Workplace Stress Among Teachers in Kosovo

Teaching has been regarded as one of the most stressful professions, and workplace stress within this professional category has been thoroughly investigated. Nonetheless, no empirical research so far has examined workplace stress among teachers in Kosovo. The present study aimed to identify age and gender-related patterns of workplace stress as well as examine the role of marital status, educational level, and working experience in a sample of Kosovo teachers. The different types of stressors reported by teachers were also examined. The sample consisted of 799 teachers (Mage = 42.94; SD = 11.50), 33.8% males and 65.2% females. The measures included the National Stress Awareness Day (NSAD) Stress Questionnaire, and one self-report questionnaire designed by the authors for the purpose of the research. Results showed that 33.2% (265 participants) of the sample reported high levels of stress. Workplace stress was significantly predicted by place of residence (β = −.442, p < .00) and level of education (β = −.191, p < .00) but not age, gender, marital status, or working experience F(6, 520) = 34.162, p < .001, R2 = .283. As regards the specific stressors, the most frequently reported were inadequate wages (36.8%), physical working environment (30.1%), and undisciplined students (26.2%). Results are discussed in the context of practical implications they have and suggestions for future research are provided.


Workplace Stress Among Teachers
The increasing rates of workplace stress and burnout are posing a significant challenge to well-being in the working environment, especially in European Union countries (Milczarek, Schneider, & González, 2009). The highest levels of workplace stress were reported in Greece (55%) and Slovenia (38%), whereas the lowest levels were reported in the United Kingdom (12%) and Germany (16%; Milczarek et al., 2009).
The European Agency for Safety and Health at Work (2009) has reported that stress was most common in education and health-related professions (up to 28.5%) as compared with other job sectors (Milczarek et al., 2009). Indeed working conditions and health-related issues of the teaching profession have been investigated over the past 30 years, mainly due to the high levels of workplace stress reported (Krause, Dorsemagen, & Alexander, 2011).
In terms of prevalence rates, research conducted by the National Union of Teachers (NUT) in 1999 found that 36% of British teachers reported stress most of the time (NUT, 1999). In addition, a report published in 2000 by the Health and Safety Executive found that 41.5% of these teachers reported high levels of stress (Health and Safety Executive, 2000). However, rates greatly differ from country to country; for instance, a study of occupational stress among Turkish and Macedonian teachers found mild stress levels in the Turkish sample and moderate stress levels in Macedonian sample (Eres & Atanasoska, 2011). On the other hand, problematically high stress rates have been reported in countries such as Slovenia (around 84%; European Foundation for the Improvement of Living and Working Conditions, 2010). These findings are especially concerning if considered in the context of an increasing trend to abandon altogether the teaching profession, which is regarded as frustrating and emotionally depleting (Hanushek, 2007;Ingersoll, 2003;Lambert, O'Donnell, Kusherman, & McCarthy, 2006).

