individual teacher. In this study, we therefore measured teachers overall sense of job satisfaction and analyzed the degree to which teachers perception of school context variables predicted overall job satisfaction. Teacher job satisfaction is closely associated with teacher absenteeism and attrition (e.g., Huberman, 1993; Sargent & Hannum, 2005; Wriqi, 2008; Zembylas & Papanastasiou, 2004). Hence, we also expect that teacher job satisfaction predicts teachers motivation to leave the teaching profession. Teacher job satisfaction and dissatisfaction are influenced by a number of variables. Dinham and Scott (1998) suggested that the sources of job satisfaction and dissatisfaction may be classified into three domains: (a) intrinsic rewards of teaching, (b) factors extrinsic to the school, and (c) schoolbased factors. The intrinsic rewards of teaching concern the actual work of teaching, working with the students, and seeing students learn and develop, all of which are primary motives for becoming a teacher and a main source of satisfaction among teachers (Scott, Stone, & Dinham, 2001). Factors extrinsic to the school include imposed educational change, external evaluation of schools, negative portrayal of teachers in the media, and a decrease in the status of teaching. Schoolbased factors or contextual variables at school may include relations with colleagues, parents, and the school leadership, as well as time pressure, disruptive student behavior, and the values emphasized at the local school. The present study focuses on relations between these schoolbased factors, which we term contextual variables, and how they relate, both directly and indirectly, to teacher job satisfaction. The indirect relations that we hypothesize are that the impact of school context variables on job satisfaction are mediated through teachers feeling of belonging at school and emotional exhaustion Feeling of belonging According to Baumeister and Leary (1995), the need to belong is a fundamental human motivation (see also Deci & Ryan, 2000). To our knowledge, teachers feeling of belonging to the school where they are teaching has not been systematically studied in research on teachers, whereas students sense of belonging to the school has been shown to be positively related to both motivation (Furrer & Skinner, 2003; Goodenow & Grady, 1993), satisfaction, and positive affect (McMahon, Parnes, Keys, & Viola, 2008; Shochet, Dadds, Ham, & Mantague, 2006). Although we lack systematic research on teachers sense of belonging to their school, we expect that similarly to findings in student research, the teachers feeling of belonging is positively related to satisfaction and positive affect. Hence, we expect that belonging is positively related to job satisfaction and negatively related to the motivation to leave the profession. An important question for this study was whether teachers feeling of belonging mediated the relation between social relations and value perception on the one hand and job satisfaction and motivation to leave the teaching profession on the other hand (see discussion of Teachers perception of the school context ) Emotional exhaustion Research in different cultures indicates that school teachers are among those professionals with the highest level of job stress (Stoeber & Rennert, 2008). Most teachers cope successfully with such stress, for instance through active problem solving, social and emotional support from colleagues, reorganizing their teaching situation, cooperating with parents, or changing their teaching strategy. However, burnout may be the endpoint of coping unsuccessfully with chronic stress (Jennett, Harris, & Mesibov, 2003). Emotional exhaustion is the core element of burnout (Maslach, Jackson, & Leiter, 1996) and results from longterm occupational stress, particularly among human service workers, including teachers (Jennett et al., 2003). Emotional exhaustion is characterized by low energy and chronic fatigue (Pines & Aronson, 1988; Schwarzer, Schimtz, & Tang, 2000). Studies in different cultures have found that measures of teacher burnout, including emotional exhaustion, predict both subjective and objective health as well as teachers motivation and job satisfaction. For instance, Hakanen, Bakker, and Schaufeli (2006) demonstrated that emotional exhaustion correlated negatively with selfrated health as well as work ability among Finnish teachers. Also, in a study of teachers in Hong Kong, Leung and Lee (2006) found that the exhaustion dimension of burnout predicted teachers intentions of leaving the profession (see also Jacobsen, Schwab, & Schuler, 1986), and Skaalvik and Skaalvik (2010a) concluded that emotional exhaustion predicted the level of job satisfaction among Norwegian teachers. An important question for the present study was whether emotional exhaustion mediates the relations between workrelated stressors and job satisfaction as well as the motivation to leave the teaching profession (see discussion of Time pressure and Discipline problems ) Teachers perception of the school context We included teachers perception of six school context variables in this study: value consonance, supervisory support, relations with

3 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e colleagues, relations with parents, time pressure, and discipline problems Value consonance According to Sahlberg (2010), teaching is a profession which is typically driven by values, ethical motives or intrinsic motivations. Explicitly or implicitly, teachers set goals for their teaching (Chang, 2009). Moreover, in their daily teaching and classroom management, teachers communicate and represent values. Thus, we may expect teachers to be most comfortable when they can teach in congruence with their own educational beliefs and values, which we refer to as value consonance, whereas teachers may be less comfortable if they have to represent values that are not congruent or consonant with their personal values. Teachers may have different goals, values, and beliefs. They may differ in their beliefs about what goals should be pursued, what content should be emphasized, and what educational means and methods should be used. Even so, the focus in this study is not on which goals the individual teacher sets, what content the teacher thinks should be emphasized, or which educational means and methods the teacher believes are the best or most proper. Our concern is whether the individual teacher feels that his or her goals and values, whatever these goals and values may be, are in congruence with the goals and values emphasized in the school where he or she is teaching. Consequently, in this study we defined value consonance as the degree to which teachers feel that they share the prevailing norms and values at the school where they are