John McCormick, Australian Centre for Educational Leadership, Wollongong University, Wollongong, Australia
Kerry Barnett, School of Education, University of New South Wales, Kensington, Australia
Purpose – It may be argued that some shared psychological mechanisms (attribution) and structures (schemas) are likely to play a role in how individuals perceive stress. This paper seeks to propose and test some hypothesised relationships between stress attribution domains and burnout dimensions.
Design/methodology/approach – The participants were 416 classroom teachers in 38 randomly selected high schools in New South Wales, Australia. Two established instruments, the Maslach Burnout Inventory, and the Teachers' Attribution of Responsibility for Stress Scale were employed in a postal survey. Data were analysed using confirmatory factor analysis and multilevel modelling.
Findings – Most variance was at the individual level, supporting the view that the stress and burnout were overwhelmingly psychological phenomena. Findings suggest the centrality of stress attributed to student misbehaviour in predicting each of the three dimensions of burnout: depersonalisation, emotional exhaustion, and personal accomplishment. Occupational stress attributed to personal failings also negatively predicted personal accomplishment.
Practical implications – The principal implication for practice is that greater emphasis should be placed on effective management of student behaviour when assisting teachers at risk of burnout.
Originality/value – This original study provides new insights into attribution schemas to assist understanding teachers' perceptions and reporting of their occupational stress and burnout in an education system.
Teachers; Stress; Australia; Individual behaviour; Psychological concepts.
International Journal of Educational Management
Emerald Group Publishing Limited
It has been argued that stress is an inherent condition of human experience (Selye, 1993). However, this relatively early conceptualisation of a common experience of stress has seemed less compelling, given the large body of evidence pointing to the importance of individual differences in explaining stress responses (Guglielmi and Tatrow, 1998; Havlovic and Keenan, 1995; McCormick, 1997a). Indeed, the importance of individual differences has even been demonstrated by rats' responses to noise stress (Macht et al., 2001).
Individual differences can be conceptualised in different ways. At the individual level, perhaps the most common differences seem to relate to human personality traits, for example, type D personalities, characterised by tendencies to experience negative affect (Sher, 2004), and type A, “driven” personalities (Rees and Cooper, 1992). However, there are other important differences, such as gender (Bellman et al., 2003; Biggam et al., 1997; Lundberg, 2002) and age (Cox et al., 2004).
Whilst differences exist between individuals, they clearly also exist between work contexts (Sparks and Cooper, 1999). That is, not only between occupations, but also between organizations and between distinct occupational contexts within organizations. This variation, at both the individual and organizational levels, lends weight to the argument that putting aside types of workplaces that are inherently toxic for human well-being, the transactional approach to explaining psychological occupational stress (Lazarus, 1995) is among the most valuable. That is, how people cognitively appraise their stress experiences is likely to be important for improvement of our understanding of the occupational stress phenomenon.
Although humans are all unique individuals, each has the same cognitive architecture, and fundamental functioning of memory, including functioning of memory of “who I am” and “what have been my life experiences”. However, it should be acknowledged that memory structures based on percepts can differ from objective reality, and we certainly do not retain all details initially stored in memory (Baddeley, 1997). Thus, although individuals are likely to appraise stressors differently, there are likely to be underlying common, fundamental psychological mechanisms and processes. There is a strong argument that our understanding of cognition should be incorporated into the study of occupational stress. Indeed, this argument for focusing on general, underlying principles and theory development rather than specific individual differences, was nicely summed up by Leon Festinger (1980, p. 246), and adopted by Weiner (1986), who wrote “The kind of analogy that existed in our minds was something like the following. It would be hopeless to have tried to discover the laws concerning free-falling objects by concentrating on measuring the different rates of descent of stones, feathers, pieces of paper, and the like”.
