Abstract
Peer violence in school has become a major issue for schools around the world. The present study examined the impacts of cultural settings and of protective individual attributes on peer bullying and victimization in school. These protective attributes were self-esteem, sense of autonomy, emotional regulation, and individual resilience. Participants were 112 Jewish and 55 Arab Bedouin pupils 10 to 11 years old. It was hypothesized that Jewish pupils would score lower than Bedouin pupils on bullying and on victimization, and will score higher than them on these protective individual attributes. It was also hypothesized that despite these differences, the investigated attributes would correlate with reduced peer violence in both groups. It was hypothesized further that individual resilience will be the major predictor of both bullying and victimization in both groups. Results have generally supported these hypotheses, suggesting alternative ways for curtailing peer aggression in school.
Peer violence was identified as a major risk factor in European (Olweus, 1991), American (Hoover, Oliver, & Hazler, 1992), and Israeli (Astor, Benbenishty, Vinokur, & Zeira, 2006) schools. Research on school violence often divided students into three categories: bullies, victims, and nonaggressive individuals (Olweus, 1994; Veenstra et al., 2005; van Dijk, Poorthuis, & Malti, 2017). These three categories cannot be regarded as mutually exclusive because victims in one context may become bullies under different circumstances, and nonaggressive students may turn out to be the targets of peer violence.
Culture and Peer Violence
Previous studies examined the effects of cultural norms on the level of peer violence in school (Aronson, Wilson, Akert, & Sommers, 2016; Benbenishty & Astor, 2005; Dwairy, 2004). Johnson et al. (2017) argued that cultural norms strongly affect school culture and peer violence in the schools. Astor et al. (2002) also suggested that differences in school violence should be attributed to school climate, that is, to risky peer-group behaviors, and personal victimization, rather than to general cultural differences. Other researchers emphasized the importance of a strong sense of connectedness to school, family, and community institutions, as well as involvement in extracurricular activities in determining peer violence (e.g., Evans & Smokowski, 2015; Resnick, 2000). Majority and minority schools employing clear, consistent, and fair rules tend to experience lower levels of violence (Khoury-Kassabri, Benbenishty, Astor, & Zeira, 2004). Similar results were obtained in other victimization studies conducted in different cultural settings (Adams, 2000; Welsh, 2000).
The present study investigates ethnic differences and similarities in expressions of violence of Bedouin and Jewish school children, and their associations with personality attributes. Arabs constitute the largest minority group in Israel, and the Bedouins comprise a subgroup of this minority. Despite internal and external modernization pressures (Gilat, Ezer, & Sagee, 2010), this minority group generally prefers to live in separate communities and to retain most of its traditional norms and perspectives of life (Gueron-Sela, Atzaba-Poria, Meiri, & Marks, 2013; Shmueli & Khamaisi, 2011). The Bedouins preserved their identity as a collectivistic society, in which loyalty to one’s group and family, interdependence, and cooperation are preferred over individualism, autonomy, self-reliance, and striving to achieve personal goals (Al-Krenawi, 1999; Triandis, 1989, 1995). While the Jewish population is oriented toward individualism, autonomy, and personal growth, the Bedouin society underscores relational-oriented values such as belongingness and loyalty (Gardner, Gabriel, & Lee, 1999). In this collectivistic and patriarchal society all family members are subordinate to the authority of the father, and respect his right to make decisions which influence their lives (Al-Krenawi, 1999). Recent Israeli studies reported higher levels of victimization and bullying in Arab and Bedouin schools compared with Jewish schools (Khoury-Kassabri et al., 2004). Similar results were obtained in other victimization studies conducted in different cultural settings (Adams, 2000; Welsh, 2000). We hypothesize, therefore, that Bedouin pupils will score higher than Jewish school children on both victimization and bullying, and will score lower than Jewish pupils on personality attributes pertaining to pupil’s autonomy, individualization, and self-esteem. Despite the emphasis of the research literature on the collectivistic nature of the Bedouin society, we assume that these individual attributes of Jewish as well as Bedouin pupils will correlate with their bullying and victimization scores and will predict them.