teaching, for instance what goals should be pursued, what content should be emphasized, and what educational means and methods should be used. A teacher who feels that the prevailing norms and values of the school are incompatible with her or his own values may experience what Rosenberg (1977, 1979) referred to as contextual dissonance, whereas a teacher who shares the prevailing norms and values of the school may experience contextual consonance. In accordance with Rosenberg (1979) we conceptualize value consonance as a particular type of contextual consonance. Contextual dissonance or consonance may have serious implications for the relationship of the individual to his or her environment. A contextual dissonance may result in a feeling of not belonging, a feeling that one does not fit, that one is out of it, somehow wrong (Rosenberg, 1977, 1979). Hence, we expected value consonance to be positively related to job satisfaction. Based on Rosenberg s analysis, we also expected that value consonance would be positively related to teachers feeling of belonging. Consequently, we expected that the relation between value consonance and job satisfaction would, at least partly, be mediated through teachers feeling of belonging. These relations may be caused by various processes. As pointed out by Rosenberg(1977,p.209), the context is a communications environment which he describes as a pool a fund of messages e some communicated directly, and others informally and unintentionally. In accordance with Rosenberg s general analysis, we therefore expect that teachers who do not share the prevailing norms and values of their school are likely to hear more deprecatory and fewer laudatory comments about themselves and their practices. Although we have defined value consonance in terms of Rosenberg s conceptualization of contextual consonance and dissonance, it may also be conceptualized as a dimension of personeorganization fit (PeO fit) or personeenvironment fit. Kristof (1996) defined a PeO fit as the compatibility between people and organizations that occurs when at least one entity provides what the other needs, or that they share similar characteristics. If teachers goals and values are not compatible with the prevailing norms and values in the school where he or she is teaching, those norms and values may be perceived as a barrier against acting according to one s values and achieving one s goals Social climate and relations with colleagues, parents and school leadership Research also indicates that a positive social climate and social support are positively related to teacher satisfaction and motivation (Day et al., 2007; Scheopner, 2010; US Department of Education, 1997) and negatively related to burnout (Hakanen et al., 2006; Leung & Lee, 2006; Schaufeli & Bakker, 2004). However, this research evidence is not consistent, and in a study of Chinese teachers, Wriqi (2008) found that collegial relations were weakly related to job satisfaction. Emotional support and positive relations with supervisors, colleagues, and parents may also promote a feeling of belonging. Hence, we expect that the relation between positive social relations (with principals, colleagues, and parents) and job satisfaction at least in part, is mediated through a feeling of belonging. Today, teachers are increasingly dependent on cooperating with parents and need positive relations with them. Therefore, experiencing that one is not trusted by the parents, that they are critical, or that cooperating with parents is difficult may be a serious strain on teachers with negative impact on their feeling of belonging. It may increase anxiety, create a feeling that one is not doing a good job, and promote a need for selfprotection Time pressure Recent research has found that teachers experience an increasing number of work assignments and a more hectic workday, thereby resulting in less time for rest and recovery (Skaalvik & Skaalvik, 2010b; Hargreaves, 2003; Lindqvist & Nordänger, 2006). The increasing work load and the hectic workday cannot be easily separated because increasing work assignments result in less time for rest and recovery. We refer to the combined work load and hectic workday as time pressure, which is due to multiple causes such as an increasing demand for documentation and paperwork, more frequent meetings, more frequent communication with parents, the administration and scoring of achievement tests, frequent changes of the curriculum, and participation in a number of school development projects (Skaalvik & Skaalvik, 2010b). According to Scott et al. (2001), this is an international tendency that has resulted in a major decline in teachers overall job satisfaction. Furthermore, time pressure has been found to correlate positively with teacher burnout (Skaalvik & Skaalvik, 2008; Hakanen et al., 2006; Kokkinos, 2007; Peeters & Rutte, 2005; Schaufeli & Bakker, 2004). Skaalvik and Skaalvik (2010a) found that time pressure was particularly strongly related to the emotional exhaustion dimension of burnout. We therefore expected time pressure to be positively related to both job satisfaction and emotional exhaustion. Moreover, we expected that the relation between time pressure and job satisfaction would be partly mediated through emotional exhaustion. We also expected that time pressure, through emotional exhaustion, would be negatively related to motivation to leave the teaching profession Discipline problems The annual Norwegian national student survey reveals that 30 percent of students from fifth to tenth grade in the compulsory school are often disturbed by other students during lessons at school (Skaalvik, Danielsen, & Skaar, 2007). Interviews with 36 teachers from 14 elementary and middle schools in Norway also show that a majority of the teachers experience disruptive student behavior as a problem (Skaalvik & Skaalvik, 2010b). Discipline problems or disruptive student behavior is recognized as a serious workrelated stressor (e.g., Buchanan, 2010), and significant correlations have been found between discipline problems and