Teacher stress and burnout
It has been acknowledged that stress affects virtually all occupations, whether they are considered blue-collar (Macdonald, 2003) or white collar or professional (McCormick, 2003), and is a world-wide phenomenon (Cox et al., 2007). However, arguably, the teaching profession is among the most studied occupations in the occupational stress and burnout literature (e.g. Guglielmi and Tatrow, 1998; Klusman et al., 2008; Mearns and Cain, 2003). There has been considerable consistency in the nature of stressors reported in the day-to-day lives of teachers, e.g. administrative workload and student indiscipline (Bradley, 2007; Gold and Roth, 1993). However, there seems to be considerable variation in the experience of teacher occupational stress, in terms of both intra-personal differences and working in different environments (Milstein and Farkas, 1988). Indeed, this complexity is compounded by a lack of consensus for a single definition, or model of, “stress”. Beehr (1998, p. 842) has pointed out:
Job stress is an area with the potential to be plagued by confusion, at least partly because of the general, nontechnical popular usage of the word stress. Even among researchers, stress had sometimes been used to mean an environmental “stressor” stimulus and sometimes to mean an individual's strain or distress reactions.
If “stress” may be considered a word in everyday use, then certainly the same would appear to apply to “burnout”, which arguably, is somewhat emotion-laden. Constituting one of the caring or human service professions, teaching has often been the context for burnout research (e.g. Ben-Ari et al., 2003; Guglielmi and Tatrow, 1998; von Kanel et al., 2008). Herbert J. Freudenberger, a psychotherapist, is credited with being the first person to use the term “burnout” in the academic literature (Pines, 1993). However, probably the most influential work in the area is that carried out by Christina Maslach and colleagues (Aluja et al., 2005; Maslach and Jackson, 1981; Maslach and Leiter, 2008) through the development of the Maslach Burnout Inventory (MBI), which has focused on the measurement of the burnout phenomenon through three dimensions, namely, emotional exhaustion, depersonalisation and reduced personal accomplishment. Emotional exhaustion is a feeling of exhaustion that is not explained by excessive physical activity. Depersonalisation is a diminishing of concern for the well-being of other persons. In service and caring professions, the other persons are typically the clients and those in care. For example, a medical practitioner experiencing burnout may be expected not to be concerned about inflicting pain on a patient when administering a medical procedure, a teacher may not care whether a student learns, or is in a safe environment. The personal accomplishment dimension relates to being effective and “making a difference” (Maslach, 1998).
Interestingly, it appears that whilst stress has been conceptualised as having the potential to be experienced in many contexts, including, for example, the home, the study of burnout seems to have largely been focused on the workplace. Stress and burnout are conceptually distinct. Pines (1993) argued that whilst stress is experienced by everybody, burnout is not. She described burnout in terms of a highly motivated individual experiencing an existential crisis: “You cannot burn out unless you were ‘on fire’ initially” (p. 386). Before burning out, workers expect to make a difference. A teacher who joins the profession expecting to be instrumental in shaping the minds and futures of adolescents may be disillusioned by her experiences in a real world high school. Whilst Pines (1993) explicitly stated “… stress as such does not cause burnout” (p. 387), Maslach and Jackson (1981, p. 99) wrote “… the chronic stress can be emotionally draining and poses the risk of ‘burnout’”. Perhaps, the common ground is that stress, even intense stress, per se, does not necessarily lead to burnout, but it can.
Notwithstanding the inherent complexities of the occupational stress and burnout phenomena, and the realization that the existence of the former does not necessarily lead to the latter, some have postulated and demonstrated empirically, associations between the two, including in the context of teaching (Burke and Greenglass, 1995; Mearns and Cain, 2003). It appears clear from what has preceded that individuals' responses to occupational stressors and the potential for occupational burnout are idiosyncratic, depending both on the individual and her or his response to the environment. However, there is a phenomenon that underlies this frame: human cognition. Whilst we differ in our perceptions, beliefs and values, each of us organizes them cognitively in much the same way.
