This is an open access article distributed under the terms of the Creative Commons
Attribution License (Teaching is one of the most stressful professions worldwide with potential detrimental effects on both teachers and their students (Johnson et al., 2005). Stress is a term that was not used in relation to teachers and instruction until 1972, while there was a vast increase in publications in the following years (Kyriacou, 2001). Teacher stress is defined as the experience of unpleasant and negative states, such as anger, tension, disappointment or depression, which arise from teaching responsibilities (Kyriacou, 2001). Stress tends to impact upon a person’s overall performance, both in terms of productivity and service delivery (McShane & Von Glinow, 2005). Almost one third of teachers consider their profession to be extremely stressful (Jepson & Forrest, 2006; Kyriacou, 2000) while, in comparison to other professionals, teachers experience much higher levels of stress. In particular, Smith, Brice, Collins, Matthews, and McNamara (2000) found that 41% of teachers reported high levels of job-related stress when compared to 31% of nurses, 29% of managers and 27% of administrators. Travers and Cooper (1993) reported that teachers experienced higher levels of stress than other employees in client-related professions, such as medical doctors, nurses, and hospital attendants.
According to the transactional model for understanding stress and coping proposed by Lazarus and Folkman (1984), when encountering life demands, a cognitive process is triggered in which perceived demands of the event are weighted against a person’s perceived capabilities for coping with those demands. Stress results from the perception that work demands exceed available resources. There are also other organizational and bio-psychological models that have attempted to explain the experience of stress, but most researchers refer to the basic assumption of Lazarus’ stress model (Hinton & Rotheiler, 1998). Stressed teachers are usually less tolerant of their students’ unwanted or challenging behaviors (Kokkinos, Panayiotou, & Davazoglou, 2005), tend to employ less ineffective classroom management strategies (Clunies-Ross, Little, & Kienhuis, 2008), and fail to support their students effectively (Jennings & Greenberg, 2009).
It is fairly easy to measure work stress that is related to a specific and concrete factor, such as exposure to loud noise, but variables related to the psychosocial work environment, such as job demands, are much more complicated to measure and investigate (Landsbergis & Theorell, 2000).
Some of the factors that contribute to stress are associated with: a) the in-class rather than organizational structure, reflected in disciplinary problems, class heterogeneity, and work overload (Forlin, 2001; Lewis, 1999; Male & May, 1998), b) individual differences that affect and determine teachers’ vulnerability against stress, such as age and gender – with younger and female teachers reporting higher stress levels (Byrne, 1991; Kantas, 2001; Offerman & Armitage, 1993), and c) factors related to the school organization and administration, such as limited governmental support and resources, inadequate training, continuous changes in the curriculum and legislation (Forlin, 2001; Travers & Cooper, 1997).
Teacher stress may be related to external factors, for example workload, time pressure, external evaluation, relationships with co-workers, classroom management (Chan, 2002; Jepson & Forrest, 2006), as well as the instruction of students with Special Educational Needs (SEN) who need more intensive and specialized training often with limited support and resources (Forlin, 2001; Male & May, 1997). Internal factors that could affect teacher stress are type of personality, professional commitment, and burnout (Jepson & Forrest, 2006; Yoon, 2002).
Burnout can be one of the primary causes of stress (Yoon, 2002), although there is also evidence that prolonged stress may cause burnout (Kyriacou, 2001). People who experience emotional exhaustion or depersonalization (which are basic components of burnout), are more likely to feel stressed when they have to face challenges in their work environment. Burnout is a work syndrome that stems out of the perceived discrepancy between the effort put into work and its corresponding rewards (Friedman, 1995) and leads to decreased prosperity and accomplishments (Burke & Greenglass, 1995). It is commonly observed in professions with intense face-to-face interactions, such as teachers or doctors (Schwab & Iwanicki, 1982) and its manifestation includes emotional and physical exhaustion, as well as many psychological symptoms – such as irritability, anxiety, and decreased self-esteem (Farber, 1991). It should be noted finally that burnout must be viewed as a continuous variable and not be classified as “all or nothing” (Girgin & Baysal, 2005).
Leiter and Maslach (1998) reported that burnout is characterized by emotional exhaustion that may result in physical symptoms such as loss of energy, depersonalization (described as cynicism), lack of idealism, negative or inappropriate attitudes against other people, and decreased personal accomplishment that is linked to decreased professional efficacy and productivity, low morale and inability to cope with work demands. It seems that both environmental–organizational and personal factors are involved in the etiology and manifestation of burnout (Farber, 1991), as put forward by the transactional model (Shirom, 1993). Since the work of teachers is extremely hard and demanding, many of them report increased burnout (Evers, Browers, & Tomic, 2002) that can prevent them ultimately from performing well in their job.