4 1032 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e1038 measures of burnout or emotional exhaustion (Skaalvik & Skaalvik, 2007, 2008, 2010a; Hakanen et al., 2006; Kokkinos, 2007; Tsouloupasa, Carson, Matthews, Grawitch, & Barber, 2010). The experiencing of discipline problems may lead to expectations of future problems, which in turn may result in stress, worry, and anxiety. Chang (2009) discusses the association between discipline problems and burnout in terms of the concept of goal congruence. Chang explains that a teacher may set different goals for his or her teaching such as maintaining order, managing students behavior, following lesson plans, and helping students reach learning goals (p. 205). Thus, as suggested by Chang, students disruptive behavior may be a threat to teachers goal achievement. We therefore expect disruptive student behavior or discipline problems to be positively related to emotional exhaustion and negatively related to job satisfaction Theoretical predictions Based on our analysis (see above), we specified a theoretical model (Fig. 1) in which value consonance, supervisory support, and relations with colleagues and parents were predictive of both teachers feeling of belonging and job satisfaction. Furthermore, we expected belonging to be positively related to job satisfaction. Hence, value consonance and social relations were predicted to be both directly and indirectly related to job satisfaction. The indirect relation was expected to be mediated through a feeling of belonging. The theoretical model also specified that time pressure and discipline problems are predictive of job satisfaction, both directly and indirectly. The indirect relation is predicted to be mediated through emotional exhaustion. Moreover, we expected that belonging, emotional exhaustion, and job satisfaction would be predictive of a motivation to leave the teaching profession. 3. Method 3.1. Norwegian public school In Norway, children are enrolled in elementary school in the calendar year when they become six years old. Elementary school lasts for seven years, from the 1st through 7th grades, and middle school lasts for three years, from the 8th through 10th grades. Both elementary and middle school are compulsory and follow a national curriculum which states the goals and content of the education, whereas the choice of instructional methods is left to the school district, the local school, or the individual teacher. Students in elementary school are given verbal and informal evaluation, while students in the middle school are graded. Elementary school teachers are normally educated through a professional teachertraining program that lasts for four years, in which they study education along with elementary school disciplines. In comparison, the majority of middle school teachers are first educated through a threeyear program not specifically oriented towards teaching. If they decide to go into teaching, they then follow a oneyear teacher education program Participants and procedure A total of 2569 teachers from 127 Norwegian elementary and middle schools (1ste10th grades) participated in this study. Norway was divided into five geographical regions. From each region about 25 schools were drawn from one city, two towns and two rural areas by a stratified random procedure. The first contact with each school was made with the school principal, and the only question put to the principal was whether he or she would agree to let data be collected at the school. Only two schools had to be replaced by other schools from the same region because of the principals not agreeing to the data collection. The next step was to contact the teachers representative at each school. The teachers representative informed the teachers about the data collection, that the purpose of the study was to explore working conditions for the teachers, and that the participation was anonymous and voluntary for the individual teachers. At that point, the decision to participate was made by the teaching staff at each school. The teachers representative also arranged for a specific period of time (60 min) to be set aside for teachers to simultaneously respond to the questionnaire. The data collection was administered in FebruaryeMarch of 2010 by two trained research assistants visiting the Value consonance Supervisory support Belonging Job satisfaction Relations with colleagues Relations with parents Time pressure Emotional exhaustion Motivation to leave Disciplin problems Fig. 1. Theoretical model of relations between the study variables.

5 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e schools and bringing the questionnaires back. Eightyfour percent of the teachers at the selected schools participated in the study. The sample consisted of 72 percent females, and the age of the teachers varied from 23 to 69. The mean age was 45, and the average number of years in the teaching profession was 16. The schools varied with respect to size from schools with five teachers to schools with 82 teachers, with the average being 38. The average number of students in the schools was 370. Sixtythree percent of the teachers taught at the elementary level (grades 1e7); whereas 37 percent taught at the middle school level (grades 8e10) Instruments Value consonance Value consonance was measured by use of a threeitem scale. It is important to note that the items referred to values, opinions, and good teaching in general, and that they did not refer to any specific values or opinions. The items referred to the school or the organization, in addition to the entire teaching staff, which is in accordance with Kristof s (1996) definition of personeorganization fit. The items were: My educational values are in accordance with the values which are emphasized at this school, My colleagues and I have the same opinion about what is important in education, and I feel that this school shares my view of what constitutes good teaching. Cronbach s alpha for the scale was.84. Cronbach s alpha is a test conducted to reveal internal consistency, and values greater than.70 are traditionally considered adequate (Cortina, 1993) Supervisory support Supervisory support was measured by a threeitem scale that measured teachers experiences of receiving cognitive and emotional support from the school leadership. The items were: In educational matters, I can always seek help and advice from the school leadership, My relationship with the principal is one of mutual trust and respect, and The school leadership is supportive and praises good work. Cronbach s alpha for the scale was Relations with colleagues Teachers relations with their colleagues were measured by the following three items: In educational matters, I can always get good help from my colleagues, The relations among the colleagues at this school are characterized by friendliness and a concern for each other, and Teachers at this school help and support each other. Cronbach s alpha for the scale was Relations with parents Teachers relations with parents were measured by means of a threeitem scale measuring the teachers feeling of being trusted by the parents. The items were: I feel that the parents have faith in my teaching, The parents are easy to work with, and The parents trust and accept my decisions. Cronbach s alpha for the scale was Time pressure Time pressure was measured by a threeitem scale. The items were: Preparation for teaching must often be done after working hours, Life at school is hectic and there is no time for rest and recovery, and