Teachers ' Burnout Profile : Risk and Protective Factors

Received: September 18, 2017 Revised: November 13, 2017 Accepted: December 12, 2017 Abstract: Background: Burnout syndrome represents a factual risk for school teachers during their career. Several factors have been analyzed as stress sources enabled to menace teachers’ general well-being; nevertheless, protective factors mostly related to their personal resources may differently characterize teachers’ profiles.


INTRODUCTION 1.Teachers' Burnout and Risk Factors
Psychological and educational research from the last twenty years have considered the role of teachers' burnout in theirdaily job at school [1 -4].According to Maslach [5 -6], burnout is a multidimensional construct comprising emotional exhaustion, cynicism, and decrease of professional accomplishment.More specifically, it is a response to chronic emotional and interpersonal stressors at work, that significantly affects helping profession workers, as school teachers [1 -3], [7 -10].Although the Maslach's approach on job burnout is widely adopted to analyze workers' strain, recently some authors have analyzed burnout as a work-related syndrome to be investigated with specific relation to its sources [11].Effectively, teachers may experience suffering due to several reasons related to their personal life, work context, and relationships with students as well.By making distinction among sources, Kristensen and colleagues pointed out that the comparison of the personal burnout and work-related burnout scales may differentiate between people at risk for personal/family demands (e.g., caring, daily life, personal health, etc.) and people at risk for work conditions (e.g., conflicts with colleagues, students, etc.) [11].With specific regard to teachers' work, it is important to note thatemotional requests are strongly involved in teaching profession and educational issues leading to consider caring-for-others the main part of their job [12 -14].With this regard, teachers-students relationships are significantly related with burnout risk.Students' misbehavior, indeed, was considered the main stressful events affecting teachers' well-being.According to several authors, teachers' negative emotional experience occurring when students misbehave was strongly related to teachers' burnout [1,10,15].Consistently, some studies showed that the stronger the negative emotions, the more intense the burnout symptoms reported by the teachers [16 -17] At the same time, teachers with high burnout levels differently perceive and express negative emotions, such as anger, shame and sadness, when compared to their colleagues with lower burnout [3].Findings from a recent study showed that burned out teachers were more at-risk to mis-perceive negative emotions during a verbal conflict with their students [8 -9].More specifically, burned out teachers perceive one's own and students' negative emotions as more intense than non-burned out teachers.
Accordingly, teachers with low levels of social and emotional competence, as typically found out in burned out teachers, show low effectiveness in their daily job in terms of absenteeism, turnover, role conflict, low self-esteem and job satisfaction [18 -21].
Furthermore, several authors have analyzed socio-demographic and career-related variables associated with at-risk teachers' profile [22 -28].Teachers' gender was the mainly studied variable, with mixed findings.While some studies didn't show significant associations, others reported that women had higher burnout risk than their men colleagues [16] [18] Furthermore, Shaheen and Mahmood found blended results showing that while female teachers were more exhausted, male teachers felt more depersonalized and less accomplished [27].Similarly, teachers' age showed incoherent results: some studies showed that younger and less experienced teachers had higher burnout risk, conversely, others researchers showed the same findings for older and more experienced teachers [22,25].
Regarding teachers' career, commonly studied variables included job roles and extra-mansions.Job roles, namely whether teacher is employed with regular class students' or with children with special needs, showed that special education teachers had high burnout risk and low job satisfaction when compared to regular class teachers.With regard to extra-mansions (or extra-role behaviors) it refers to behaviors and roles that go beyond the teaching role itself, usually not paid (teachers do not receive an extra reward for performing them) [29 -30].According to some authors the way teachers participate and respond to extra-mansions requests may be due to efficacy beliefs, work identification and personal engagement [29 -30].Extra mansions and related behaviors may influence (and be influenced by) how teachers perceive their own well-being and stress levels [29 -30].
To sum up, the existing literature addressing teachers' burnout takes into account the role played by emotional demands in teaching profession, which may enhance their risk level to feel exhausted in their job.Moreover, there is a wide agreement to consider that teachers' burnout needs to be analyzed by considering their socio-demographic variables as well as job characteristics, even though findings remain inconsistent.At the same time, in order to deeply analyze teachers' risks, the recent research suggests taking into account teachers' positive resources able to protect them from risks factors.Effectively, it is possible to think that protective factors enhance teachers' abilities to cope with work-related stressors.