Protective Psychological Factors
The present study focuses on protective psychological factors which may reduce peer bullying and victimization (Aronson et al., 2016; Berger & Caravita, 2016). As indicated above, school violence is affected by social factors such as limited social opportunities (Elbedour, ElBassiouny, Bart, & Elbedour, 2013; Sherer & Karnieli-Miller, 2004) or the nature of the school climate (Berger & Caravita, 2016; Johnson et al., 2017). Bernat, Oakes, Pettingell, and Resnick (2012) argued further that peer violence in school is substantially contingent on the students’ psychological attributes. School violence was found to be negatively associated with students’ self-efficacy, low sense of helplessness, realistic planning, and high intelligence (Lösel & Farrington, 2012). Reduced student violence was also linked to a higher level of emotional regulation (Garofalo, Holden, Zeigler-Hill, & Velotti, 2016), higher self-esteem (Lee, 2014; Teng, Liu, & Guo, 2015), and perceived level of autonomy (Kaplan & Assor, 2012). The present study examines the role of these three protective student attributes as well as the contribution of individual resilience in reducing peer violence in school.
Self-Esteem
Self-esteem pertains to an individual’s assessment of his or her own worth. Individuals with high self-esteem are more satisfied with life (Myers & Diener, 1995; Proctor, Linley, & Maltby, 2009), experience more positive emotions (Pelham & Swann, 1989), adjust better to school (Kurman, Rothschild-Yakar, Angel, & Katz, 2018), and are less likely to be anxious or depressed (Crandall, 1973). Lee (2014) further found that level of self-esteem negatively associated with children’s violence.
Sense of Autonomy
Student autonomy often reflects parental support of independence, parents’ attempts to promote intrinsic values in their children, and providing a rationale and demonstrating intrinsic values that allow a sense of choice about behavioral enactment. Parental recognition of children’s feelings allows children to concurrently feel close to their parents and to have a sense of volition when enacting their parents’ expectations (Assor, Roth, & Deci, 2004). Parental autonomy support promotes internalization of desired behaviors by explaining and demonstrating the value of those behaviors to their children, as well as by respecting their perspectives, including their negative feelings and disagreements about these behaviors (Deci & Ryan, 2000). Barber, Stolz, Olsen, Collins, and Burchinal (2005) found that this kind of parental support was linked with adolescents’ social initiative, and with lower antisocial behavior and violence. Kaplan and Assor (2012) have found further that teacher support of student autonomy may reduce school violence in the long run.
Emotional Regulation
Emotional regulation, defined as the ability to control and manage one’s feelings, is an essential component of social adjustment. Children who were able to regulate their emotions paid more attention, worked harder, and achieved more in school. They were better behaved, more capable of resolving conflicts with their peers, more caring toward others, and exhibited lower levels of stress (Gross, 2015; Thompson & Meyer, 2007). Inappropriate emotional regulation was often associated with breaking social norms (Southam-Gerow & Kendall, 2002). Gross (2015) contended that emotional regulation could be achieved by cognitive reappraisal or by emotional inhibition. Reappraisal regulation was more strongly linked to positive outcomes than inhibited expression of emotions.
Individual Resilience
Aggression is strongly associated with psychological stress. Stressful school environments (Meyer-Adams & Conner, 2008) as well as stressful home environments (Wiesner & Windle, 2004) were associated with increased violent behavior in school. Continuous stress caused emotional vigilance in children, which was likely to be expressed by aggressive reactions in the long run (Cassidy & Taylor, 2005). Furthermore, compared with children whose antisocial conduct breached in adolescence, those whose antisocial conduct erupted earlier had a worse prognosis in terms of future aggression and other antisocial activities (Patterson, Forgatch, Yoerger, & Stoolmiller, 1998). We assumed that individual resilience to stress will play a major role in decreasing peer violence. This assumption concords Block’s (1971) conception of ego resilience which claimed that resilience under-control is a risk factor for violence and antisocial behavior.