The attribution of responsibility for teacher stress model
Attribution theory is an umbrella term under which a number of theories and approaches are located (Forsterling, 1988). An underlying principle of attribution theory is that humans are “naïve scientists” hardwired to make sense of what happens in the world through causal explanations (Hewstone, 1983). This is functional (Effler, 1984) and may have served evolutionary purposes. Whilst attribution research has largely focused on causal explanations of specific events, for example, “Why was my job application unsuccessful?”, it can be integrated with schema theory. Schemas are knowledge structures stored in long-term memory. When a person has consistent experiences, followed by consistent attributions, causal attribution schemas may be expected to form.
McCormick and colleagues (De Nobile and McCormick, 2010; McCormick and Solman, 1992; McCormick, 1997a,b; McCormick and Shi, 1999; McCormick, 2003; McCormick et al., 2006) have posited and tested empirically the attribution of responsibility for teacher stress model for teachers working in schools within educational systems. The point of departure for the model is the concept of loose coupling (Weick, 1976), which seems to be particularly relevant to educational organisations. Loose coupling is essentially about the extent to which work is closely controlled. In a large educational system, the coupling is between subsystems. In normal circumstances, a school principal will not closely supervise the work of teachers. Indeed, perhaps with the exception of situations where there are concerns about the quality of a teacher's work, department heads generally will not closely supervise the work of a classroom teacher. Similarly, even more so, head office usually will not closely control day-to-day activities in a school. Hence, some subsystems may be conceptualised by teachers as being less or more close to the teacher than others. Teachers may be expected to develop organisational schemas more consistent with a loosely coupled structure than with formal, hierarchical organization charts. The latter could portray teachers embedded within the school, which is embedded within the school system, which is embedded within society. However, the teachers may be expected to develop schemas that locate these entities at varying distances from self, with fellow teachers relatively close and head office and Society relatively distant.
There is considerable empirical support for the self-serving or hedonic attribution bias (Forsterling, 1988). In general terms the hedonic bias suggests that in individualistic countries, such as the USA, the UK and Australia, people tend generally to take personal credit for their successes (Kashima and Triandis, 1986), and deny responsibility for their failures (Weiner, 1985). How people respond to stressful environments is commonly conceptualised as successful coping, or failure to cope. Thus, failure may be associated with teachers' occupational stress.
Drawing the preceding elements together, teachers in an educational system may generally be expected to perceive their occupational stress as a negative phenomenon, and being responsible for their own stress as a form of failure. Moreover, teachers may generally be expected to attribute responsibility for their occupational stress to specific domains, conceptually organised in relation to self schemas. The hedonic bias suggests that the teachers will attribute responsibility for their stress increasingly to entities increasingly distant from self. These may be conceptualised as attribution schemas, developed and elaborated over the time the teachers spend in the organisational milieu. Thus, they would be relatively stable, although they may be altered by salient organisational experiences. Applying the attribution of responsibility for teacher stress with different samples and contexts, McCormick and associates consistently have empirically measured four attribution stress domains. The personal domain relates to stress attributed to self-perceived adequacy as a teacher. The student domain relates to stress attributed to student misbehaviour. The school domain relates to stress attributed to a lack of a supportive school environment. The external domain relates to stress attributed to distant entities such as the system's head office, and government educational policies.
Conceptual framework and hypotheses
The conceptual framework of this research comprises some components discussed above. First, a teacher's cognitive organization of explanations for his occupational stress may be expected to play a role in the extent to which he perceives himself burned out. Second, (attribution) stress domains will be related differentially to different burnout dimensions. The latter leads to a set of hypotheses.
Stress attributed to the personal domain relates to a teacher's assessment of her effectiveness as a teacher. It follows logically that a teacher who experiences stress because of perceptions of not being suited to the work of teaching is likely also not to perceive high levels of personal accomplishment. Of course a complementary argument can be made for a teacher who perceives little stress from feelings of professional inadequacy. This hypothesis follows:
H1. Personal domain stress will be related negatively to personal accomplishment.