Meetings, administrative work, and documentation take much of the time that should be used for teaching preparation. Cronbach s alpha for the scale was Discipline problems and disruptive student behavior Discipline problems were measured by the following three items: My teaching is often disrupted by students who lack discipline, Some students with behavioral problems make it difficult to carry out lessons as planned, and Controlling students behavior takes a lot of time and effort. Cronbach s alpha for the scale was Feeling of belonging The teachers feeling of belonging was measured by a threeitem scale measuring their general feeling of belonging to the school, their feeling of being accepted by their school s leadership, and their feeling of being trusted by their colleagues. Translated into English, the items were: I feel that I belong to this school, I feel that I am accepted by my school s leadership, and I feel that my colleagues have faith in me. Cronbach s alpha for the scale was Emotional exhaustion Emotional exhaustion was measured by a short sixitem modified version of the emotional exhaustion dimension of the Maslach Burnout Inventory e Educators Survey (MBI: Maslach et al., 1996). The items were drawn from a Norwegian version of the MBI. The participants rated statements indicating that their work made them feel emotionally drained or exhausted. The short sixitem version has previously been shown to have a Cronbach s alpha of.88 (Skaalvik & Skaalvik, 2010a). Cronbach s alpha in this study was Job satisfaction The teachers overall job satisfaction was measured by means of a fouritem scale. The items were: I enjoy working as a teacher, I look forward to going to school every day, Working as a teacher is extremely rewarding, and When I get up in the morning, I look forward to going to work. Cronbach s alpha for the scale was Motivation to leave the teaching profession The teachers motivation for leaving the teaching profession was measured by a threeitem scale. The items were: I wish I had a different job to being a teacher, If I could choose over again I would not be a teacher, and I often think of leaving the teaching profession. Cronbach s alpha for the scale was.92. Responses on all scales were given on a sixpoint scale from Completely disagree (1) to Completely agree (6). In the questionnaire, the items measuring school context variables were mixed and placed as the first block of items followed by scales measuring value consonance, belonging, exhaustion, job satisfaction, and motivation to leave the profession Data analysis We tested the theoretical model (see Fig. 1) by means of SEM analyses (structural equation modeling) using the AMOS 7 program to explore relations between the variables. SEM is a statistical methodology that takes a confirmatory approach to the analysis (Byrne, 2001). In this approach, a hypothesized model of relations between the variables is statistically tested to determine the extent to which it is consistent with the data, which is referred to as the goodness of fit. If the goodness of fit is adequate, it supports the plausibility of the relations between the variables. In order to assess the model fit, we used wellestablished indices such as CFI, IFI, TLI, and RMSEA, as well as the chisquare test statistics. For the CFI, IFI, and TLI indices, values greater than.90 are typically considered acceptable and values greater than.95 indicate a good fit to the data (Bollen, 1989; Byrne, 2001; Hu & Bentler, 1999). For wellspecified models, an RMSEA of.06 or less reflects a good fit (Hu & Bentler, 1999). We first tested two measurement models by means of confirmatory factor analysis using the AMOS 7 program. The purpose of the measurement models is to test that the factors specified are

6 1034 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e1038 independent constructs as well as the correlation between the factors. The first measurement model specified the six school context variables as separate constructs whereas the second measurement model specified belonging, emotional exhaustion, job satisfaction, and motivation to leave the teaching profession as four separate constructs. We then tested the theoretical model of relations between the constructs (Fig. 1). 4. Results Table 1 shows the zero order correlations between the study variables as well as statistical means and standard deviations. Job satisfaction was positively related to value consonance, supervisory support, relations with colleagues, relations with parents, and belonging, whereas it was negatively related to time pressure, discipline problems, emotional exhaustion, and motivation to leave the teaching profession. Teachers feeling of belonging was positively and moderately to strongly related to value consonance, supervisory support, and positive relations with colleagues. Belonging was also positively but weakly related to teachers relations with parents. With the exception of moderate to strong relations between value consonance, supervisory support, and positive relations with colleagues (r ¼.49,.50, and.50, respectively) we found only weak correlations between the six school context variables. We first tested a measurement model with six latent school context variables: value consonance, supervisory support, relations with colleagues, relations with parents, time pressure, and discipline problems. Each of these latent variables was indicated by three to five items as described in the Methods section. The model had good fit to the data (c 2 (155, N ¼ 2569) ¼ , CFI ¼.957, IFI ¼.957, TLI ¼.942 and RMSEA ¼.055). We then tested a second measurement model with four latent variables: feeling of belonging, emotional exhaustion, job satisfaction, and motivation to leave the teaching profession. Each of these latent variables was indicated by three to six items as described in the Methods section. The model had acceptable fit to the data (c 2 (98, N ¼ 2569) ¼ , CFI ¼.943, IFI ¼.943, TLI ¼.921 and RMSEA ¼.080). The relations between the study variables were further analyzed by means of SEM analysis using the AMOS 7 program. In the initial analysis, we tested the theoretical model described in the Methods section. We then deleted nonsignificant paths one by one, starting with the path with the lowest regression weight. The final empirical model, showing standardized regression weights, is displayed in Fig. 2. The model had acceptable fit to the data (c 2 (566, N ¼ 2569) ¼ 4500,601, CFI ¼.934, IFI ¼.934, TLI ¼.922 and RMSEA ¼.052). The c 2 /df ratio was equal to 7.952, which is high compared to Kline s (1998) rule of values of less than 3 being considered adequate. However, the c 2 /df ratio is sensitive to the number of observations. Testing the model on a smaller subsample from the southeast region of the country (N ¼ 265) revealed the same pattern of results, but with an c 2 /df ratio equal to In accordance with our expectations, value consonance, supervisory support, relations with colleagues, and relations with parents were all significant and independent predictors of teachers feeling of belonging. Discipline problems and time pressure were significant predictors of emotional exhaustion. Both feeling of belonging and emotional exhaustion were significantly related to job satisfaction. Motivation to leave the teaching profession was negatively related to job satisfaction and positively related to emotional exhaustion. We also explored additional direct paths to the final model from value consonance and social relations to emotional exhaustion, and from time pressure and discipline problems to belonging. None of these paths were statistically significant at the.05 level. Contrary to expectations, only two school context variables, relations with parents and time pressure, were directly, but weakly related to job satisfaction. However, as perceived by the teachers, all school context variables were indirectly related to job satisfaction. Value consonance, supervisory support, relations with colleagues, and relations with parents were positively related to job satisfaction through feeling of belonging. Moreover, time pressure and discipline problems were negatively related to job satisfaction through emotional exhaustion. Also, contrary to expectations, teachers feeling of belonging was not directly related to motivation to leave the profession. The relation between belonging and motivation to leave the profession was mediated through job satisfaction and emotional exhaustion. As described above, the empirical model reveals several indirect relations. If interpreted in causal terms, an indirect effect implies a hypothesis that an independent variable causes a mediating variable which in turn causes a dependent variable (MacKinnon, Lockwood, & Williams, 2004; Sobel, 1990). For example, teachers feeling of belonging mediates the relations between value consonance, supervisory support, relations with colleagues, and relations with parents, on the one hand, and job satisfaction on the other. In comparison, the relations between discipline problems and time pressure, on the one hand, and job satisfaction and motivation to leave the profession, on the other, were mediated through emotional exhaustion. Both belonging and emotional exhaustion were also indirectly related to motivation to leave the teaching profession, mediated through job satisfaction. We also tested the model for eight different subgroups: older teachers, younger teachers, males, females, elementary school teachers, middle school teachers, teachers at city schools and teachers at rural schools. All these models demonstrated the same pattern of results and had the similar significant regression Table 1 Zero order correlations and descriptive statistics. Study variables Value consonance e Supervisory support e Relations with colleagues e Relations with parents e Time pressure e Discipline problems e Belonging e Emotional exhaustion e Job satisfaction e Motivation to leave e M SD

7 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e Value consonance.30 R 2 =.53 R 2 =.45 Supervisory support Relations with colleagues Relations with parents Belonging Job satisfaction.53 Time pressure.48 Emotional exhaustion.27 Motivation to leave Disciplin problems.21 R 2 =.39 R 2 =.52 Fig. 2. Structural model of relations between teachers perception of six school context variables, teachers feeling of belonging, emotional exhaustion, job satisfaction, and motivation to leave the teaching profession. Standardized regression weights reported. Only significant regression weights are included in the figure. weights as the model shown in Fig. 2. Additionally, all models had similar fit to the data; for instance RMSEAs for the eight models were.053,.050,.054,.051,.051,.052,.053, and.051, respectively. Thus, the pattern of relations between the study variables was very similar for different subgroups of teachers. Table 2 shows the statistical means and standard deviations for different subgroups of teachers according to age, gender, location, and school level. We used Cohen s d to estimate effect size (Cohen, 1988). Cohen s d is a measure of the difference in statistical means between two groups in terms of standard deviation. According to Cohen (1988) a difference that is lower than.2 standard deviation is considered not to be significant. A difference between.2 and.5 standard deviations is considered significant, but small whereas a difference larger than.8 standard deviation is considered large. Table 2 also shows that there were few and only minor differences between the subgroups. Still, we found small tendencies toward older teachers reporting more positive relations with parents, less discipline problems, and higher job satisfaction than did younger teachers, whereas younger teachers reported more positive relations with their colleagues. Regarding Table 2, we should note that younger teachers are defined as those aged 35 years or younger and older teachers are defined as being over the age of 50. No significant differences were found between either of these two groups and teachers between 35 and 50 years. Table 2 also shows some small differences between male and female teachers. Compared to males, females reported higher value consonance, better relations with colleagues, higher job satisfaction, and lower motivation to leave the teaching profession. Nonetheless, they also reported higher time pressure. We found no significant differences between teachers at urban and rural schools, though compared to teachers in middle school, teachers in elementary school reported higher value consonance Table 2 Means and standard deviations for different subgrops of teachers. Variables Younger Older d Female Male d Urban Rural d Elem Middle d <35 >50 Value Consonance * * (2.46) (2.44) (2.42) (2.37) (2.41) (2.47) (2.39) (2.39) Supervisory support * (3.32) (3.33) (3.36) (3.28) (3.33 (3.36) (3.17) (3.41) Relations colleagues * * (2.28) (2.45) (2.46) (2.26) (2.42 (2.42) (2.46) (2.33) Relations parents * (2.30) (2.15) (2.22) (2.18) (2.19) (2.28) (2.22) (2.21) Time pressure * (3.14) (3.11) (3.98) (3.75) (3.94) (3.86) (3.98) (3.79) Discipline problems * (3.78) (3.79) (3.98) (3.75) (3.94) (3.86) (3.98) (3.79) Belonging (2.48) (2.53) (2.48) (2.53) (2.55) (2.43) (2.51) (2.47) Exhaustion (7.03) (6.96) (7.06) (7.03) (6.97) (7.18) (7.02) (7.09) Job satisfaction * * (4.25) (3.69) (4.06) (3.82) (3.93 (3.91) (3.86) (4.01) Motivation to leave * (4.18) (3.89) (4.13) (4.00) (4.15 (4.06) (4.12) (4.02) Note. d ¼ effect size (Cohen, 1988). * = Cohen s d between 0.2 and 0.5 indicating small but significant differences.

