A Circle of Violence: Are Burnout, Disengagement and Self-E ﬃ cacy in Non-University Teacher Victims of Workplace Violence New and Emergent Risks?

: Workplace violence (WV) is deﬁned as an intentional misuse of power, including threats of physical force against another person or group, which can cause physical, mental, spiritual, moral or social harm. The aim of this study was to describe the prevalence, characteristics and consequences of WV in a sample of Italian teachers. Our hypothesis was that WV impacted workplace satisfaction, self-e ﬃ cacy and burnout. A self-administered questionnaire was answered by 331 teachers. A total of 192 (58%) subjects reported experiencing a physical or psychological form of WV. Overall, ﬁndings conﬁrmed our hypothesis: teacher victims of WV showed high levels of burnout—both in terms of exhaustion (F = 3.96; p = 0.04) and disengagement (F = 5.85; p = 0.016), lower levels of workplace satisfaction (F = 13.24; p < 0.001) and regulatory emotional self-e ﬃ cacy—especially for negative emotions (F = 5.45; p = 0.02) compared with teachers who have never experienced WV. This investigation suggests the importance of preventing WV and o ﬀ ering support to victims. Doing so will increase teachers’ ability to manage and cope with violent behavior. Prevention and intervention may also decrease serious consequences in relation to victims’ wellbeing, and improve the general stability of the classroom, as well as motivation and academic commitment.


Introduction
In the literature, workplace violence (hereafter WV) is a well-known phenomenon that causes physical, emotional and psychological suffering [1]. The World Health Organization [2] (p. 4) defines WV as "the intentional use of power, including threat of physical force, against another person or group, that can result in harm to physical, mental, spiritual, moral or social development. It includes verbal abuse, bullying/mobbing, harassment and threats". Investigating WV and its consequences for individuals and can offer useful information to improve prevention and intervention efforts. According to the Fourth European Working Condition Survey report [3], 6% of the 44,000 European workers participating in the survey were exposed to threats of physical violence by colleagues (2%) or other people (4%). In the USA, the Occupational Safety and Health Administration [4] noted that an average of nearly two million U.S. workers reported having been a victim of WV.
According to the Bureau of Labor Statistics [5], one of the occupational sectors at greatest risk of WV is the educational sector (see also [6]). This is important because many studies involving teachers have emphasized that WV has a negative impact not only on the wellbeing of victims of violence, but also on the quality of their teaching [7][8][9][10][11]. In this sector, the phenomenon of WV has usually been investigated as bullying perpetrated among students, without any consideration of teachers as victims [12,13]. In investigations that involved teachers, findings showed that the prevalence of WV ranged between 7.5-84.8%, evidencing the great variability of the phenomenon. In 2008, Russo and In the past decade, there have been numerous social and organizational changes in Italy that have had a strong impact on the role played by teachers [52], and led to numerous new risk factors in the educational sector. Economic and political reforms have led to additional duties and responsibilities for teachers, such as managing for diversity, multiculturalism, special educational needs, and the use of new technologies [53,54]. In the meantime, Italian teachers have also experienced an increase in precarious contracts [55], a lack of social support from colleagues and school managers, an increase in the age of teachers [56], a loss of social prestige, inadequate educational equipment, an increase in parental disinterest and a decrease in students' motivation to learn [52]. These social and working conditions have a consequence on perceived stress and self-efficacy [57]: continued exposure to stressors increases the risk of burnout and disengagement [58].
Given the above, the major purpose of the current study was to describe the prevalence of WV in a sample of Italian teachers and also the consequences of WV in terms of workplace satisfaction, burnout and regulatory emotional self-efficacy. The study had a descriptive aim and the research design was cross-sectional.
The hypotheses were as follows: (a) Teacher victims of workplace violence report a lower level of workplace satisfaction than teachers who have never experienced WV. (b) Teacher victims of workplace violence report a higher level of burnout (both in terms of exhaustion and disengagement) than teachers who have never experienced WV. (c) Teacher victims of workplace violence report a lower level of regulatory emotional self-efficacy than teachers who have never experienced WV.

Participants
The participants were 472 teachers from different kindergartens and primary and secondary schools in the Metropolitan City of Turin in Northwest Italy. The inclusion criteria for participation in the survey was working as a teacher in a non-university organization and working at a school in the Metropolitan City of Turin. Therefore, we excluded cases in which the worker was a pre-service teacher, a university teacher or an employee not in charge of teaching (e.g., caretaker or administrative clerk). After the approval of the local ethical committee (University of Torino N.133260/2019), a self-administered anonymous questionnaire was distributed to all teachers in schools that had volunteered to participate in the research. Alongside a hard copy of the questionnaire, participants received an information letter and an informed consent form in accordance with the Declaration of Helsinki [59]. A box reporting the title of the project and the date of return of the questionnaires was placed on school premises. Participants had ten working days to complete the questionnaire and put it in the box. All questionnaire respondents participated on a voluntary basis and did not receive any compensation or extra credit for their participation in the investigation. Research assistants trained by the researchers collected the data.