Stress-Related Individual Factors: Gender, Age, Working Experience, and Marital Status
Research on stress-related factors in the workplace has focused on the role of variables such as age, gender, working experience, marital status, and so on (Bhadoria & Singh, 2010;Lau, Yuen, & Chan, 2005 (Jick & Mitz, 1985). However, some researchers have failed to find any significant gender differences in their studies (Fontana & Abouserie, 1993). For instance, figures from the European Working Conditions Survey carried out in 1995, 2000/2001, and 2005 show small differences between levels of stress reported by men and women. However a more recent survey did show that stress is a little more prevalent among men (23%) as compared with women (20%; Milczarek et al., 2009). Also in a study among 400 teachers in Pakistan in primary and secondary schools, it was found that gender was a very strong predictor of teachers' stress (Rubina, Sadaf, & Masood, 2011). Mondal, Shrestha, and Bhaila (2011) also found a significant difference between male and female teachers, with male teachers reporting more stress than the female teachers (Mondal et al., 2011). This finding was in line with research in the field (Brember, Brown, & Ralph, 2002;Chaplain, 1995;Cheng Kaa-Lee, 1993;Gursel, Sunbul, & Sari, 2002). Eres and Atanasoska (2011) also reported significantly higher levels of stress among men in the Turkish sample but not in the Macedonian one (Eres & Atanasoska, 2011). Despite this, studies actually show that female teachers tend to complain more of burnout than male teachers (Bhadoria & Singh, 2010;Chan, Chen, & Chong, 2010;Ravichandran & Rajendran, 2007). To sum up, gender still remains a quite interesting factor to be examined in the context of workplace stress.
Age-related findings on workplace stress also vary a lot. According to the European Working Conditions Survey (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005), the highest stress levels were observed among middle-aged (45-54 years) workers, and the lowest among older and younger workers (Milczarek et al., 2009). According to Health and Safety Laboratory and Health and Safety Executive (2005) report, the highest levels of stress were experienced by women aged between 25 and 34 years and men aged between 35 and 44 years (Health and Safety Laboratory and Health and Safety Executive, 2005. This finding has been replicated in other studies showing that younger and less experienced teachers suffered more burnout than older or more experienced teachers (Abdul Majid, 1998;Bhadoria & Singh, 2010;Hadi, Naing, Daud, Nordin, & Sulong, 2009;Lau et al., 2005). On the other hand, numerous studies did not find any relationships between age, job experience, and stress levels (Chona & Roxas, 2009;Jepson & Forrest, 2006;Johannsen, 2011;Mondal et al., 2011;Platsidou & Agaliotis, 2008). Yet another study reported higher levels of stress among younger teachers, but results did not reach statistical significance (Koruklu, Feyzloôlu, Ozenoölu-Kiremit, & Aladag, 2012). Eres and Atanasoska (2011) found similar results with significant age differences found only for the Turkish sample (but not the Macedonian one; Eres & Atanasoska, 2011).
Inconclusive findings also exist on the influence of marital status on workplace stress. Marital status has been related to stress levels; more specifically, individuals who were widowed, divorced, or separated have reported higher stress levels (Smith, Brice, Collins, Mathews, & McNamara, 2000). On the other hand, some studies have reported no significant differences by marital status (Cheng Kaa-Lee, 1993;Chona & Roxas, 2009;Yahaya, Hashim, & Kim, 2008). Apart from marital status, educational level is also important to consider. Research has found an increase of stress levels with levels of education, so that more educated individuals experience higher stress (Blaug, Kenyon, & Lekhi, 2007). However, there are also studies which have found no significant differences by educational level (Darmody & Smyth, 2011;Eres & Atanasoska, 2011).

Types of Stressors
Research investigating types of stressors among teachers is vast and in several cases inconsistent. Examples of stressors include the lack of opportunities for career progress, lack of status or respect at work, work overload, instability, and lack of discipline among students (Akhlaq, Amjad, & Mehmood, 2010;Platsidou & Agaliotis, 2008). Other studies have reported stressors such as time constrains (Kyriacou, 1987), poor quality of relationships with colleagues (Troman, 2000), and large numbers of students (Nagel & Brown, 2003;Trendall, 1989). In terms of stressor rankings, student behavior is reported as the most stressful factor followed by staff support, relationship with parents, personal and professional competences, and so on (Paulse, 2005). Other examples of stressors include, role conflict, role ambiguity, discipline problems, time pressure, bad working conditions, selfrespect, inadequate support from friends, family and colleagues, and low student motivation (A. Chan et al., 2010;Detert, Caravella, Derosia, & Duquette, 2006).

Workplace Stress Among Teachers in Kosovo
Research on workplace stress in Kosovo is scarce, whereas studies on teacher stress, inexistent. Bytyqi, Reshani, and Hasani (2010) conducted a general workplace survey in public organizations which examined employees' levels of work stress, job satisfaction and organizational commitment, and the impact these factors have on each other. The quantitative research findings revealed high levels of workplace stress (Bytyqi et al., 2010). Nonetheless, no research so far has been conducted among teachers. Such a study is particularly relevant considering that teachers in Kosovo have been exposed to multidimensional stressors related to the specific socio-political context and difficulties before and after the Kosovo war (educational system reforms, large numbers of students, work shifts, low wages, etc.). Such specificity might yield quite interesting findings not only in terms of understanding stress in the specific socio-cultural context, but also in suggesting ways in which existing theories might be modified or adapted to reflect cultural diversity. This aspect is especially relevant considering that most research has been conducted in Western countries or the United States, while developing countries are only minimally contributing to the field.
Moreover, the increasing number of individuals choosing to work as teachers especially during recent years (a pattern reflecting the increasing need for this profession with the construction of new schools and increasing number of students; Kosovo Agency of Statistics, 2014) suggest an urgent need to examine stress and stress-related factors in this specific target group.
In this context, a model constructed on specific objective characteristics (e.g., age, gender) has more of a practical rather than theoretical relevance. Thus, it might be important to examine stress patterns in terms of gender, age, marital status, or working experience because such investigation would allow the identification of specific target groups for intervention (e.g., middle-aged divorced men). The consideration of specific target group characteristics would subsequently guide the design of tailored interventions toward stress management. On the other hand, the identification of specific job stressors is important in terms of policy making (e.g., at an institutional level or even at a national level).