Culture seems to constitute an important component of work-related stress, as proposed by the Culture-Work-Health Model (Peterson & Wilson, 2002).There have been very few relevant studies on Greek primary schoolteachers. However, before discussing them, it is important to provide some information about the Greek situation. Greek primary schoolteachers work less hours than their European colleagues, have considerably longer vacations, a permanent job and no official evaluations. On the other hand, they have limited to non-existent resources, they are relocated often to different schools (very often away from their families), they have limited promotional opportunities, and low salaries with no extra benefits (Kantas & Vassilaki, 1997). Actually, since the financial crisis started in Greece, primary schoolteachers have also experienced 30%-40% cuts in the salaries.
The Greek literature on the topic includes work by Koustelios and Kousteliou (1998) and Kantas and Vassilaki (1997), who studied teachers from both primary and secondary education and focused mainly on burnout, Leontari, Kiridis, and Yalamas (1996), who obtained a general index of stress measured with a general single item, and Pomaki and Anagnostopoulou (2003), who measured aspects of stress using scales on job characteristics. These studies concluded that Greek schoolteachers experience relatively low levels of stress, whereas other studies have indicated that Greek teachers experience considerably high levels of stress and psychosomatic symptoms (Kantas, 2001; Papastylianou, 1997).
As such, there is a documented need to measure stress among Greek teachers focusing on a particular level of education (due to work differences between primary and secondary education), with a large sample and appropriate instruments. This has become more imperative in order to monitor the extent to which the recent massive educational and financial changes may have impacted on teacher stress (Kyriacou, 2001). Therefore, the main aim of the present study is to examine levels of teacher stress among Greek primary schoolteachers. The secondary aim is to explore whether internal (gender, age, years of teaching experience, and burnout) and external (teaching students with SEN) factors can predict levels of teacher stress.
Method
Participants
A number of 384 primary schoolteachers took part in this study, aged 25 - 59 years old (mean age = 41 years and 4 months). There were 146 males (38%) and 238 females (62%). More than half (61.7%) had never taught a child with SEN, while the remaining 38.3% had taught a student with SEN. Their teaching experience varied from 3-33 years (mean years of teaching experience = 14 years and 9 months). All the teachers were recruited from primary schools in Middle and Northern Greece.
Measures
The participants were asked to offer some demographic data and then to complete two questionnaires: the Questionnaire on Teacher Stress (Kyriacou & Sutcliffe, 1978) and the Maslach Burnout Inventory – Educators Survey (MBI-ES - Maslach, Jackson, & Leiter, 1996). The Questionnaire on Teacher Stress (Kyriacou & Sutcliffe, 1978) is one of two instruments that have been constructed to measure teacher stress (developed by Kyriacou, who also has offered the first definition of teacher stress). It has been used in other similar studies (e.g., Van Dick & Wagner, 2001; Kyriacou, 1979). It is made up of 50 items and, for the present study, the following scales were selected: unwanted student behavior (6 items), professional demands and responsibilities (8 items), school management (7 items), time pressure (5 items) and health problems (10 items). The instrument also makes it possible for teachers to identify 14 sources of stress, but this was beyond the scope of the present study. The answers are provided on a Likert scale from 1 to 7, where 1 = never and 7 = daily. None of the scores are reverse coded and all items are added to provide a total score. The minimum score is 36 and the maximum is 252, with a higher score indicating higher levels of stress. The Cronbach α reliability of the scale for the current study is 0.87 and is considered satisfactory.
The MBI-ES (Maslach et al., 1996) comprises 22 items that measure variables related to the teacher profession. It is the most widely used tool to measure burnout not only among teachers, but also among all mental health professionals – more than 90% of studies (Hastings, Horne, & Mitchell, 2004; Schaufeli & Enzmann, 1998). It yields scores along three dimensions: a) emotional exhaustion, which consists of 9 items that ask respondents to rate how frequently they experience such things as fatigue, frustration, and interpersonal stress in their jobs; b) depersonalization, which consists of 5 items that ask respondents to rate how frequently they have negative experiences with colleagues and clients; and c) professional accomplishment, which consists of 8 items that ask respondents to rate how frequently they have positive experiences in their job. The answers are provided on a Likert scale from 1 to 7, where 1 = never and 7 = daily and some items are reverse coded.
The minimum score is 22 and the maximum is 154. The Cronbach α reliability of the scale for the current study is 0.89 and is considered satisfactory.
Procedure
The first step was to gain ethical approval for the study and then to randomly select 35 schools from a list of all primary schools in Northern Greece. School principals were contacted and informed about the purpose of the study. Permission to contact the teachers was granted by 24 school principals and the researcher visited them to brief potential participants. It was stressed that participation was optional and that data collection would take place at the school after teaching hours. The questionnaires were anonymous and participants were informed that they could withdraw at any time without penalty. Out of the 400 teachers who were approached initially, 384 returned their questionnaire fully completed and the return rate was excellent (96%). The completion of the questionnaire lasted approximately 30 minutes and the researcher was available to answer any questions.
Results
Linear multiple regression was conducted to analyze the data. Teacher stress scores were regressed on gender, age, years of teaching experience, teaching students with SEN, and burnout (see Table 1). These five predictors accounted for almost one third of the variance in teacher stress scores (R2 = .33).
