ارتباط زورگویی سنتی و ارتکاب مزاحمت سایبری در میان دانش آموزان استرالیا

Abstract This study investigated the associations of gender, age, trait anger, moral disengagement, witnessing of interparental conflict, school connectedness and the religious makeup of the school setting in the involvement in traditional bullying and cyberbullying perpetration. Five hundred Australian students completed an anonymous self-report, paper-based questionnaire. According to the results, 25.2% of the participants reported having engaged in traditional or cyberbullying perpetration. While trait anger and moral disengagement were associated with being a traditional bully, trait anger, interparental conflicts, moral disengagement and school connectedness were associated with being a traditional bully-victim. Additionally, trait anger and moral disengagement were associated with being a traditional-and-cyberbully. Our findings indicated that besides individual variables, the family and school environment have an impact on traditional and cyberbullying perpetration behavior. Results imply that any prevention attempts to reduce traditional and cyberbullying should consider students' experiences both at home and at school.

1. Introduction Bullying is a social relationship problem which can be defined as an imbalance of power characterized by an intention to hurt others which is repeated (Olweus, 1993). With the advent of technology such as the Internet and mobile phones widely available to young people, cyberbullying, or bullying using technology has emerged (Campbell, 2005). Although there has been some controversy over whether the three criteria of an imbalance of power, intentionality and repetition of traditional bullying apply to cyberbullying (Dooley et al., 2009 and Slonje et al., 2013), many researchers are in agreement that they are applicable, although with some differences in appearance depending on the different mediums (Menesini et al., 2013 and Ybarra et al., 2012). Hence, cyberbullying can be defined as aggressive, deliberate and repeated behaviors of an individual or group of individuals by using information and communication technologies to inflict harm on others (Smith et al., 2008). The prevalence rates for traditional and cyberbullying in Australia seem to be similar to other developed countries with about 20–30% of students being traditionally victimized, 15% being cyberbullied and 7–8% being bullied in both modes (Campbell et al., 2011 and Hemphill et al., 2012). These prevalence rates are despite Australia's adoption of a National Safe Schools Framework (Cross, Epstein, & Hearn, 2011) where every school is required to develop an anti-bullying policy and evidenced-based programs to reduce bullying are available (Cross et al., 2012). Victims of both traditional bullying and cyberbullying suffer many negative consequences according to the existing research evidence. Victims have reported experiencing psychological, social, physical and school related problems. Anxiety, depressive symptoms, anger, sadness, guilt, shame and frustration have been reported among the psychological problems (Chin, 2011, Mishna et al., 2010 and Wang et al., 2011). Negative social impacts have been shown to be withdrawal from friends, loneliness and peer rejection (Hinduja and Patchin, 2007, Kroon, 2011 and Ybarra and Mitchell, 2004). Physically, victims have been shown to self-harm, sustain physical injuries and abuse drugs (Hinduja and Patchin, 2007, Shariff, 2008, Wang et al., 2010 and Ybarra and Mitchell, 2004). Additionally, victims have attendance problems and low grades at school (Cross et al., 2015 and Johnson, 2011). Students who bully others in physical or cyber space appear to be at risk as well. Compared to the victims, traditional bullies were reported having low levels of school attendance, school satisfaction and higher levels of irritability (Arslan, Hallett, Akkas, & Akkas, 2012); and in comparison with the non-involved students, cyberbullies were found to have social problems, higher levels of stress, depression and anxiety (Campbell, Slee, Spears, Butler, & Kift, 2013). When compared with the non-involved students, perpetrators of traditional bullying and cyberbullying were also at highest risk in terms of substance usage and weapon carrying (Wang, Iannotti, & Luk, 2012). Therefore, understanding the risk factors for students who bully their peers is important to inform prevention and intervention programs to reduce all forms of bullying, including cyberbullying.