Protective Factors in Teachers' Burnout
Current literature indicates several positive dimensions that could act as protective factors towards burnout enhancing teachers' well-being and lowering the risk to develop a stress/burnout condition.More specifically, several studies addressed the associations between teachers' burnout levels and such dimensions including job satisfaction, selfand collective efficacy, positive attitudes toward their profession.Job satisfaction, namely how content an individual is about his/her job experience, plays a crucial role in promoting well-being [8,9].Similarly, self-efficacy at work may positively influence teachers' job experience [31 -35].Indeed, teachers with high levels of self-efficacy are more able to ask and receive support from their colleagues and principals and manage students' behaviors [33 -35, 39].Consistently, several studies, showed that higher self-efficacy is inversely associated to burnout [31, 33, 36, 39 -40].Teachers with high self-efficacy are less at-risk for anxiety, burnout, suffering from job demands and work-related stress in general and are more likely to perceive themselves as satisfied, motivated and creative professionals [31,32,41].
Nevertheless, the role played by protective factors may be not directly related to teaches' positive outcomes.Some studies indeed, showed that collective efficacy (namely, the belief that one's own organization is able to manage daily demands and tasks) did not reduce teachers' burnout levels neither promote their well-being [36, 42 -43].Effectively, while self-efficacy is more related to teachers' personal and psychological resources, collective efficacy is significantly affected by school-related variables, such as leadership styles and schools' mission and vision [44].
With regard to teachers' positive attitudes toward their profession, previous studies have analyzed teachers' engagement with their work as well as their reported positive relationships with students and colleagues., Engaged teachers show high motivation toward their professional tasks as well as attitude to enhance their professional abilities [14,45].Finally, positive attitudes toward their profession can be evaluated by taking into account teachers' positive relationships at school.More specifically, teachers with higher scores in social and emotional competence are more effective in relationships at school, with students and colleagues [46].Moreover, they are able to regulate their own emotions and behaviors in emotion-related events at school, as well as to modulate their teaching practices and strategies according to the characteristics and abilities of their students [4, 47 -49].According to Diener and colleagues emotion regulation could be synthesized in terms of hedonic balance which refers to the balance between positive and negative emotions [50 -52].This construct is currently considered as the affective component of subjective well-being [53].
Overall, despite current literature indicates several potentially protective factors, high levels of each mentioned dimension (i.e., job satisfaction, efficacy beliefs, emotion regulation/hedonic balance) don't automatically lead to lower perceived of professional burnout.Consequently, it is not certain that positive factors act as protective factors.It is possible, for example, that other conditions (e.g., macro-social changes, organizational context, etc.) modify the protective power of some of the mentioned variables [54].For these reasons, studying burnout profiles may help shedding new light about what differentiates teachers at risk for burnout from their colleagues, and better understanding how they perceive their job experiences [6].

Aims and Hypotheses
The current study aimed to define different teachers' profiles based on their burnout levels and attitudes towards job.More specifically, we expected that burned out teachers showed lower levels of job satisfaction, self-efficacy, attitudes toward professional growth, collective efficacy, positive emotions (regarding both students and professional role) and hedonic balance (regarding both students and professional role), and higher levels of negative emotions (regarding both students and professional role).Furthermore, we expected that higher age, higher experience and being involved in extra-mansions would characterize at-risk teachers.