8 1036 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e1038 and more positive relations with their school leadership (supervisory support). 5. Discussion In this study, we examined the relations between teachers perception of six school context variables and their feeling of belonging, emotional exhaustion, job satisfaction, and motivation to leave the teaching profession. As expected, teachers feeling of value consonance as well as supervisory support and positive relations with colleagues and parents were predictive of belonging, whereas time pressure and discipline problems were predictive of emotional exhaustion. Also, in accordance with expectations, both teachers feeling of belonging and emotional exhaustion were predictive of job satisfaction, while emotional exhaustion and job satisfaction were predictive of motivation to leave the teaching profession. The positive relation between value consonance and teachers feeling of belonging supports the theoretical analysis of contextual consonance and dissonance, and raises important questions which should be addressed in future research. To the best of our knowledge, the construct of value consonance and its relation to teachers feeling of belonging has not been previously studied in teacher research. Value consonance in this study was based on Rosenberg s (1977, 1979) theory of contextual consonance and dissonance. According to Rosenberg, contextual dissonance may result in a feeling of not belonging or a feeling that one does not fit in. This notion was supported in the present study, in which we focused on value context; the degree to which teachers feel that they share the prevailing norms and values of the school where they teach. Future research should analyze the processes through which value consonance relates to teachers feeling of belonging. One possibility is that the sheer feeling of sharing goals and values with colleagues and the school administration increases a feeling of belonging in itself. Such sharing of goals and values may increase teachers beliefs in what they do and make them feel as part of a team. Another possible explanation may be that not sharing the predominant goals and values at the school is perceived as a barrier against teaching in accordance with one s own goals and values, which may also result in a reduced feeling of belonging. This explanation may be of particular importance in Norwegian schools because teachers often work in teams and may therefore have to compromise their personal ideas about teaching. An important task for future research is also to explore means to establish shared goals and values among the teaching staff. We particularly call for analysis of the impact of the school leadership or management on the school climate and such variables as teachers engagement, feeling of value consonance, and identification with schoolbased norms and values. This task calls for a variety of approaches, including qualitative case studies and experimental studies. The finding that supervisory support and relations with colleagues and parents were predictive of belonging supports the notion that relatedness is a fundamental human need (Deci & Ryan, 2000). Although this result was expected, it is interesting to note that the teachers relations to principals, colleagues, and parents were independent predictors of the teachers feeling of belonging. This indicates the importance of creating an atmosphere of mutual trust and respect among all groups and interested parties in a school. The finding that time pressure and discipline problems were predictive of emotional exhaustion is also in accordance with theoretical expectations. Emotional exhaustion, which is the core element of burnout (Maslach et al., 1996), results from longterm occupational stress (Jennett et al., 2003). Such stress may be caused by both time pressure and discipline problems. Previous research has found that the school context or working conditions are predictive of teachers job satisfaction. In particular, job satisfaction has been shown to be positively related to social support and social climate (US Department of Education, 1997). However, contrary to expectations, we found no strong and direct relation between teachers perception of the school context and job satisfaction. Although the zero order correlations (see Table 1) demonstrated that all school context variables were significantly related to job satisfaction, the SEM analysis revealed that only two of these constructs, relations with parents and time pressure, were directly though only weakly related to job satisfaction. However, the analysis revealed that all school context variables were indirectly related to job satisfaction, mediated through belonging or emotional exhaustion. It is important to note that two groups of school context variables were related to job satisfaction through different processes. Social relations with colleagues, parents and the school leadership were related to job satisfaction through the feeling of belonging. Value consonance, the degree to which teachers feel that they share the prevailing norms and values at the school where they are teaching, were also related to job satisfaction through belonging. None of these school context variables was directly related to emotional exhaustion. In comparison, time pressure and discipline problems were related to job satisfaction through emotional exhaustion. These variables were not significantly related to belonging. As explained above, an indirect effect implies a hypothesis that an independent variable causes a mediating variable which in turn causes a dependent variable (MacKinnon et al., 2004; Sobel, 1990). For example, based on the analysis of this study, we may assume that value consonance and positive social relations result in a stronger feeling of belonging and that this stronger feeling of belonging in turn results in higher job satisfaction. Following the same reasoning, we assume that time pressure and discipline problems increase emotional exhaustion, which in turn reduces teachers job satisfaction. We had expected that belonging and emotional exhaustion would partially mediate the relation between teachers perception of the school context and their job satisfaction (see Fig. 1). Even so, an important finding in this study is that these relations were almost entirely mediated through belonging and emotional exhaustion. Thus, our analysis suggests that belonging and emotional exhaustion are key variables in mediating the impact of school context variables on job satisfaction and motivation to leave the teaching profession. Teachers feeling of belonging was also negatively related to their motivation to leave the teaching profession. The zero order correlation between these variables was.32. Hence, an interesting result of the SEM analysis was that there was no significant direct relation between belonging and motivation to leave the profession. The relation between these constructs was mediated through job satisfaction and emotional exhaustion. It is especially important to note the negative relation between belonging and emotional exhaustion. Though the present study does not prove causal directions, a possible interpretation is that belonging works as a barrier against exhaustion. Such an interpretation is consistent with the notion of belonging is a fundamental psychological need. A lack of belonging may therefore be perceived as a psychological strain that may cause emotional exhaustion. Emotional exhaustion was both directly and indirectly related to motivation to leave the profession. The indirect relation was mediated through job satisfaction. These results indicate that job satisfaction is a key variable for mediating the effect of belonging and exhaustion on the motivation to leave the profession. One implication of this study is that school administrators should pay more attention to teachers feeling of belonging, emotional exhaustion, and job satisfaction. In the SEM model, the constructs of exhaustion and job satisfaction are related to