Measures
Workplace violence (WV): The Violent Incident Form (VIF) [60] was used to investigate workplace violence (Italian version by Magnavita and Bergamaschi [61]). The instrument uses a broad definition of violence that includes different forms. Participants were asked if they had ever experienced physical aggression, verbal aggression, threatening or stalking during the last 12 months, during the course of their working life or on the way from their home to their workplace (or vice versa). Victims of WV were asked to refer to the frequency of the episodes and the characteristics of the abuse and the perpetrator (role, gender, and age) and to identify the circumstances of the aggression and its consequences.
Workplace satisfaction: The Copenhagen Psychosocial Questionnaire (COPSOQ II) [62] was used to investigate workplace satisfaction. The scale was composed of six items. An example item is "How much are you satisfied with your job as a whole, everything taken into consideration?"; the response categories range from one (=strongly unsatisfied) to five (=strongly satisfied). In this study, the Cronbach's alpha for the scale was 0.80.
Burnout: The Oldenburg Burnout Inventory (OLBI) [63] was used to investigate burnout (in this study, the Cronbach's alpha = 0.86). The OLBI includes two different dimensions to measure burnout: exhaustion and disengagement. Each dimension is measured through eight items (four positively worded items and four negatively worded items). The specific characteristics of the questionnaire are as follows: i.
The eight items of the exhaustion subscale refer to general feelings of emptiness, a strong need for rest, and a state of physical exhaustion. Example items are "After my work, I regularly feel worn out and weary" and "After my work, I regularly feel totally fit for my leisure activities" (reversed). ii.
The eight items of the disengagement subscale refer to distancing from the object and the content of work and to experiencing negative attitudes and behaviors toward work. Example items are "I frequently talk about my work in a negative way" and "I get increasingly engaged in my work" (reversed). The response categories for both subscales range from one (=strongly disagree) to four (=strongly agree).
Regulatory emotional self-efficacy: The Regulatory emotional self-efficacy (RESE) scale [64] was used to investigate regulatory emotional self-efficacy. The scale is composed of 12 items measuring the perceived ability of managing emotions (in this study, the Cronbach's alpha = 0.88). Four items measure the perceived ability to manage positive emotions, an example item is "How well can you express your happiness when something good happens to you?", while six items measure the perceived ability to cope with negative emotions; an example item is "How well can you dominate your anger when someone unfairly mistreats you?" and the response categories range from one (=completely unable) to five (=completely able).
Demographic questions: Participants were asked to indicate their gender, age, qualifications and marital status.

Data Analysis
Data were processed using version 26 of IBM SPSS Statistics (license owner: Department of Psychology, University of Turin) to produce mainly descriptive and inferential statistics. χ 2 tests were used to measure the differences between groups. Descriptive measures (means ± SD) were calculated to compare victims of WV with nonvictims. After controlling for the normal distribution of the sample and the homogeneity of variance (with Leven's test), a one-way ANOVA was calculated to test the mean score differences between victims and nonvictims on the scales of workplace satisfaction, burnout and regulatory emotional self-efficacy. Differences were considered statistically significant if p < 0.05.