Stress attributed to the student domain may be related to each of the burnout dimensions. First, students as the “clients” of teachers are arguably the most likely candidates to be depersonalised by a burned out teacher. Second, stress in this domain is likely to be related to the teacher not managing student behaviour successfully, and the likely consequence is that this is associated with less effective learning and teaching. Third, the day in and day out struggle with student misbehaviour is likely to prove exhausting. Hence the following hypotheses are posited:
H2. Student domain stress will be related positively to depersonalisation.
H3. Student domain stress will be related negatively to personal accomplishment.
H4. Student domain stress will be related positively to emotional exhaustion.
Teaching as a profession is not only about teaching students. It also involves interacting with colleagues in the school. Stress attributed to the school domain (lack of a supportive environment) is likely to be associated with unsatisfying relationships with fellow members of the profession. This can lead to a sense of diminished personal accomplishment:
H5. School domain stress will be related negatively to personal accomplishment.
Similarly, stress attributed to more distant entities such as the Ministry/Department of Education bureaucracy and government policies, is likely to be perceived by teachers as problematising their work, making it more difficult for them to work effectively:
H6. External domain stress will be related negatively with personal accomplishment.
In addition to the hypotheses, the following research question was formulated:
RQ. Are the demographic variables, sex, age, and years of teaching experience predictors of depersonalisation, emotional exhaustion, and personal accomplishment?
One high school was randomly selected from each of the 40 government school areas in New South Wales (NSW), Australia. The approximate number of non-executive, classroom teachers was established at the time of securing the cooperation of the schools, and sufficient numbers of questionnaires for approximately half the teachers at each school were sent to coordinators. The latter were asked to distribute randomly the questionnaires, each with an accompanying letter requesting participation, and a prepaid return envelope. In total, 416 teachers returned survey forms, rendering a return rate of approximately 36 per cent. Survey forms from two schools were subsequently eliminated because of the very small number of returned questionnaires. The sample comprised 57 per cent females; the mean age was 42.5 years. The mean number of years of teaching experience was 17.9.
The data reported here were gathered by three sections of the questionnaire. The first asked participants for demographic details: age, number of years of teaching experience and gender. The second comprised a version of the Maslach Burnout Inventory (e.g. Chan and Hui, 1995), which has been found to have three distinct subscales: depersonalisation (e.g. “I feel I treat some students as if they were impersonal objects”; five items), personal accomplishment (e.g. “I deal very effectively with the problems of my students”; eight items), and emotional exhaustion (e.g. “I feel emotionally drained from my work”; nine items). Participants were asked to report how often they experienced particular feelings on a seven-point scale ranging from 0 – never to 6 – a few times a week. The third section comprised the Teachers' Attribution of Responsibility for Stress Questionnaire (TARSQ) (McCormick et al., 2006), which has four distinct subscales, personal domain (e.g. “Feeling of not being suited to teaching”; two items), student domain (e.g. “Verbal abuse by students”; four items), school domain (e.g. “Lack of a supportive and friendly atmosphere”; five items), and external domain (e.g. “The Government's education policies”; four items). Although these specific scales do not measure the degree of responsibility attributed for the stress, they are consistent with the theoretical model in that they seek to measure perceived stress in the domains. Participants were asked to report how stressful experiences were for them by responding on a five-point Likert-type scale, ranging from 1=no stress to 5=extreme stress.
