School violence is an issue of national and international concern, especially because of its adverse impact on students' physical and emotional well-being (Akiba, LeTendre, Baker, & Goesling, 2002). However, little attention has been paid to victimization of students by their own teachers (Hyman & Perone, 1998).
Previous studies that examined this issue found that students are exposed to high levels of corporal punishment by school staff in different countries around the world, such as Australia (Delfabbro et al., 2005), Korea and China (Kim et al., 2000), and the United States (Office for Civil Rights, 2006). Khoury-Kassabri, Astor, and Benbenishty (2008), in a study of 27,316 students in grades 4 through 11 in Israel, found that almost one-third of Israeli students reported emotional maltreatment by a staff member. Close to one-fifth of all students reported being a victim of physical maltreatment.
This study examined both physical and emotional maltreatment. Emotional maltreatment may be verbal or nonverbal and includes name calling, mocking the student's appearance and disabilities, humiliating the student in front of classmates, and behaving in blatantly discriminatory ways against certain students (Benbenishty, Zeira, Astor, & Khoury-Kassabri, 2002). Hyman (1990) defined corporal punishment in school as the infliction of pain or confinement as a penalty for an offense committed by a student. Some forms of physical maltreatment may be regarded as intentional and acceptable corporal punishment. In many other instances, physical maltreatment by staff in school is not part of an educational policy advocating corporal punishment. Rather, staff may react to infractions of discipline and provocation by using various degrees of force, such as pushing, shoving, slapping, pinching, punching, or kicking (Hyman & Snook, 2000).
Previous studies that attempted to identify the risk factors for staff maltreatment have not examined teachers' characteristics. The aim of the current study was to examine the association between teachers' reports on use of violence toward their students and teachers' characteristics. More specifically, in this study, the contribution of teachers' self-efficacy, attitudes toward violence, and background variables (for example, gender, teaching experience) to the understanding of child maltreatment by teachers is explored.
MALTREATMENT WITHIN ARAB SCHOOLS
This study focused on Arab homeroom teachers in Israel. In previous works in Israel, it was found that children in Arab schools reported much more staff maltreatment than children in Jewish schools (Khoury-Kassabri, 2006), and a preliminary analysis of this study by the author (Khoury-Kassabri, 2008) found that Jewish teachers reported almost no violence toward their students. Physical violence was reported by 1.9% of teachers, and emotional maltreatment was reported by about 3%. Because of these low numbers, the current work focuses on Arab teachers. It should be noted that Arab students are being maltreated by Arab staff.' There is almost no Jewish staff in Arab schools and vice versa (Benbenishty, Zeira, & Astor, 2002). Some of these differences were attributed to the socioeconomic differences between Arabs and Jews in Israel. Arabs in Israel form a minority that is characterized by higher rates of poverty and unemployment and much lower expenditure of public funds for social services (such as education) compared with the Jewish majority (Hareven, 2002; Kop, 2004). However, previous studies found that, even after controlling for socioeconomic status, there were still clear gaps between Jewish and Arab students in reports of perpetration. The aim of the current study was to examine the contribution of the teachers' characteristics to their reports of violence toward students.
Teachers" Self-Efficacy Many studies on teachers' self-efficacy were influenced by Bandura's (1977, 1982) conceptualization of self-efficacy in his cognitive social learning theory. In 1977, Bandura introduced the concept of self-efficacy that "refers to the beliefs in one's capabilities to organize and execute the course of action required to produce given attainment" (Bandura, 1997, p. 3). Bandura suggested that self-efficacy contains two dimensions: outcome expectation, which is defined as the person's estimate that a given behavior will produce a certain outcome, and efficacy expectation, which is defined as the individual's belief that he or she is capable of demonstrating the behavior required to produce the outcome. Applying this conceptualization to the construct of teacher self-efficacy, Gibson and Dembo (1984) developed a 30-item scale that also yielded two factors. The first is teaching efficacy, which refers to outcome expectation as suggested by Bandura, and reflects the degree to which teachers believe that students can be taught, given their family background, socioeconomic status, and school conditions. The second, personal teaching efficacy, refers to Bandura's efficacy expectation, and reflects a teacher's belief that he or she has the necessary skills and knowledge to help students do better than usual and to achieve positive student change.