Results
The questionnaire was responded to by 331 teachers (70%; µ = 2.95). A total of 110 participants (33%) were males, 185 (56%) were females, and 36 (11%) did not report their gender. Sixty-six respondents (20%) were under 40 years old, 88 (27%) were aged from 40 to 50, and 159 (48%) were over 50. Eighteen subjects (5%) did not provide their age. Most of the participants (192, 58%) experienced a physical/psychological form of WV. Fifteen subjects (4.5% of the total sample; 13 females) stated that they had been victims of physical aggression in the last 12 months, 47 (14%; 36 females) during their lifetime and four (1%; three females) on the way to or from their workplace. Forty-nine victims of physical aggression, when describing their experience, reported that the perpetrator was a male in 80% of cases, a student in 57%, a parent in 20%, a colleague in 14%, a superior in 2%, and a stranger in 4% (the remaining 3% did not specify). Eighty-two percent of victims described the perpetrator as emotionally upset. Fourteen percent considered changing their job because of aggression, and 12% actually changed it. Twenty-nine subjects (9% of the total sample; 18 females) reported that they had been victims of threats in the last 12 months, 74 (22%; 40 females) during their lifetime and seven (2%; four females) on the way to or from their workplace. Seventy-nine subjects (24% of the total sample; 40 females) indicated that they had been victims of verbal aggression at work in the last 12 months, 145 (44%; 84 females) during their lifetime and 22 (7%; 11 females) on the way to or from their workplace. Nineteen subjects (6% of the total sample; 12 females) reported that they had been victims of stalking in the last 12 months. Among victims of threats, verbal aggression and stalking, 143 described at least one episode. They reported that the perpetrator was a male in 54% of cases, a student in 19%, a student's parent in 33%, a colleague in 31%, and a superior in 17%. Thirty-seven percent of victims considered changing their job because of aggression, and 14% actually changed it.
To better understand the phenomenon and its consequences for teacher victims, a dichotomic variable (victim/nonvictim of WV) was created. In the category "victim", teachers who had experienced WV once or more were included. No significant difference was found based on gender (χ 2 = 0.015; p = 0.90) or age (χ 2 = 3.7; p = 0.8). The variable "victim/nonvictim of WV" was used to compare the two groups regarding workplace satisfaction, burnout and regulatory emotional self-efficacy.
Workplace satisfaction: A one-way ANOVA was performed to compare victims and nonvictims of WV. The Workplace Satisfaction Scale was introduced as the dependent variable. The results are reported in Table 1. Note: mean (M); standard deviation (SD); Fisher's ratio (F); p value (p); 1 The number of victims is lower than 192 (the total number of victims of WV) because subjects who did not answer all questions on the workplace violence scale were excluded from the analysis.
As illustrated in Table 1 Burnout: One-way ANOVA OLBI analyses were performed to compare victims and nonvictims of WV. The general OLBI, measuring burnout, and the two subscales of exhaustion and disengagement were introduced as dependent variables. The results are reported in Table 2. Table 2. Burnout (exhaustion and disengagement): comparison between victims and nonvictims (one-way ANOVA). Note: mean (M); standard deviation (SD); Fisher's ratio (F); p value (p); 2 The number of victims is lower than 192 (the total number of victims of WV) because subjects who did not answer all questions on the OLBI were excluded from the analysis.

M (SD) F p
As noted in Table 2 Regulatory emotional self-efficacy: One-way ANOVA analyses were performed to compare victims and nonvictims of WV. The RESE, scale measuring regulatory self-efficacy, and the two subscales related to positive and negative emotions were introduced as dependent variables. The results are reported in Table 3. Table 3.
Regulatory emotional self-efficacy: comparison between victims and nonvictims (one-way ANOVA). The results reported in Table 3