Analysis and discussion
The first step was to apply confirmatory factor analysis to the stress domains and burnout variables. The four factor stress domains measurement model was confirmed (χ2=154.24 p <0.001; χ2/df =1.84; RMR =0.05; RMSEA =0.05; NFI =0.97; CFI =0.99; AGFI =0.99; GFI =0.99). The three factor burnout measurement model could not be accepted as three items had R-squared statistics less than 0.3. These were sequentially eliminated, resulting in a three factor model with acceptable fit (χ2=591.80 p<0.001; χ2/df =3.54; RMR =0.08; RMSEA =0.08; NFI =0.94; CFI =0.96; AGFI =0.96; GFI =0.97). The next step was to estimate an overall full measurement model comprising the stress and burnout factors. This model has good fit and is shown schematically with fit statistics in Figure 1. The full measurement model was used to generate latent variable scores for each of the stress domain and burnout factors. Table I shows raw scale means, standard deviations, intercorrelations of the factors estimated in the full measurement model, and scale Cronbach alphas on the diagonal. Intercorrelations provide support for hypotheses one to four. However, it was deemed appropriate to submit the data, given their structure, to multilevel analysis.
A summary of variance decomposition models is shown in Table II, which shows that only two stress variables, external domain and student domain, had statistically significant variance at the school level with relatively small effect sizes. That most of the variance occurred overwhelmingly at the individual level reinforces that these essentially are internal, psychological phenomena. Three multilevel models were developed with the burnout variables as dependent variables. These are shown in Tables III–V. A hierarchical approach was adopted. The demographic variables were entered first, sequentially in the order of sex, age, then years of teaching experience. It was argued that sex was likely to be the most salient as it generally is determined at birth; age preceded teaching experience as it is a universal characteristic of all human beings. The stress domain variables were then added sequentially in the order: external domain, school domain, student domain, personal domain for the model with depersonalisation as the dependent variable, reflecting the conceptual distancing from the self. This order was reversed for models with emotional exhaustion and personal accomplishment as these may be conceptualised as being closer to the self. The models at each step were checked for a decrease in the log-likelihood statistic, and statistically non-significant variables were eliminated. The latter resulted in a slight increase in the log-likelihood function for the final model in two of the three analyses.
Table III shows the development of a multilevel model with depersonalisation as the dependent variable. Sex is a statistically significant predictor of depersonalisation, suggesting that male participants were likely to score more highly on depersonalisation than females. This is consistent with earlier studies (e.g. … The other statistically significant predictor of depersonalisation is student domain), hence, H2 is supported.
Table IV shows the development of a multilevel model with emotional exhaustion as the dependent variable. Of the demographic variables, only years of teaching experience is a statistically significant predictor of emotional exhaustion. Logically, burnout in the form of emotional exhaustion is likely to rise to higher levels the longer a person works in a demanding environment. Table IV also shows that student domain is a statistically significant predictor of emotional exhaustion, hence, H4 is supported. Although not hypothesised, external domain stress also is a statistically significant predictor of emotional exhaustion. This may reflect a belief of many of these teachers that a lack of support from the educational bureaucracy and the government made their job difficult.
Table V shows the development of a multilevel model with personal accomplishment as a dependent variable. Sex is a statistically significant predictor of personal accomplishment, suggesting male participants were more likely to report lower personal accomplishment than females. Additionally, personal domain and student domain are statistically significant negative predictors of personal accomplishment. Hence, H1 and H3 are supported. H5 and H6, that school domain stress and external domain stress were negative predictors of personal accomplishment are not supported. External domain is a statistically significant positive predictor of personal accomplishment. However, significance was only marginal (0.05 level), and it may be unwise to speculate on this result.
The results of this study suggest that for these teachers, the extent of stress attributed to the personal domain and the student domain predicts specific burnout dimensions. The student domain is a predictor of each of the three burnout dimensions. Generally, the greater the stress attributed to the student domain, the greater the depersonalization, and vice versa. Arguably, teachers who generally perceive misbehaving students to be sources of stress, are less likely to see them in terms of their individual humanity than if this were not the case. Similarly, if teachers are consistently experiencing stress from badly behaved students, this is likely to exact an emotional toll, as the teachers could experience a range of emotions, perhaps most notably, anger. Moreover, this could work in both directions as emotion-charged teaching is less likely to be effective. Consistent with the scenarios described above, is the view that personal accomplishment as a teacher is likely to encompass, at least in part, successful management of student behaviour, or, the latter is a necessary condition for accomplishment of other goals in the classroom, like effective teaching of content. Relatedly, generally, the greater the stress these participants attributed to their personal inadequacy as teachers, the lower their sense of personal accomplishment, and vice versa. The unexpected result that external domain stress is associated with emotional exhaustion possibly could mean that attributing stress to distant entities such as the head office and the government involves negative emotions. However, this explanation is speculative and really needs to be probed in further research.