Many studies have found that teacher self-efficacy is an important factor for the teaching and learning process (Ho & Hau, 2004). However, Emmer and Hickman (1991) argued that a substantial amount of teacher attention is focused on behavioral problems and achieving order and cooperation in class. Therefore, teachers' efficacy with respect to students' behavioral problems should be examined (Soodak & Podell, 1996).
To the best of my knowledge, no studies have investigated the relationship between teachers' self-efficacy and their violent behavior toward students. However, studies have addressed the relationship between parental self-efficacy and use of violence. It was reported that parents who are confident in their ability to parent successfully are less likely to use physical discipline and abuse (Jones & prinz, 2005).
Furthermore, it has been argued that many teachers use corporal punishment because they lack alternative skills to deal with students, especially those who engage in repeated violence or disruptive behavior (Greydanus et al., 2003; Khoury-Kassabri, 2006). Previous studies found that maltreatment of students by teachers is most prevalent in more stressful settings, such as those in which teachers feel victimized or unsafe, and in schools with fewer resources (Khoury-Kassabri, 2006). This study tests the hypothesis that teachers lacking in resources and support to help deal with behavioral problems and those who feel unsafe may use more violence with their students.
Beliefs and Attitudes
Another argument that has been made with respect to teachers' use of violent behavior is related to teachers' attitudes toward the use of violence as a way to instill discipline and solve behavioral problems. According to Dubanoski, Inaba, and Grekewicz (1983), some teachers believe in the effectiveness of corporal punishment. They think it builds character, is essential to the development of the child's conscience, and teaches respect; without it, behavioral problems might escalate. In a study of 1,082 school principals in Washington state, 68% of principals who permitted corporal punishment believed that it is an effective method of dealing with student misbehavior, and almost all of them reported that it was effective in both the short and long term (Grossman, Rauh, & Rivara, 1995).
No published studies that empirically examined the relationship between teachers' beliefs and their violent behavior toward students were found. However, studies were found of parents and adolescents that examined the relationships between these variables. In these studies, significant relationships were found between attitudes toward violence and actual use of it (Holden, Miller, & Harris, 1999; Huesmann & Guerra, 1997).
The focus on both attitudes and self-efficacy (or "perceived behavioral control," as it is called in Ajzen's theory) gets strong support from studies that applied the theory of planned behavior; they found that these two variables play an important role in explaining human behavior (Ajzen, 1991, 2002). Perceived behavioral control was found to be well correlated with behavioral performance. With respect to attitudes, Ajzen (1991) argued that the influence of general attitude on specific actions in specific situations is generally attenuated. Therefore, the current work examines attitudes specifically with regard to the educational setting and discipline methods used by teachers to handle behavioral problems.
Several other studies have examined separately the effects of attitudes and self-efficacy on parental use of violence and found them to be significantly associated (Andreou, Vlachou, & Didaskalou, 2005; Nebbitt, 2009). For this reason, the current study aims to test whether teachers' use of violence is directly associated with self-efficacy and whether the relationship is mediated through teachers' attitudes toward the use of violence. This examination might increase our understanding of the dynamic relationships between attitudes and self-efficacy and nfight have important implications for prevention and intervention recommendations.
Sociodemographic and Contextual Factors The effects of teachers' gender, education, and years of experience on their use of violence toward students will be explored.
SUMMARY OF HYPOTHESES AND RESEARCH QUESTIONS
The current study examines the contribution of teachers' characteristics to student victimization by educational staff as reported by homeroom teachers. The following two hypotheses were proposed:
Higher levels of teacher self-efficacy are associated with lower levels of violence toward students.
Supportive attitudes toward the use of violence are associated with higher levels of violence toward students.
The following two...