Discussion
The aim of this study was to compare workplace satisfaction, burnout and regulatory emotional self-efficacy in teacher victims and nonvictims of WV. The first interesting result pertains to the prevalence of the phenomenon. In this investigation, the prevalence of reported victimization was 58%. This finding is placed in about the middle of the range based on previous studies [6,9,10,[14][15][16][17][18][19]. In cases of victims of threats, verbal aggression and stalking, the perpetrator was a student's parent in about a third of cases and a colleague was the perpetrator in another third. Regarding the perpetrator, physical violence was perpetrated almost half the time by colleagues and superiors. While the sources of tension in the school context can vary, teachers' work takes place within an organized hierarchical structure in which relationships with colleagues and continuous adaptation to new norms are organizational factors that can increase stress. Moreover, tension can derive from difficulties in classroom management and also from "indirect users", i.e., students' parents. The reasons for this include personal expectations not being satisfied, which can, in turn, create tension, distress and aggressive behavior [65]. The negative outcomes of the experience of victimization can lead the victims to become disaffected with their job and their workplace: some of them considered changing their job or effectively changed it. As stressed by Newman and colleagues [66], teachers who worry about their safety are more likely to feel demotivated and leave the teaching profession.
As expected, victims of WV were more prone than nonvictims to experience lower levels of workplace satisfaction, higher levels of burnout (both in exhaustion and disengagement) and lower levels of perceived self-efficacy. Thus, our hypothesis was confirmed. These findings highlighted not only the higher prevalence of at least an episode of WV in this population, but also the psychological conditions associated to WV. These conditions, probably consequences of WV, affect teachers' ability to manage the classroom as a whole and also their relationships with students, students' parents and colleagues. The experience of victimization is associated to teachers' physical and psychological condition. Feeling exhausted and disengaged can increase teachers' absenteeism and negatively influence their educational ability, with important repercussions on the general stability of the classroom, on the continuity of students' experience and on teachers' ability to support students' motivation and academic commitment [67].
Moreover, as stressed by Allen [45], poor self-efficacy is a predictor of an increase in violence and bullying. Thus, the circle of violence (that is, the tendency to perpetrate violence by victims of abuse and maltreatment [68]) can affect the quality of life of both teachers and students. Students who perceive low efficacy among teachers in managing the classroom and/or in their relationships with colleagues and superiors may be able to detect their inability to cope with the dynamics of the classroom, especially those that involve bullying [69,70].
Maintaining emotional self-efficacy is particularly significant for the teaching profession, which is an emotional job. It is a type of work characterized by a need for the worker to implement emotional regulation in response to the specific demands of his/her task [71]. The teaching profession requires workers to manage their emotions in relationships with others (students, students' parents, superiors or colleagues) and to continuously adapt their behaviors and facial expressions to the specific speaker and context. The consequence of victimization on maintaining self-efficacy may increase the circle of violence: the difficulty in regulating emotions, especially negative emotions, may lead teachers to express feelings such as anger and impatience inappropriately, arousing responses with negative consequences for teachers themselves and for the institution [71].
Due to the possible adverse consequences of WV in teachers and their classrooms, it is mandatory to prevent this phenomenon. In the meta-analysis by Wilson and Lipsey [72], their findings showed that prevention programs are generally effective in reducing some behaviors, such as fighting, name calling, and intimidation. Moore and colleagues [73] (see also [74]) proposed an eight-step procedure to develop a WV prevention program that involves management, staff, the community, teachers, students and families. One of the steps is focused on policies and procedures, such as the reporting of WV episodes (step 5). Reporting is considered an important component of the prevention program. Staff members first need to be aware that WV is not tolerated; secondly, they need to know that they will receive support from their managers and that the perpetrator will be sanctioned. Without a consequence (e.g., detention, suspension, expulsion), violent behavior can be reinforced, encouraging the perpetrator to continue being violent. Espelage and colleagues [12] proposed a multisystem approach to better understand and prevent the phenomenon in this population. The program involves students, teachers, classrooms, and schools. The goal is to reduce aggressiveness and violence using methods such as mentoring (to understand emotions and how to manage them), functional assessment-based interventions (FABIs), or individual interventions to understand the motives of violent behavior [75][76][77], problem-solving techniques to reduce conflict [78,79], and school staff support for teacher victims of violence [80,81]. Additionally, Guglielmi and Fraccaroli [65] proposed methods to prevent WV in school. These methods involve both individuals and schools, and the aim is to provide individuals with resources to cope with stressful events. Regarding the school, the goal is to offer the opportunity to reflect on organizational aspects, such as norms, the climate and social support. The intervention could include mentoring with teachers who have the most experience (mentor) and those who only have a few years of experience (mentee). Another intervention addresses the creation of a professional community in schools; the coordination of social interaction among teachers to promote positive feelings (such as belonging and the acceptance of diversity); the ability to cope with conflict; and the cultivation of mindfulness, resilience and self-confidence [82]. To help teacher victims of WV, interventions (e.g., cognitive behavior therapy, counseling, social support) could focus on a reduction in consequences such as distress and rumination.
This study has some limitations. Firstly, since participants were not randomly selected from a convenience sample, the results should be interpreted with caution and should not be generalized. Moreover, gender differences were not investigated; however, it could be interesting to analyze the correspondence between the genders of victims and perpetrators and to investigate whether this variable impacts the characteristics of prevarication. Finally, this study had a descriptive design and the results are based on cross-sectional data; a longitudinal investigation would be necessary to better understand the complex interactions between variables across time. Moreover, the analysis performed allowed us to compare victims and nonvictims in different aspects, but did not permit us to assess the causal relations between variables. For this reason, the results must be interpreted with caution and considered as preliminary knowledge.

Conclusions
Teachers are workers at risk of WV perpetrated by students, students' parents and colleagues (including supervisors). This behavior may negatively affect not only teachers, but also students, classrooms and schools. Prevention and intervention are mandatory to increase teachers' ability to manage and cope with violent behavior. For this purpose, it is important to provide information about the phenomenon-and how to manage it-in training courses for both pre-service and in-service teachers. At an organizational level, intervention programs must be preceded by an analysis of the context, the prevalence of WV, the consequences for victims, the typology of perpetrators and other factors. Such an analysis allows for tailor-made intervention, which reflect the needs and the characteristics of the organization, as well as internal (school) and external (community) cultural and organizational norms and values.