Participants
Two hundred and sixty-six secondary school teachers (F=69.1%)were involved.Teachers' age ranged from 26 to 65 years old (M=48.95;SD=8.31).Teachers' years of experience ranged from 1 to 41 years (M=21.72;SD=10.36).The most part of them were regular (87.1%) and permanent (83.1%) teachers, while the rest of the sample was a special education teacher (around 13%), with a temporary contract (around 16%).Teachers were recruited in two ways: 1) at school, because the principal agreed to take part to the study as whole community; 2) online, via forum, facebook pages, and/or mailing lists.Consequently, data were gathered face-to-face, during collective meeting with teachers in each school, or online, by Google modules.In each situation, teachers were instructed by informed consent that they could leave the study at any time, and they could ask the researchers for further information either in personal contact, or using a specific online module.Data were gathered from September 2015 to March 2016.

Instruments
Copenhagen Burnout Inventory (CBI [11,22]).The CBI evaluates three dimensions of burnout level: 1) Personal burnout, namely how often teachers felt drained, or without energy (6 items; ex., How often do you feel tired?; How often do you feel worn out?); 2) Job burnout, namely how often teachers felt their job were stressing them out (7 items; ex., Do you feel that every working hour is tiring for you?; Does your work frustrate you?) Students-related burnout, namely how often teachers thought working with students was exhausting (6 items; ex., Do you feel that you give more than you get back when you work with students?;Do you find it hard to work with students?).Overall, it is, thus, composed by 19 items measured with a 5-points Likert scale (from 1= Never, to 5=Always).Alpha's Cronbach were respectively: .889,.858,.813.[45].For the current study, we used the following sub-scales: job satisfaction, emotions related to relationships with students and to the professional role, attitudes towards professional growth, and selfefficacy.Job satisfaction is measured by 5 items: teachers were asked to define how much they agreed with each item on a 7-point Likert scale (1=Strongly disagree, 7=Strongly agree).Emotions related to relationships with students, as well as emotions related to the professional role are measured by 30 items: teachers were asked to define how frequently they perceived certain emotions, by using a 5-point Likert scale (1=Almost never, 5=Almost always).Starting from these scales, two further scales were measured: Hedonic balance related to Students (HB-Student) and Hedonic balance related to Professional role (HB-Professional Role), both measured as the difference between positive and negative emotions [50 -52].Consequently, when the hedonic balance is high, the respondent perceives more positive than negative emotions.Attitudes towards professional growth are measured by a 16-item scale: teachers were asked to define how much they thought they could improve specific teaching abilities thanks to experience or training opportunities, using a 9-points Likert scale (1=not improvable, 9=totally improvable).Self-efficacy is measured by 24item: teachers were asked to evaluate how much they felt effective when approaching daily teaching tasks, on a 10points Likert scale (1=not effective, 10=totally effective).Alpha's Cronbach were .835 to .975.[53].This 9-item scale measures how much teachers feel their school is effective in managing daily educational tasks.Teachers were asked to evaluate their agreement with items on a 7-point Likert scale (1=Totally disagree, 7=Totally agree).Alpha's Cronbach was .952.

School collective efficacy scale
Finally, socio-demographic (i.e., age and gender) and job-related characteristics were collected by ad hoc questions.With regard to job-related characteristics, teachers were asked about job role (regular class vs. special needs), enrollment positions (tenured vs. temporary), extra mansions at school (yes vs. no).