9 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e belonging and explain 52 percent of the variance in teachers motivation to leave the teaching profession. Moreover, belonging and exhaustion are together strongly related to various aspects of the school context. In order to increase the feeling of belonging and job satisfaction and decrease emotional exhaustion, it seems important to create a supportive school environment, to clarify and develop mutual goals and values, to reduce time pressure on teachers, and to establish schoolbased directions for student behavior. Even though the six school context variables that were analyzed in this study explained 53 percent of the variances in teachers feeling of belonging and 39 percent of the variance in emotional exhaustion, much of the variance in these variables is not explained in this study. Future research should therefore include other school context variables as well as the personal characteristics of teachers, teachers life situation, and teachers status within society. This study demonstrates the association between teachers perception of selected variables in the school context and their job satisfaction as well as their motivation to leave the teaching profession. Though firm conclusions about causality cannot be drawn from this study, it indicates that the impact of teachers perception of the school context on their job satisfaction is an indirect one, mediated through teachers feeling of belonging and emotional exhaustion. Yet, the study has several limitations. Firstly, in this study we did not examine the reasons for actually leaving the teaching profession. We merely asked about the motivation for leaving the profession among teacher who had not yet left the profession. Still, the results are important for two main reasons: (a) for a number of teachers, the lack of satisfaction and a growing motivation to leave the profession may result in actually leaving the teaching profession, and (b) a lack of satisfaction and a growing motivation to leave the profession is likely to affect teachers work engagement, how they relate to students and colleagues and hence the quality of their teaching. Secondly, we should also note that although this study indicates that belonging and exhaustion are important variables mediating the impact of school context variables on job satisfaction and motivation to leave the teaching profession, these variables may not be the only mediating variables. Future research should also explore other possible mediating variables. Thirdly, we examined six school context variables and the analysis showed that all six variables were independent predictors of either belonging or exhaustion. Other school context variables should be examined in future research, both variables which are directly related to job satisfaction and motivation to leave the teaching profession and variables which are indirectly related to job satisfaction, for instance through belonging and exhaustion. Fourthly, it should be noted that several variables in this study were measured with short threeitem scales. Though these scales had a satisfactory reliability, it is important to test our results with more extensive and wideranging scales. More specifically, our measure of belonging includes items that capture teachers feeling of being accepted by the school leadership and trusted by their colleagues. Because the content of these items are close to our conceptualization of belonging, more research is needed using scales which discriminate more clearly between these constructs. This research was supported by a grant from the Research Council of Norway. References Baumeister, R. F., & Leary, M. R. (1995). The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497e529. Bollen, K. A. (1989). A new incremental fit index for general structural models. Sociological Methods & Research, 17, 303e316. Borman, G. D., & Dowling, N. M. (2008). Teacher attrition and retention: a metaanalytic and narrative review of the research. Review of Educational Research, 78, 367e409. Buchanan, J. (2010). May I be excused? Why teachers leave the profession. Asia Pacific Journal of Education, 30, 199e211. Byrne, B. M. (2001). Structural equation modelling with AMOS. Basic concepts, applications, and programming. Mahwah, N.J.: Lawrence Erlbaum Ass. Chang, M.L. (2009). An appraisal perspective of teacher burnout: examining the emotional work of teachers. Educational Psychology Review, 21, 193e218. Cohen, C. (1988). Statistical power in the behavioral sciences (2nd ed.). Hillsdale, N.J.: Lawrence Erlbaum Ass. Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98e104. Day, C., Sammons, P., Stobard, G., Kington, A., & Gu, Q. (2007). Teachers matter: Connecting work, lives and effectiveness. Berkshire, England: Open University Press. Deci, E. L., & Ryan, R. M. (2000). The what and why of goal pursuits. Human needs and the selfdetermination of behaviour. Psychological Inquiry, 11, 227e268. Dinham, S., & Scott, C. (1998). A three domain model of teacher and school executive career satisfaction. Journal of Educational Administration, 36, 362e378. Furrer, C., & Skinner, E. A. (2003). Sense of relatedness as a factor in children s academic engagement and performance. Journal of Educational Psychology, 95, 148e162. Goodenow, C., & Grady, K. E. (1993). The relationship of school belonging and friends values to academic motivation among urban adolescent students. Journal of Experimental Education, 62, 60e71. Hakanen, J. J., Bakker, A. B., & Schaufeli, W. B. (2006). Burnout and work engagement among teachers. Journal of School Psychology, 43, 495e513. Hargreaves, A. (2003). Teaching in the knowledge society: Education in the age of insecurity. Milton Keynes: Open University Press. Hong, J. Y. (2010). Preservice and beginning teachers professional identity and its relation to dropping out of the profession. Teaching and Teacher Education, 26, 1530e1543. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1e55. Huberman, M. (1993). The lives of teachers. London: Cassell. Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: an organizational analysis. American Educational Research Journal, 38, 499e534. Jacobsen, S. E., Schwab, R. L., & Schuler, R. S. (1986). Toward an understanding of the burnout phenomenon. Journal of Applied Psychology, 71, 639e640. Jennett, H. K., Harris, S. L., & Mesibov, G. B. (2003). Commitment to philosophy, teacher efficacy, and burnout among teachers of children with autism. Journal of Autism and Developmental Disorders, 33, 583e593. Kline, R. B. (1998). Principals and practice of structural equation modelling. New York: Guilford. Kokkinos, C. M. (2007). Job stressors, personality and burnout in primary school teachers. British Journal of Educational Psychology, 77, 229e243. Kristof, A. L. (1996). Personeorganization fit: an integrative review of its conceptualizations, measurement and implications. Personnel Psychology, 49, 1e49. Köber, T., Risberg, T., & Texmon, T. (2005). Hvor jobber førskolelærere og lærere?. In Utdanning 2005 (pp. 232e253) Oslo: Statistics Norway. emner/04/sa_utdanning/arkiv/sa74/kap11.pdf. Leung, D. Y. P., & Lee, W. W. S. (2006). Predicting intention to quit among Chinese teachers: differential predictability of the component of burnout. Anxiety, Stress, & Coping, 19, 129e141. Lindqvist, P., & Nordänger, U. K. (2006). Who dares to disconnect in the age of uncertainty? Teachers recesses and offtheclock work. Teachers and Teaching: Theory and Practice, 12, 623e637. Locke, E. A. (1976). The nature and causes of job satisfaction. In M. Dunette (Ed.), Handbook of industrial and organizational pshychology (pp. 1297e1349). Chigago: RandMcNally. Loeb, S., SarlingHammond, L., & Luczak, J. (2005). How teaching conditions predict teacher turnover in California schools. Peabody Journal of Education, 80, 44e70. McDonald, D. (1999). Teacher attrition: a review of literature. Teaching and Teacher Education, 15, 835e848. MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for the indirect effect: distribution of the product and resampling methods. Multivariate Behavioral Research, 39, 99e128. McMahon, S. D., Parnes, A. L., Keys, C. B., & Viola, J. J. (2008). School belonging among lowincome urban youth with disabilities: testing a theoretical model. Psychology in the Schools, 45, 387e401. Maslach, C., Jackson, S. E., & Leiter, M. P. (1996). Maslach Burnout Inventory manual (3rd ed.). Mountain View, California: CPP, Inc. Moe, A., Pazzaglia, F., & Ronconi, L. (2010). When being able is not enough. The combined value of positive affect and selfefficacy for job satisfaction in teaching. Teaching and Teacher Education, 26, 1145e1153. Peeters, M. A. G., & Rutte, C. G. (2005). Time management behaviour as a moderator for the job demandecontrol interaction. Journal of Occupational Health Psychology, 10, 64e75. Pines, A. M., & Aronson, E. (1988). Career burnout: Causes and cures. New York: Free Press. Rinke, C. (2007). Understanding teachers careers: linking professional life to professional path. Educational Research Review, 3, 1e13. Rosenberg, M. (1977). Contextual dissonance effects: nature and causes. Psychiatry, 40, 205e217.