Aims
The purpose of the present study was to investigate workplace stress and stress-related variables among teachers in Kosovo. The study aims to identify age and gender-related patterns of workplace stress as well as examine the role of marital status, educational level, and working experience on stress levels. As suggested by research from other countries, we expected gender differences in stress levels. Also age was expected to be a significant factor influencing stress, that is, younger teachers were expected to experience more stress than older ones. As regards marital status, married men/ women were expected to report less stress as compared with single, divorced, or widowed men/women. Finally, the study also aimed to explore specific workplace stressors reported by teachers, especially to compare findings with those from other countries (any context-specific stressor?).

Participants
Participants included teachers of primary and secondary schools in two municipalities of Kosovo (Prishtina and Ferizaj). Data were collected during the years 2012-2013. Choice of participants and schools was random without using any inclusion or exclusion criteria. The sample consisted of 799 teachers (Prishtina n = 617, Ferizaj n = 182) estimated as 18% of the actual educational workers in Prishtina and Ferizaj. Based on gender, 33.8% of the sample were males, 65.2% females, and 1% did not report their gender; based on place of residence, 41.6% of participants came from urban areas, 50.1% from rural areas, and 8.4% did not report on this item. The age range of participants was from 20 to 65 years (M age = 42.94; SD = 11.50). The working experience reported by the teachers varied from 0 to 45 years (M = 16.25; SD = 11.49). Further characteristics of the sample are provided in Table 1.

Measures
The measures included the NSAD Stress Questionnaire (International Stress Management Association [ISMA], 2012), and one self-report questionnaire designed by the authors for the purpose of the research. Also participants were asked to report age, gender, level of education, working experience, marital status, and residence, all variables of interest in the study. The NSAD Stress Questionnaire assessed levels of workplace stress and included 25 questions, with YES/NO answers. Results were interpreted as follows: (a) 4 points or less-You are least likely to suffer from stress-related illness; (b) 5 to 13 points-You are more likely to experience stress-related ill health either mental, physical, or both; (c) 14 points or more-You are the most prone to stress showing a great many traits or characteristics that are creating unhealthy behaviors. In our interpretation, scores up to 4 are considered as low levels of stress; scores in the range 5 to 13 as moderate levels of stress; and scores above 14 are considered as high levels of stress. The questionnaire had a good internal consistency, α = .859.
The second measure was constructed for the specific purpose of the study and was based on existing research on workplace stressors. It included 12 items assessing stressors such as work overload, lack of student discipline, lack of control, low wages, and so on. Each stressor was evaluated for its impact on a Likert scale from 0 (not at all) to 4 (a lot).

Procedure
After acquiring relevant permission, questionnaires were distributed in the selected schools. Participants were ensured anonymity and confidentiality of their reports and provided informed consent to participate in the research. The administration of the questionnaires was done both individually and in groups, depending on the conditions and the directions provided by the participating schools.

Results
Results showed that 33.2% (265 participants) of the sample reported high levels of stress, 38% (304 participants) reported moderate levels of stress, and 10.3% (82 participants) reported low levels of stress.
Multiple standard regression analysis was used to predict the level of stress from a model of variables including gender, age, residence, marital status, working experience, and level of education (Table 2 and 3). For the purpose of this analysis, age, working experience, and stress levels were considered as continuous variables; also level of education was recoded as a dichotomous variable into currently married and currently unmarried participants (single, divorced, widowed). Finally, level of education was also recoded as a dichotomous variable into high school education and university-level education (university, masters, doctorate). The model reached statistical significance, F(6, 520) = 34.162, p < .00, and explained 28.3% of the variance in stress levels (R 2 = .283) (Table 2 and 3). The only variables contributing significantly to the model were residence (β = −.442, p < .00) and level of education (β = −.191, p < .00) ( Table 3).

Specific Factors
As regards the specific stressors, the most frequently reported as causing much or too much stress were inadequate wages  (36.8%), physical working environment (30.1%), and undisciplined students (26.2%). For further details please see Table 4. Chi square tests were used to assess the relationship of each specific stressor to age, gender, marital status, education, working experience, and place of residence. No significant relationship was found between the stressors and gender, marital status, or level of education. There were significant associations between place of residence and two stressors: (a) undisciplined students, χ 2 (4, N = 415) = .16, p = .019, φ = .16, and (b) lack of respect from supervisors, χ 2 (4, N = 413) = .15, p = .039, φ = .15. Both stressors were stronger in rural as compared with urban areas.
The groups with 0 to 10 years experience, 21 to 30 years experience, and 31 to 40 years experience reported work overload as being a stronger stressor, as compared with the group with 11 to 20 years experience. The groups with 11 to 20 years experience, 21 to 30 years experience, and 31 to 40 years experience reported inadequate wages as being stronger stressor (as compared with group with 0 to 10 years experience).