Analyses Plan
In order to distinguish two or more groups of teachers according to their burnout levels, a TwoStep Cluster analysis (SPSS, v. 20) was performed.TwoStep analysis is effective within social studies research, above all when inserted variables are continuous.This analysis unfolds in two steps.The first one forms pre-clusters of densely-packed records, by merging cases that have identical or very similar patterns of responding on the variables of interest.The main aim of this step is to reduce the number of records that will be compared in the formal clustering phase.The second step uses a standard agglomerative hierarchical clustering method to group these pre-clusters into distinct clusters.Each pre-cluster is merged with the two closest pre-clusters to create a single cluster.Subsequently, this cluster is merged with the next two closest pre-clusters to create a larger cluster, and so on, until all pre-clusters are merged into one cluster (that includes all the participants).The program then calculates Bayesian Information Criterion (BIC) values for all the produced cluster solutions, to identify how well each successive cluster solution includes distinctive clusters.Finally, it automatically calculates the optimal cluster number based on the point where an additional cluster does not appreciably improve the BIC value.The main benefits of the TwoStep method are that it manages large data sets more effectively than traditional methods (namely, k-means and expectation-maximization) (SPSS Inc., 2001).
In order to verify whether emerging cluster groups showed different profiles (i.e., protective factors, demographic, and job-related variables) a set of ANOVAs (for continuous variables) and Chi-square analysis (for categorical variables) were run.Protective factors included: self and collective efficacy, job satisfaction, emotions related to job role and students, attitudes toward professional growth, hedonic balance.Demographic and job-related variables included: age, gender, years of experience, job role, job contract, extra mansions (other than teaching).

RESULTS
Table 1 shows means and standard deviations for the whole sample and for two teachers groups obtained by twostep cluster analysis which split sample in two groups: with low and high levels of personal, professional and studentrelated burnout which were labeled, respectively, "at-risk" teachers (N= 140) and "non-at-risk" teachers (N= 136).Table 2 shows one-way ANOVA results.At-risk and not-at-risk teachers significantly differ for: self-efficacy (F 274,1 =9.998; p=.002); positive emotions related to job role (F 274,1 =5.846; p=.016); relationships with students (F 274,1 =4.800; p=.029); and hedonic balance related to their professional role (F 274,1 =2.189; p=.050).Moreover, at-risk and not-at-risk teachers didn't differ for: job satisfaction, attitudes toward professional growth, collective efficacy, negative emotions and hedonic balance regarding relationships with students and negative emotions regarding professional role (p>.05).Table 3 shows for each cluster group their means and frequencies differences for their socio-demographic and jobrelated variables obtained by one-way ANOVAs and Chi-Squares analysis.No significant differences have been found for age and years of experience (p>.05).At the same time, Chi-Square analysis highlighted significant difference between at-risk and not-at-risk teachers for extra mansions (Χ 2 =4.346, p<.05).Specifically, teachers which are employed in several professional mansions fall in at-risk cluster.Gender, job role and job contract, instead, did not significantly characterize teachers' profiles.

DISCUSSION
The present study aimed to define teachers' profiles on the basis of their burnout risk and attitudes towards job.We expected to find different at-risk teachers' profiles distinguished by their job satisfaction, self-efficacy, attitudes toward professional growth, collective efficacy, positive emotions (regarding both students and professional role) and hedonic balance (regarding both students and professional role), and higher levels of negative emotions (regarding both students and professional role).
Results partially confirmed our hypotheses.We identified two different profiles, named "at-risk" and "not-at-risk", on the basis of their burnout levels.Starting from these profiles, we analyzed more deeply the differences and similarities between at-risk and not-at-risk teachers, in order to define protective factors that are differently associated with their profiles .Techers' profiles significantly differed with regards to measures of self-efficacy, positive emotions and hedonic balance towards their professional role, and positive emotions in relationships with their students, which were all higher in teachers with low burnout level (not-at-risk profile).These results confirmed our hypothesis that atrisk teachers showed lower levels of self-efficacy, positive emotions (regarding students and professional role) and hedonic balance (regarding professional role).Surprinsingly, no differences have been found between the two profiles regarding: job satisfaction, attitudes toward professional growth, negative emotions (towards both relationships with students and professional role), hedonic balance regarding relationships with students and collective efficacy.Conversely to our hypothesis, indeed, teachers at risk show the same levels in these attitudes towards job than not-atrisk colleagues.Furthermore, our hypothesis about the role of demographic and career-related variables was partially confirmed.Indeed, while no differences have been found regarding age, gender, years of experience, job role, and kind of enrollment., at-risk teachers were more frequently engaged in the extra mansions than not-at-risk colleagues.The following discussion will focus on similarities and differences as emerged in our findings.