This study has a number of limitations that should be acknowledged. First, the data are self-report, and, although the survey was confidential, and participants were anonymous, it is possible that the latter did not accurately respond to the survey items. Second, only a limited number of theoretical constructs were used in the study. Other constructs could be just as, or even more, relevant. Third, the data are cross-sectional in nature and causality cannot be inferred. Fourth, the statistical models are only a few of many possibilities, although, it should be emphasized that models are based on theory and logical argument.
The main purpose of this study was to test hypothesised relationships between two sets of theoretical constructs, namely, stress domains, and burnout. The attribution of responsibility for stress model (McCormick, 1997b) proposes that teachers develop and elaborate relatively stable attribution schemas for their occupational stress in specific domains. Burnout is a well-established phenomenon in client-centred occupations, including teaching (Aluja et al., 2005; Burke and Greenglass, 1995). Although, as has been emphasized previously, one may not assume occupational burnout is a consequence of occupational stress, it seems likely that, at least in some instances, they are dynamically related. Perhaps one of the most important conclusions that can be made in the context of this study is that stress attributable to student misbehaviour appears to be the most salient in relation to burnout. Arguably, this makes sense in terms of the centrality of managing student behaviour in the work of teachers. To experience high levels of stress consistently from an inability to manage student behaviour successfully, is tantamount to failure, and likely to be associated with the depersonalising of students, experiencing of an exhausting range of negative emotions, and a sense of diminished accomplishment. Conversely, teachers who treat students impersonally, experience a range of negative emotions, and feel that they are not achieving professional goals, are unlikely to manage student behaviour effectively. This research suggests that programs designed to assist teachers identified as being at risk of burning out should focus on improving the management of student behaviour. Such teachers could be retrained to equip them to carry out this part of their work more successfully, and provided with increased supervisor and peer support to enable improvement. Notwithstanding, whilst it seems logical to identify the burned out teacher as the person at risk, the fact is that there are likely to be many students being taught ineffectively, which ultimately could be more costly for society.
Figure 1Full measurement model incorporating burnout and stress domain latent variables
Table IRaw scale means and standard deviations and inter-correlations of latent variables estimated in the full measurement model with Cronbach alpha statistics on the diagonal
Table IIFully unconditional variance decomposition models
Table IIIDevelopment of a multilevel model with depersonalisation as the dependent variable
Table IVDevelopment of a multilevel model with Emotional Exhaustion as dependent variable
Table VMultilevel models with personal accomplishment as the dependent variable
Aluja, A., Blanch, A., Garcia, L.F. (2005), "Dimensionality of the Maslach Burnout Inventory in school teachers: a study of several proposals", European Journal of Psychological Assessment, Vol. 21 No.1, pp.67-76.
Baddeley, A. (1997), Human Memory: Theory and Practice, Psychology Press, Hove, .
Beehr, T.A. (1998), "Research on occupational stress: an unfinished enterprise", Personnel Psychology, Vol. 51 No.4, pp.835-44.
Bellman, S., Forster, N., Still, L., Cooper, C.L. (2003), "Gender differences in the use of social support as a moderator of occupational stress", Stress and Health, Vol. 19 No.1, pp.45-58.
Ben-Ari, R., Krole, R., Har-Even, D. (2003), "Differential effects of simple frontal versus complex teaching strategy on teachers' stress, burnout and satisfaction", International Journal of Stress Management, Vol. 10 No.2, pp.173-95.