A recent literature review indicated that resilience has several different definitions (Bonanno, Romero, & Klein, 2015). Previous definitions of psychological resilience tended to define it in terms of a low level of deficit-based pathogenic processes, such as levels of distress symptoms after a potentially traumatic event (Bonanno, Rennick, & Dekel, 2005). Resilience theory gradually shifted away “from looking at risk factors that led to psychosocial problems to the identification of strengths of an individual” (Richardson, 2002, p. 309). Resilience was thus defined as “protective factors which modify, ameliorate or alter a person’s response to some environmental hazard that predisposes to a maladaptive outcome” (Rutter, 1987, p. 316; 2006), or as people’s ability to survive stress and adversity (Bonanno, 2004).
In their quest for uncovering the basic determinants of psychological resilience, Masten (2011) and Hadi, Llabre, and Spitzer (2006) questioned the position that resilience reflects mainly protective factors. They argued that understanding resilience to extreme adversity requires a concurrent examination of positive adaptation and pathological processes. In line with this reasoning, Kimhi and Eshel (2015) maintained that a sense of resilience involves a constant collision of positive forces and pathogenic elements, both of them should determine resilience simultaneously. They defined resilience accordingly as “the current psychological outcome of the ongoing conflict between stress-resistant strength and vulnerability of an individual or a community, following a potentially traumatic experience” (p. 181).
Individual Strength and Vulnerability
Pupils’ strength component of resilience was defined by their satisfaction with their everyday lives (Eshel, Kimhi, Lahad, & Leykin, 2017; Kimhi, Eshel, Leykin, & Lahad, 2017). This indicator of individual resilience was positively predicted by sense of coherence, self efficacy, and social support of a sample of 3,415 Israelis (Eshel, Kimhi, Lahad, Leykin, & Goroshit, 2018), and positively correlated with community resilience of Israeli adults who encountered continuous security stress (Eshel, Kimhi, & Marciano, Accepted). The vulnerability component of resilience was determined by level of distress symptoms. Painful events and adversities could shake pupils’ basic sense of security and give rise to posttraumatic symptoms, which might linger long after the termination of these events. Distress symptoms may take the form of emotional and behavioral problems, depression, anxiety, and grief (Dyregrov, Gjestad, & Raundalen, 2002; Thabet & Vostanis, 2000). Previous studies validated this index of resilience indicating that individual resilience which was determined by strength to vulnerability ratio (SVR) constituted a major predictor of future sense of coherence and social support, under continuous terror attacks in Israel (Eshel et al., 2017). Furthermore, a regression analysis, in which the effects of several predictors were controlled for each other, indicated that SVR significantly predicted the adjustment of Arab widows to the adversities of widowhood, whereas other predictors, such as well-being, social support, or sense of coherence, did not add significantly to the explained variance of this adjustment (Yasien-Esmael, Eshel, & Rubin, 2017). SVR also mediated the links between demographic predictors and war-related personality attributes (Eshel, Kimhi, & Goroshit, 2016). The present study concentrated on the role of individual resilience in decreasing peer violence, assuming SVR would serve as a major predictor of restrained violence among school children.
The following hypotheses were examined:
Method
Participants
The participants included 112 Jewish students (mean age = 10.95 years, SD = .72) and 55 Arab Bedouin schoolchildren (mean age = 10.70 years, SD = .79) in Grades 5 and 6 enrolled at two Jewish and two Bedouin Israeli schools. The Jewish sample included 70 girls and 42 boys, and the Bedouin sample consisted of 28 girls and 27 boys.
Instruments
Peer violence
The Peer Violence–Victimization in School Scale was devised by the Israeli National Authority for Measurement and Evaluation in Education (2016). Its 15 items refer to different violent acts ranging from cursing or pushing to a physical harm caused by an instrument (e.g., “Did you bring a knife or pen-knife to school to defend yourself or attack someone?”). Cronbach’s alpha reliability for the violence scale in the present study was .89. Additional 15 items pertained to victimization (e.g., “Did some student attempt to convince others not to talk to you or not to be your friend?”). Cronbach’s α for the victimization scale was .89. The items of both scales were rated by a scale ranging from 0 = never occurred to 4 = happened 4 times or more during the last month. These scales were validated in several large-scale Israeli studies (e.g., Benbenishty & Astor, 2005).