At-Risk and Non-At-Risk Teachers: Two Different Profiles
Focusing on the different patterns resulting from the cluster analysis, the main result was that high levels of positive emotions at school, positive relationships with students, hedonic balance towards professional role and self-efficacy protects from burnout; conversely to previous research, factors usually considered as protective, including (lower) negative emotions, job satisfaction, collective efficacy and attitudes toward professional growth, do not have a protective role towards burnout levels.Particularly interesting is the difference between the profiles regarding the protective role of positive emotions, opposed to the neutral role of negative ones.
On one hand, this is consistent with previous research, stating that positive and negative emotions are not two opposite ends of a bipolar continuum, but two independent factors with differentiated biological and behavioral patterns, and consequently, have an independent role on people's perceptions, beliefs and behaviors [55 -56].On the other hand, the differentiated role of positive and negative emotions may inform about the impact of perceived emotions at school on teachers' burnout.Several studies underline that people tend to show a negativity bias when it comes to perceive and notice emotions [56 -57].This implies that, to overcome the perception of negative emotions in daily life, in favor of positive ones, a conscious effort should be made [12, 58 -59].People with high levels of self-reported positive emotions seem to be more able to buffer negative ones and acknowledge personal resources and well-being states [60 -61].With regard to student-related positive emotions, it may be that positive relationships with students protect teachers in the not-at-risk profile from developing burnout.This is consistent with the idea that burnout is not due to the people (clients, patients, students) with whom professionals primarily relate when they are at work, but to the work conditions they experience.With regard to professional role-related positive emotions, according to previous studies, when teachers notice social and organizational resources within their job context, their risk to develop burnout symptoms is reduced [2,62] On the contrary, perceiving high job demands (such as high workload or role stress and ambiguity) increases the risk to develop burnout symptoms [2,62].Therefore, positive emotions, related to both relationships with students and job conditions, could act as a protective mechanism against the risk of developing burnout, Moreover, hedonic balance regarding professional role (but not student-teacher relationships) distinguished not-at-risk from at-risk teachers.Therefore, it seems that teachers at-risk for burnout have more difficulties regulating emotions related to professional role than not-at-risk colleagues.Consistently, higher levels of perceived positive emotions are linked to higher emotion regulation, namely the ability to maintain, decrease and increase positive and negative emotions [63 -66].Teachers with higher emotion regulation report higher personal accomplishment (i.e., higher quality of social relationships and lower burnout [63, 67 -68]).Therefore, it is possible that at-risk and not-atrisk teachers differ in terms of noticing and reporting positive emotions, but not negative ones, not-at-risk teachers are more engaged in positive interactions and positive attitudes toward their profession, than at-risk colleagues.Not-at-risk teachers could perceive and evaluate more positively their relational and professional experience at work.Another interesting point that distinguished at-risk from not-at-risk teachers was the engagement in extra-mansions at work.Previous studies showed that teachers' positive emotions are strongly related to teachers' commitment at school, in terms of motivation to engage in extra-mansions behaviors, personal initiative and pursuit of professional quality [69].Despite this, at-risk teachers who took part in our study are specifically characterized for covering extramansions at school.Previous research showed a connection between work overload, and specifically extra-mansions, and job-related stress [29 -30].Moreover, feeling overloaded at work may have a role in reducing the ability to maintain hedonic balance and positive emotions related to professional role.Some authors, indeed, showed that professional identification and related emotions are related to how teachers manage their workload [29 -30].Despite needing further researches, this point may shed new light about the effect of job overload not only on teachers' emotions or burnout, but even on how teachers perceive themselves as professionals (for example, with regard to their motivation to engage in extra-mansions, and the efficacy beliefs related to them).Finally, this research indicates another protective factor: self-efficacy.Our results show, consistently with previous researches that feeling effective as a professional protects from developing burnout symptoms [10, 31, 35, 70 -71].More specifically, some studies report that teachers with high self-efficacy tend to perceive students' misbehavior and work conditions as less relevant for their stress levels [31].