10 1038 E.M. Skaalvik, S. Skaalvik / Teaching and Teacher Education 27 (2011) 1029e1038 Rosenberg, M. (1979). Conceiving the self. New York: Basic Books. Sahlberg, P. (2010). Rethinking accountability in a knowledge society. Journal of Educational Change, 11, 45e61. Sargent, T., & Hannum, E. (2005). Keeping teachers happy: job satisfaction among primary school teachers in rural northwest China. Comparative Education Review, 49, 173e204. Schaufeli, W. B., & Bakker, A. D. B. (2004). Job demands, job resources, and their relationship with burnout and engagement: a multisample study. Journal of Organizational Behavior, 25, 293e315. Scheopner, A. (2010). Irreconcilable differences: teacher attrition in public and catholic schools. Educational Research Review, 5, 261e277. Schøne, P. (1999). Avgang og rekruttering i undervisningssektoren e betydningen av lønn. Oslo, Norway: Institutt for samfunnsforskning. Schwarzer, R., Schmitz, G. S., & Tang, C. (2000). Teacher burnout in Hong Kong and Germany: a crosscultural validation of the Maslach Burnout Inventory. Anxiety, Stress & Coping, 13(3), 309e326. Scott, C., Stone, B., & Dinham, S. (2001). I love teaching but.. International patterns of teaching discontent. Education Policy Analysis Archives, 9(number 28), 1e18, Available at Shochet, I. M., Dadds, M. R., Ham, D., & Montague, R. (2006). School connectedness is an underemphasized parameter in adolescent mental health: results of a community prediction study. Journal of Clinical Child and Adolescent Psychology, 35, 170e179. Skaalvik, E. M., Danielsen, I. J., & Skaar, K. (2007). De viktige få. Analyse av Elevundersøkelsen Kristiansand, Norway: Oxford Research. Skaalvik, E. M., & Skaalvik, S. (2007). Dimensions of teacher selfefficacy and relations with strain factors, perceived collective teacher efficacy, and teacher burnout. Journal of Educational Psychology, 99, 611e625. Skaalvik, E. M., & Skaalvik, S. (2008). Teacher selfefficacy: conceptual analysis and relations with teacher burnout and perceived school context. In H. W. Marsh, R. G. Craven, & D. M. McInerney (Eds.), Selfprocesses, learning, and enabling human potential (pp. 223e247). Connecticut: Information Age Publishing. Skaalvik, E. M., & Skaalvik, S. (2009). Does school context matter? Relations with teacher burnout and job satisfaction. Teaching and Teacher Education, 25, 518e524. Skaalvik, E. M., & Skaalvik, S. (2010a). Teacher selfefficacy and teacher burnout: a study of relations. Teaching and Teacher Education, 26, 1059e1069. Skaalvik, E. M., & Skaalvik, S. (2010b). Utfordringer i lærerrollen. In R. A. Andreassen, E. J. Irgens, & E. M. Skaalvik (Eds.), Kompetent skoleledelse (pp. 147e163). Trondheim, Norway: Tapir akademisk forlag. Smithers, A., & Robinson, P. (2003). Factors affecting teachers decisions to leave the profession. Liverpool, England: University of Liverpool, Centre for Education and Employment Research, Department for Education and Skills, Research Report RR430. Sobel, M. E. (1990). Effect analysis and causation in linear structural equation models. Psychometrica, 55, 495e515. Stoeber, J., & Rennert, D. (2008). Perfectionism in school teachers: relations with stress appraisals, coping styles, and burnout. Anxiety, Stress, & Coping. An International Journal, 21, 37e53. Tsouloupasa, C. N., Carson, R. L., Matthews, R., Grawitch, M. J., & Barber, L. K. (2010). Exploring the association between teachers perceived student misbehaviour and emotional exhaustion: the importance of teacher efficacy beliefs and emotion regulation. Educational Psychology, 30, 173e189. US Department of Education. (1997). Job satisfaction among America s teachers: Effects of workplace conditions, background characteristics, and teacher compensation. Washington, DC: National Center for Educational Statistics, U.S. Department of Education. Available at: html. Weiss, E. M. (1999). Perceived workplace conditions and firstyear teachers morale, career choice commitment, and planned retention: a secondary analysis. Teaching and Teacher Education, 15, 861e879. Weiss, H. M. (2002). Deconstructing job satisfaction: Separating evaluations, beliefs and affective experiences. Human Resource Management Review, 12(2), 173e194. Wriqi, C. (2008). The structure of secondary school teacher job satisfaction and its relationship with attrition and work enthusiasm. Chinese Education and Society, 40(5), 17e31. Zembylas, M., & Papanastasiou, E. (2004). Job satisfaction among school teachers in Cyprus. Journal of Educational Administration, 42, 357e374.

11 ID Title Pages Teacher job satisfaction and motivation to leave the teaching profession: Relations with school context, feeling of belonging, and emotional exhaustion