Discussion
The present study aimed to investigate levels of stress and explanatory factors among teachers in Kosovo. Results showed moderate to high levels of stress among teachers as compared with rates from other countries (Milczarek et al., 2009). Also out of all specific stress-related factors, only level of education and place of residence were significant predictors. These findings are in line with those reported by Lokanadha and Vijaya Anuradha (2013). More specifically, findings on level of education showed that lower levels of education were related to higher stress; this result is in line with findings reported by Mondal et al. (2011) but not those reported by Blaug et al., 2007and Mokdad, 2005(Blaug et al., 2007Mokdad, 2005;Mondal et al., 2011). A possible explanation could be that a lower level of education might make it more difficult to successfully manage some challenges of the teaching role; nonetheless, this claim should be tested in further research, as results actually showed no differences in the types of stressors by level of education. As regards place of residence, results showed that teachers from rural areas reported higher levels of stress as compared with those from urban areas. Undisciplined students and lack of respect from supervisors were suggested as two of the main stressors. These findings suggest that it might be particularly important to further assess context-specific stressors in rural areas, as it seems teachers in these contexts are faced with particular challenges (different from those in urban areas).
Although the other variables did not contribute significantly to the predictive model of stress, they showed significant correlations with stress levels. For instance, results showed higher stress levels among older participants; these results are not in line with research showing that young teachers are more stressed due to lack of experience (Abdul Majid, 1998;Bhadoria & Singh, 2010;Koruklu et al., 2012;Lau et al., 2005).Nonetheless they are in line with findings from the Health and Safety Laboratory and Health and Safety Executive (2005) report which showed that the age group between 45 and 54 years reports the highest levels of stress. These results might also be explicable in terms of the specific factors reported by the different age groups; in fact, the strongest stressors reported in the older age groups were unclear responsibilities, physical working environment, and inadequate wages. A possible explanation is that older teachers find it more difficult to adapt to change (new or unclear responsibilities). Also because older age and experience go together, older teachers might think they deserve higher wages due to their greater working experience. Finally, unsuitable physical environment seems to be a greater burden for older rather than younger teachers. The specificity of stressors within different age groups suggests that this might be an important aspect to consider in future research, as well as in targeted stress-management programs. As regards working experience, results were not in line with existent research as individuals with more experience reported higher stress levels (Abdul Majid, 1998;Bhadoria & Singh, 2010;Lau et al., 2005;Platsidou & Agaliotis, 2008). Older age and greater experience in fact seem to be risk rather than protective factors for experiencing higher levels of stress. In fact, teachers with more working experience reported inadequate wages and work overload as significant stressors; thus, their expectations deriving from the years of experience seem not to be met adequately (they probably think they should be paid more and work less than they actually do). This finding has implications for policy makers, especially in providing feedback on the current situation and identifying factors that might improve performance particularly among experienced teachers.
In terms of gender, results showed no significant differences in stress levels; thus, men and women did not differ in reported stress levels, a finding which is in line with some research in the area (Bhadoria & Singh, 2010;D. W. Chan & Hui, 1995;Fontana & Abouserie, 1993;Platsidou & Agaliotis, 2008;Ravichandran & Rajendran, 2007). In addition, both genders do not differ in terms of the specific stressors they report, suggesting no gender specificity in the appraisal of stressors in the working environment. These findings were quite interesting considering that Kosovo is a country where specific gender roles are still very prominent; thus, the approach to work and specific job-related stressors were expected to also reflect this difference. Nonetheless, results demonstrated this was not the case, at least for the specific sample investigated. Future research (maybe with national representative samples) might further investigate this aspect, as well as provide explanations for the current state.
Finally, it should also be mentioned that the three toprated stressors as reported by teachers were inadequate wages, the physical working environment, and undisciplined students. These results have clear implications for researchinformed policies in Kosovo, which should aim toward improving performance by reducing levels of stress and improving overall quality of life among teachers.

Conclusions
The levels of stress experienced by teachers in Kosovo are explicable through levels of education and place of residence. Nonetheless, there are complex relationships between different types of stressors and characteristics such as age, working experience, educational level, or place of residence. Findings have implications for future research as well as public policies aiming to reduce stress and improve quality of life among teachers in Kosovo.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.