Similarities Between Teachers' Profiles
With regard to neutral factors, these included collective efficacy, job satisfaction and attitudes towards professional growth.Concerning collective efficacy, previous studies state that the professionals' beliefs about the capacity of one's own organization to prevent and manage regular and special tasks influence well-being, preventing burnout from developing [71 -72].At the same time, recent studies showed that collective efficacy does not always have a role in promoting teachers' wellbeing [36].Several considerations could be done with this regard.Firstly, according to Bandura context and circumstances may have a role in constructing efficacy beliefs, as the conditions in which a task is performed may influence the outcomes [42].With regards to the specific contemporary context in which Italian teachers work, we launch the hypothesis that events at the macro-level within Italian education reforms might have somehow influenced these results.Data were gathered right after the implementation of a new educational reform (promulgated in July 2015) that highly impacted Italian schools both as regards the didactic programming and the schools' organization.Among the most important changes, schools have been required to find trainings (consistent with the studied subject) for the students from the last two years in firms, organizations or associations, to be done during school hours; school principals are asked to hire teachers autonomously; chairs are divided among a higher number of teachers.Secondly, as reported by Malinen and Savolainen the general organization of the school system may have a role in predicting the effects of collective efficacy on teachers' well-being perceptions [36].Bracci reported that the increasing autonomy of public Italian schools from ministry and regional authorities is creating a confusion web of accountability, in which individualistic and collectivistic cultures and practices are merge together [73].It is possible that the characteristics of the Italian educational system make individual-level efficacy more significant in influencing burnout perceptions.Further researches are needed in order to verify whether contextual variables may influence teachers' sense of community and, consequently, how effective they perceive their school as a community.
With regard to job satisfaction, it is well-established its association with burnout [74 -75].Effectively, the more teachers' experienced distress, the more they feel less involved and satisfied with their job.Despite this, in the current study at-risk teachers are not significantly less satisfied than not-at-risk colleagues.These findings suggest that, for this sample of Italian teachers, job satisfaction is not a discriminating variable for burnout risk.Again, information about the Italian social context may shed new light on this result.Apart from dissatisfaction coming from managing and implementing the recent reform, OECD TALIS report (2013) showed that Italian teachers perceive their professional role as underestimated within the society.Recent studies showed that when teachers perceive poor social recognition, they are dissatisfied about their job [76].
Teachers' attitudes toward professional growth do not have a role in differentiating at-risk from not-at-risk teachers, too.At the best of our knowledge, there are no previous studies acknowledging the role of this variable in preventing stress or burnout.Despite this, some authors suggest that professional education may influence teachers' well-being.Further researches are needed in order to better conceptualize teachers' interpretation of their professional growth [77].
Overall, this study suggests that when addressing teachers' burnout, it is necessary to take into account risk factors as well as positive and protective dimensions in a wider perspective, in which contextual variables may play a significant role.Recently, some authors have analyzed teachers' burnout from an ecological point of view [78 -79].In this perspective proximal and distal systems of teachers' job may differently impact teachers' burnout development.Future research should consider how teachers' characteristics (e.g., socio-demographic variables, individual's emotional competence, professional career) interact with school-related factors (as proximal system) as well as society dimensions linked to teaching profession (as distal system).

LIMITATIONS
This study has several limitations.A first consideration must be made about contextual influences.As stated above, some results could be better deepened and interpreted in light of the examination of contextual variables as, particularly important for the contemporary Italian teachers, the recent implementation of the educational reform.In this regard, a longitudinal study would inform better about the effects of the socio-political changes on teachers' perceptions of their profession and well-being at school.Moreover, it is possible that the multiple methods used to gather data may have had an impact on participants' answers to self-report questionnaires: a more homogeneous method would have eliminated this possible bias.