Biggam, F.H., Power, K.G., Macdonald, R.R., Carcary, W.B., Moodie, E. (1997), "Self-perceived occupational stress and distress in a Scottish police force", Work & Stress, Vol. 11 No.2, pp.118-33.
Bradley, G. (2007), "Job tenure as a moderator of stressor-strain relations: a comparison of experienced and new-start teachers", Work & Stress, Vol. 21 No.1, pp.48-64.
Burke, R.J., Greenglass, E.R. (1995), "A longitudinal study of psychological burnout in teachers", Human Relations, Vol. 48 No.2, pp.187-202.
Chan, D.W., Hui, E.K.P. (1995), "Burnout and coping among Chinese secondary school teachers in Hong Kong", British Journal of Educational Psychology, Vol. 65 No.1, pp.15-25.
Cox, T., Karanika, M., Griffiths, A., Houdmont, J. (2007), "Evaluating organizational-level work stress interventions: beyond traditional methods", Work & Stress, Vol. 21 No.4, pp.348-62.
Cox, T., Leka, S., Ivanov, I., Kortum, E. (2004), "Work, employment and mental health in Europe", Work & Stress, Vol. 18 No.2, pp.179-85.
De Nobile, J.J., McCormick, J. (2010), "Occupational stress of catholic primary school staff: a study of biographical differences", International Journal of Educational Management, Vol. 24 No.6, pp.492-506.
Effler, M. (1984), "Attribution theories of lay epistemology?", European Journal of Social Psychology, Vol. 14 No.4, pp.431-7.
Festinger, L. (1980), "Looking back", in Festinger, L. (Eds),Retrospections on Social Psychology, Holt, Rinehart & Winston, New York, NY, pp.236-54.
Forsterling, F. (1988), Attribution Theory in Clinical Psychology, Wiley, Chichester, .
Gold, Y., Roth, R.A. (1993), Teachers Managing Stress and Preventing Burnout: The Professional Health Solution, The Falmer Press, London, .
Guglielmi, R.S., Tatrow, K. (1998), "Occupational stress, burnout, and health in teachers: a methodological and theoretical analysis", Review of Educational Research, Vol. 68 No.1, pp.61-99.
Havlovic, S.J., Keenan, J.P. (1995), "Coping with work stress: the influence of individual differences", in Crandall, R., Perrewe, P.L. (Eds),Occupational Stress: A Handbook, Taylor & Francis, Washington, DC, pp.179-92.
Hewstone, M. (1983), "Attribution theory and common-sense explanations: an introductory overview", in Hewstone, M. (Eds),Attribution Theory: Social and Functional Extensions, Basil Blackwell, Oxford, pp.1-26.
Kashima, Y., Triandis, H.C. (1986), "The self-serving bias in attributions as a coping strategy", Journal of Cross-cultural Psychology, Vol. 17 No.1, pp.83-97.
Klusman, U., Kunter, M., Trautwein, U., Ludtke, O., Baumert, J. (2008), "Teachers' occupational well-being and quality of instruction: the important role of self-regulatory patterns", Journal of Educational Psychology, Vol. 100 No.3, pp.702-15.
Lazarus, R.S. (1995), "Psychological stress in the workplace", in Crandall, R., Perrewe, P.L. (Eds),Occupational Stress: A Handbook, Taylor & Francis, Washington, DC, pp.3-14.
Lundberg, U. (2002), "Psychophysiology of work: stress, gender, endocrine response, and work-related upper extremity disorders", American Journal of Industrial Medicine, Vol. 41 No.5, pp.383-92.
McCormick, J. (1997a), "Occupational stress of teachers: biographical differences in a large school system", Journal of Educational Administration, Vol. 35 No.1, pp.18-36.
McCormick, J. (1997b), "An attribution model of teachers' occupational stress and job satisfaction in a large education system", Work and Stress, Vol. 11 No.1, pp.17-32.
McCormick, J. (2003), "Stress and professional work", in Peterson, C. (Eds),Work Stress: Studies of the Context Content and Outcomes of Stress, Baywood, Amityville, pp.167-79.