Self-esteem
Rosenberg’s (1979) self-esteem questionnaire was employed (sample items: “I feel I have a number of good characteristics”; “In general, I am satisfied with myself”). The questionnaire contains 10 items rated on a 4-point scale ranging from 1 (totally disagree) to 4 (totally agree). The scale’s internal consistency in the present study was α = .86 for both groups.
Sense of autonomy
Pupil’s attitude toward autonomy versus dependence was assessed by the Parenting Attitudes Toward Autonomy Support Scale devised and validated by Assor et al. (2004). The scale’s 20 items are rated on a 4-point response scale ranging from 1 = does not agree at all to 4 = agrees very much (e.g., “A child has a right to have an opinion of his or her own, and he or she should be given an opportunity to express it”). Previous results indicated that when parents provided an autonomy-supportive socializing context, children tended to integrate the attributes and values endorsed by the parents (Grolnick, Deci, & Ryan, 1997). Cronbach’s alpha was .68 for Jewish and Bedouin pupils.
Emotional regulation
The 10-item emotional regulation scale (Gross & John, 2003) is aimed at assessing two aspects of this concept: emotional regulation-reappraisal (six items, for example, “I control my feelings by changing the way I think about situations”) and emotional regulation-inhibition (four items, for example, “I keep my emotions to myself”). The response scale ranges from 1 (very untrue for me) to 7 (very true for me). The Cronbach reliabilities of emotional reappraisal were .84 for Jewish students and .80 for their Bedouin peers. Alphas of the emotional inhibition scale for Jews and Bedouins were .71 and .66, respectively.
My life today
The nine-item self-report My Life Today scale developed and validated by Eshel, Kimhi, Lahad, and Leykin (2016, 2016, 2017) measured individual strength. The 6-point response scale ranges between 1= not good at all and 6 = very good. The scale’s current reliability was α = .88 for both samples.
Distress symptoms
The Brief Symptom Inventory (BSI; Derogatis & Savitz, 2000) was used to assess anxiety, depression, and somatization symptoms. The response scale for this 18-item inventory ranges from 2 = not suffering at all to 5 = suffering to a very large extent. The scale’s current Cronbach’s alpha value was .83 for both samples.
Individual resilience
The IND-SVR resilience score was determined by dividing the mean standardized My Life Today score by the mean standardized BSI score (Kimhi & Eshel, 2015).
Procedure
The research questionnaire was administered to fifth and sixth grade pupils who were taking extracurricular courses in science, medicine, technology, and art. The participants’ parents signed an informed consent form agreeing that their children would participate in this study. The participating children were informed that they could stop their participation whenever they wished to do so. Bedouin participants responded to questionnaires that were back translated into Arabic.
Results
The differences between the two investigated groups in their victimization, bullying, and personality attributes were examined by MANOVA (multivariate analysis of variance) analyses. Two MANOVAs compared the peer violence subscales (Table 1). The multivariate effect for the bullying subscales was significant (Hotelling T = .20, F = 6.53,
MANOVA Comparing Reported Victimization and Bullying Among Jewish (n = 112) and Bedouin (n = 55) Students.
Note. MANOVA = multivariate analysis of variance.
p < .1. *p < .05. **p < .01. ***p < .001.
Bedouin students scored higher on moderate, serious, social, and cyber bullying. No significant difference was found for verbal bullying. The multivariate effect of victimization was also significant (Hotelling T = .24, F = 7.58,
Table 2 showed that in line with Hypothesis 2 the Jewish sample scored higher than the Bedouin sample on three out of four attributes—SVR, self-esteem (marginally), and sense of autonomy—and scored lower than the Bedouin sample on inhibition of emotions. Bedouin students scored higher on total bullying. A similar difference between the two groups indicated that Jewish adolescents claimed to be more verbally victimized than Bedouin students. No significant differences were found between boys and girls in their personality attributes or in their reported peer bullying and victimization.
MANOVA Comparing Violence and Personality Scores of Jewish (n = 98) and Bedouin (n = 49) Pupils.