Table 1
Means, Standard Deviations and Intercorrelations Among the Variables
| M | SD | 1 | 2 | 3 | 4 | 5 | 6 | |
|---|---|---|---|---|---|---|---|---|
| 1. Gendera | 1.62 | 0.49 | 1.00 | |||||
| 2. Age | 41.40 | 5.01 | -0.12* | 1.00 | ||||
| 3. Years of teaching experience | 14.92 | 6.46 | -0.09* | 0.75** | 1.00 | |||
| 4. Teaching children with SENb | 1.62 | 0.49 | 0.08 | 0.15* | 0.13* | 1.00 | ||
| 5. Burnout | 61.66 | 12.13 | 0.58 | 0.09* | 0.14* | 0.16* | 1.00 | |
| 6. Stress | 92.00 | 49.20 | 0.05 | 0.04 | 0.04 | 0.15* | 0.29** | 1.00 |
a1 = male and 2 = female. b1 = taught students with SEN and 2 = did not teach students with SEN.
*p < 0.05. **p < 0.001.
Burnout (β = .27, p < .001) and teaching children with SEN (β = .15, p = .045) demonstrated significant effects on teacher stress scores, while gender (r(384) = .05, p = .190), age (r(384) = -.04, p = .227), and years of teaching experience (r(384) = -.04, p = .240) were not associated with teacher stress.
Discussion
The main aim of this study was to examine the levels of stress that Greek primary schoolteachers experience after the recent massive educational and financial reforms, using an appropriate measure that relates to this specific profession. It was found that the mean scores obtained from the teacher stress questionnaire were low (with a mean of 92 out of 252), indicating low levels of stress. This finding is contradictory to those of other studies (e.g., Jepson & Forrest, 2006; Smith, Brice, Collins, Matthews, & McNamara, 2000) that have been conducted in other countries, while evidence from Greece is contradictory. Although direct comparison with other studies is quite difficult due to the different nature of the sample and the different measurement techniques, this study is in line with the findings of Leontari et al. (1996) and Pomaki & Anagnostopoulou (2003). This could be attributed to the fact that Greek primary schoolteachers have long vacations, a permanent job, and no official evaluation (Kantas & Vassilaki, 1997). It was expected that the recent developments would have contributed to increased teacher stress but it is plausible that, since they are largely an external factor and everybody in Greece experiences these changes, people have started to adjust to them.
Levels of teacher stress were predicted by the burnout they experience and whether they have taught students with SEN regardless of their age, gender, and years of teaching experience. The variance that was explained by those two variables was quite satisfactory (33%). Burnout has been associated with stress in other studies conducted in Greece (e.g., Leontari et al., 1996; Pomaki & Anagnostopoulou, 2003), but there were issues with the measurement of stress and/or the nature of the sample (since stress was measured with more generic questionnaires and the sample contained also secondary schoolteachers and this group faces different professional challenges). Very few studies so far have identified burnout as a predictor of stress among primary schoolteachers, although this has been found for other professions (Jepson & Forrest, 2006; Yoon, 2002). This can have important practical implications, since techniques used to reduce burnout could be effective also in decreasing the levels of stress reported by teachers.
Teachers who have taught students with SEN in this study reported lower levels of stress than their colleagues who did not have such an experience. This finding is contradicted by some studies (e.g., Forlin, 2001; Williams & Gersch, 2004) and supported by others (e.g., Trendall, 1989). It could be attributed to the fact that Greek primary schoolteachers who instructed students with SEN have a considerably smaller class and a more flexible curriculum (Koustelios & Kousteliou, 1998), which seems to be the main difference from teachers who teach non-SEN students. More research is needed in this area to explore this factor that seems to predict the level of teacher stress and should, thus, be investigated further.
Although gender and age were found to differentiate levels of teacher stress in other studies (Byrne, 1991; Kantas, 2001; Offerman & Armitage, 1993), this was not the case in the present study, despite the fact that the sample was properly balanced. It is possible that these factors differentiate between teachers who experience high levels of stress which was not the case in this dataset.
This study, which is the first to explore teacher stress in Greece after the beginning of the financial crisis using appropriate questionnaires and a large sample of primary schoolteachers, has certain limitations. More specifically, it relies on self-reports that are more prone to bias than biological measures (Avramidis & Kalyva, 2006); also, some other variables could affect stress levels, such as personality traits or work conditions (e.g., Jepson & Forrest, 2006; Yoon, 2002), and they were not considered in the present study.
Levels of stress among Greek primary schoolteachers are not as high as in other European countries due to specific work conditions and despite the fact that this population has suffered from massive salary cuts and radical educational reforms. This is something worth exploring more in future research, as there seem to be some protective factors against stress even in extremely stressful times that need to be identified and enhanced through appropriate support mechanisms. It was also found that burnout and teaching students with SEN predicted almost one third of the variance in teacher stress. Since this was one of the few studies that used burnout as a predictor of stress, and not the other way around, further research is needed to explore this relationship in more detail.