McCormick, J., Ayres, P.L., Beechey, B. (2006), "Teaching self- efficacy, stress and coping in a major curriculum reform: applying theory to context", Journal of Educational Administration, Vol. 44 No.1, pp.53-70.
McCormick, J., Shi, G. (1999), "Teachers' attributions of responsibility for their occupational stress in the People's Republic of China and Australia", British Journal of Educational Psychology, Vol. 69 No.3, pp.393-407.
McCormick, J., Solman, R. (1992), "The externalised nature of teachers' occupational stress and its association with job satisfaction", Work and Stress, Vol. 6 No.1, pp.33-44.
Macdonald, W.A. (2003), "Work demands and stress in repetitive blue-collar work", in Peterson, C.L. (Eds),Work Stress: Studies of the Context, Content and Outcomes of Stress, Baywood, Amityville, pp.139-63.
Macht, M., Krebs, H., Weyers, P., Janke, W. (2001), "Effect of stress on feeding behavior in rats: individual differences", Personality and Individual Differences, Vol. 30 No.3, pp.463-9.
Maslach, C., Jackson, S.E. (1981), "The measurement of experienced burnout", Journal of Occupational Behavior, Vol. 2 No.1, pp.99-113.
Maslach, C., Leiter, M.P. (2008), "Early predictors of job burnout and engagement", Journal of Applied Psychology, Vol. 93 No.3, pp.498-512.
Maslach, C.A. (1998), "A multidimensional theory of burnout", in Cooper, C.L. (Eds),Theories of Organizational Stress, Oxford University Press, New York, NY, pp.68-85.
Mearns, J., Cain, J.E. (2003), "Relationships between teachers' occupational stress and their burnout and distress: roles of coping and negative mood regulation expectancies", Anxiety, Stress and Coping, Vol. 16 No.1, pp.71-82.
Milstein, M.M., Farkas, J. (1988), "The overstated case of educator stress", Journal of Educational Administration, Vol. 26 No.2, pp.232-49.
Pines, A.M. (1993), "Burnout", in Goldberger, L., Breznitz, S. (Eds),Handbook of Stress: Theoretical and Clinical Aspects, The Free Press, New York, NY, pp.386-402.
Rees, D., Cooper, C.L. (1992), "Occupational stress in health service workers in the UK", Stress Medicine, Vol. 8 No.1, pp.79-90.
Selye, H. (1993), "History of the stress concept", in Goldberger, L., Breznitz, S. (Eds),Handbook of Stress: Theoretical and Clinical Aspects, Free Press, New York, NY, pp.7-17.
Sher, L. (2004), "Type D personality, stress, and cortisol", Journal of Psychosomatic Research, Vol. 57 No.2, pp.117-8.
Sparks, K., Cooper, C.L. (1999), "Occupational differences in the work-strain relationship: towards the use of situation-specific models", Journal of Occupational and Organizational Psychology, Vol. 72 No.2, pp.219-29.
von Kanel, R., Bellingrath, S., Kudielka, B.M. (2008), "Association between burnout and circulating levels of pro- and anti-inflammatory cytokines in schoolteachers", Journal of Psychosomatic Research, Vol. 65 No.1, pp.51-9.
Weick, K.E. (1976), "Educational organizations as loosely coupled systems", Administrative Science Quarterly, Vol. 21 No.1, pp.1-19.
Weiner, B. (1985), "An attributional theory of achievement motivation and emotion", Psychological Review, Vol. 92 No.4, pp.548-73.
Weiner, B. (1986), An Attributional Theory of Motivation and Emotion, Springer-Verlag, New York, NY, .
Anderson, R.C., Pichert, J.W., Shirey, L.L. (1983), "Effects of the reader's schema at different points in time", Journal of Educational Psychology, Vol. 75 No.2, pp.271-9.
John McCormick can be contacted at: email@example.com