Note. MANOVA = multivariate analysis of variance; SVR = strength to vulnerability ratio; BSI = Brief Symptom Inventory.
p < .1. *p < .05. ***p < .001.
A Pearson correlation matrix examined the associations between the investigated variables (Table 3). These data showed that in most cases SVR, self-esteem, sense of autonomy, and emotion regulation positively correlated with each other. However, while victimization and bullying of both Jews and Bedouin negatively and significantly correlated with SVR, a different pattern of correlations was found for the two ethnic groups. Victimization and bullying of Bedouin pupils were significantly and negatively correlated with self-esteem, sense of autonomy, and the positive component of SVR (My Life Today), whereas these indicators of peer violence were not significantly correlated with these personality attributes of Jewish pupils. The negative component of SVR (distress symptoms) was concurrently positively correlated with victimization and bullying of the Jewish sample, but was not significantly correlated with these indices of violence of the Bedouin sample. These data partly supported Hypothesis 3.
Pearson Correlation Matrix of the Investigated Variables.
Note. SVR = strength to vulnerability ratio.
*p < .05. **p < .01. ***p < .001.
An additional research question investigated the relative impact of the personality attributes as predictors of bullying and victimization among Bedouin and Jewish students. Four regression analyses (Table 4) indicated that SVR significantly predicted victimization as well as bullying of Jewish and Bedouin students, while pupils’ self-esteem, sense of autonomy, or emotional regulation did not have additional significant contributions to these predictions beyond the effects of SVR. These results supported our claim that individual resilience constitutes a major predictor of violent behavior in school, whose contributions to predicting bullying and victimization overpower the effects of the rest of the predictors.
Regression Analyses for the Effects of SVR, Self-Esteem, Sense of Autonomy, and Emotional Regulation on Victimization and Bullying of Jewish and Bedouin Pupils.
Note. SVR = strength to vulnerability ratio.
Discussion
This study examined determinant of peer violence in school among Jewish and Bedouin pupils. Bedouin pupils in the present study reported a higher level of bullying compared to Jewish pupils and in some respects also a higher level of victimization. Nisbett and Cohen (1996) as well as Elbedour et al. (2013) described the Bedouin culture as a “culture of honor” in which defending personal honor was an obligatory social value. Individuals in these cultures were expected to respond aggressively to any expression of insolence in order to preserve their reputation, and gain the respect of others. Bedouin adolescents could be expected, therefore, to exhibit a higher level of peer violence compared with Jewish students, as was the case in the present study.
A large number of studies of correlates of peer violence found that relational aggression was positively correlated with depression, loneliness, social anxiety, externalizing symptoms, and peer rejection during middle childhood (Craig, 1998; Paquette & Underwood, 1999; Prinstein, Boergers, & Vernberg, 2001), and negatively correlated with high theory of mind (van Dijk et al., 2017), as well as higher level of emotional regulation (Garofalo, Holden, Zeigler-Hill, & Velotti, 2016). Other studies reported that self-esteem (Lösel & Farrington, 2012; Teng et al., 2015) and support of pupils’ autonomy (Kaplan & Assor, 2012) negatively affected peer violence. Eisenberg, Liew, and Pidada (2004) found further that inept emotional regulation was associated with breaking social norms, while Cassidy and Taylor (2005) reported that early development of inappropriate emotional regulation predicted increased aggressive behaviors in the future. Another study established the association of stress with anger, fighting and bullying behaviors among schoolchildren (Centeio, Whalen, Kulik, Thomas, & McCaughtry, 2015). The present study indicated that the generality of these findings might be culturally bounded, as the correlations of bullying and victimization with personality attributes differed in the two investigated ethnic groups.
A number of recent Israeli studies indicated that individual resilience constituted a protective factor that helped confining detrimental reactions to various stressful conditions (e.g., Eshel & Kimhi, 2016). Yasien-Esmael et al. (2017) demonstrated as well that in the context of bereavement, SVR was a major predictor of coping of Israeli Arab Muslim widows with the loss of their husbands. Furthermore, the role of SVR as a mediator of the effects of other individual attributes on outcome variables was established by previous studies of coping with adversities (Eshel et al., 2016; Eshel, Majdoob, & Goroshit, 2014; Nadim, Pearlman-Avnion, Zysberg, & Eshel, Submitted). The present study revealed the major role of individual resilience in reducing peer violence. We are not aware of previous studies that examined the role of individual resilience in diminishing peer violence among school children.
One explanation for peer violence, derived from dominance theory, claimed that bullying should be regarded as an attempt to gain social dominance at the group and individual levels (Long & Pellegrini, 2003). According to this perspective, the aim of bullying was to maintain domination and to gain power and status to establish an individual-based social hierarchy. Engagement in such aggression would probably involve a conflict between proper peer interaction and the temptation of peer domination. No wonder that students with higher SVR, that is, with psychological strength that surpassed their vulnerability, were involved to a lesser degree in violence than peers with a lower SVR.
We believe that the study of peer violence should identify individual characteristics which would predict reduced violence, after controlling for the effects of other potential predictors. The multiple regressions conducted in this study to predict bullying and victimization revealed that self-esteem, sense of autonomy, and emotional regulation did not make any significant contribution to the explained variance of these indices of violence, beyond the contribution of SVR. These data indicated that resilience defined by both individual strength and frailty constitutes a cardinal predictor of lower relational violence.
It was observed that Israeli Jewish pupils who grew up in an individualistic cultural setting tended to endorse values of autonomy and individual coping with adversities, whereas Bedouins children who grew up in a traditional collectivistic society were characterized by lower levels of individuation and greater dependence on their families (Dwairy, 2004). Cultural norms were likely to account for the present findings of higher self-esteem, sense of autonomy, and SVR of Jewish pupils compared with Bedouin counterparts. However, our data showed clearly that individual characteristics played an important role in determining peer violence of Bedouin school children as well as of Jewish age mates.
It should be noted that studies of aggressive behavior and social hierarchy in school did not reach a consensus concerning the relative impact of personal attributes as compared to school culture and values. Klein, Cornell, and Konold (2012) claimed that school climate explains 66% of the variance of student aggression, whereas other researchers in this field contended that psychological factors were major determinants of peer violence (Aronson et al., 2016; Berger & Caravita, 2016).
Direct methods of restraining student violence may constitute necessary educational measures employed by teachers and schools. Yet the current levels of student violence may point at the limited impact of these measures on school bullying and victimization. The present data suggested that additional indirect methods based on supporting and enhancing individual attributes such as SVR may contribute further to decreasing this violence. More extensive research on these issues should be conducted to help the educational system cope with this serious problem.
Limitations
Although the present results were encouraging, additional research should be devised to reaffirm them. Much larger samples would be required to corroborate the interactions among cultural factors, elements of school culture and student characteristics in the Jewish and the Bedouin school systems, and to establish the role of individual resilience as a major protective psychological factor in decreasing peer violence in school.
Conclusion
Previous research examined the role of protective and risk personality factors in determining peer violence in school. Donnellan, Trzesniewski, Robins, Moffıtt, and Caspi (2005) identified higher self-esteem as a violence-reducing feature, and Lösel and Bliesener (1994) found that higher self-efficacy was associated with lower peer violence. Temperament characteristics (Donnellan et al., 2005) and attention deficit hyperactivity disorder (ADHD; Bernat et al., 2012) were identified, on the contrary, as violence-enhancing factors. The present study suggested two reservations for these conclusions. First, the correlations of personality characteristics with peer violence might differ in different cultural settings. Our data showed that victimization and bullying of Bedouin pupils were significantly and negatively correlated with self-esteem and sense of autonomy, whereas these indicators of violence were not significantly correlated with these personality attributes of Jewish pupils. Second, predictors of peer violence should be studied by multiple methods such as regression analyses, to examine the possibility of hierarchy among them. Our data suggested that measures of individual resilience such as SVR may have a greater role in predicting adolescents’ violence in school than other personality attributes. Furthermore, it should be remembered that diversity is a complex phenomenon which exists concurrently between different cultural groups as well as within each group.
Footnotes
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, authorship, and/or publication of this article.