3. Results 3.1. Data analysis The frequencies obtained from Traditional Bullying and Cyberbullying Questionnaire were used to identify the perpetrators and victims of traditional bullying and cyberbullying. The categories were created on the basis that the frequency of the reported perpetration was once or more. Please note that only-victim status participants who reported having been victimized once or more in traditional or cyber ways but never bullied others were disregarded since perpetrators were the specific focus of the current research. All participants were categorized first descriptively by their bully status; whether it was traditional bullying or cyberbullying and whether the participant was a bully only or a bully-victim. This resulted in 6 categories of pure traditional bully (traditionally bullied others once or more but was never victimized), pure cyberbully (cyber bullied others once or more but was never cyber victimized), combined traditional and cyber bully (bullied others in traditional and cyber environments once or more but was never victimized), traditional ‘bully-victim’ (not only bullied others but was also victimized in traditional ways), cyber ‘bully-victim’ (not only bullied others but was also victimized in cyber settings) and traditional and cyber ‘bully-victim’(not only bullied others but was also victimized in physical and cyber settings). Additional combinations were also identified such as a traditional bully who was also a cyber ‘bully-victim’ resulting in 12 categories as can be seen in Table 1. Table 1. Descriptive statistics for traditional bullying/cyberbullying perpetrators. Perpetrator types Whole sample f(%) Gender f(%) Grade f(%) School setting f(%) Girls Boys Primary High Islamic Others Trad. bully-only 26(5.2) 14(4.8) 11(5.8) 4(4.9) 22(5.3) 14(7.9) 12(3.7) Trad. bully-victim 21(4.2) 14(4.8) 7(3.7) 8(9.9) 13(3.1) 13(7.3) 8(2.5) Cyber bully-only 7(1.4) 4(1.4) 3(1.6) – 7(1.7) 3(1.7) 4(1.2) Cyber bully-victim 2(0.4) 1(0.3) – 1(1.2) 1(0.2) 1(0.6) 1(0.3) Trad. and cyber bully 8(1.6) – 8(4.2) – 8(1.9) 3(1.7) 5(1.6) Trad. bully-victim + Cyber bully-victim 21(4.2) 8(2.7) 11(5.8) 1(1.2) 19(4.6) 7(3.9) 14(4.3) Trad. and cyber bully + Cyber victim 6(1.2) 3(1.0) 3(1.6) – 6(1.5) 1(0.6) 5(1.6) Trad. and cyber bully + Trad. victim 8(1.6) 4(1.4) 3(1.6) – 8(1.9) 1(0.6) 7(2.2) Trad. bully + Trad. and cyber victim 16(3.2) 11(3.8) 5(2.6) 1(1.2) 15(3.6) 3(1.7) 13(4.0) Trad. bully + Cyber victim 4(0.8) 3(1.0) 1(0.5) 2(2.5) 2(0.5) 2(1.1.) 2(0.6) Cyber bully + Trad. victim 4(0.8) 2(0.7) 2(1.0) – 4(1.0) – 4(1.2) Cyber bully + Trad. victim + Cyber victim 3(0.6) 2(0.7) 1(0.5) – 3(0.7) – 3(0.9) Total 126(25.2) 65(22.6) 55(28.9) 17(20.9) 108(26.0) 49(27.5) 78(24.2) Notes. Trad. = traditional. Ns vary (N = 500 for the whole sample; N = 483 for the gender; N = 5494 for the grade; and N = 500 for the school setting). Table options However, as there were such low frequencies in seven of the 12 categories in Table 1, the categories were collapsed. The new categories were created with the first five perpetrator groups in Table 1. These groups were traditional bully, traditional ‘bully-victim’, cyberbully, cyber ‘bully-victim’ and traditional and cyber bully (Table 2). The participants in the other groups in Table 1 were combined under these five perpetrator groups on the criterion that they were involved in the relevant perpetration category. For example, if a participant reported having traditionally bullied someone but was also cyber victimized once or more, this participant was combined with the traditional bully category (see Trad. bully category in Table 2). Or if a participant reported having bullied others in traditional and cyber environments but was also cyber victimized once or more, this participant was combined with the traditional and cyber bully category (see Trad. and cyber ‘bully’ category in Table 2). The rationale behind these combinations was dependent on the research indicating that there is an overlap between traditional bullying and cyberbullying (e.g., Erdur-Baker, 2010 and Wachs, 2012). This overlap implies that traditional bullies also use online environments to bully others. Table 2. Combinations of the five perpetrator categories. Category Combination n Trad. bully Trad. bully only group and trad. bully + cyber victim group 30 Trad. ‘bully-victim’ Trad. bully-victim group, traditional bully-victim group + cyberbully-victim group and trad. victim + trad. bully + cyber victim group 58 Cyberbully Cyberbully only group, trad. victim + cyberbully group and trad. victim + cyber victim + cyberbully group 14 Cyber ‘bully-victim’ Cyber bully-victim group 2 Trad. and cyber ‘bully’ Trad. bully + cyberbully group, trad. bully + cyber victim + cyberbully group and trad. victim + trad. bully + cyberbully group 22 Notes. Trad. = traditional. Table options Independent sample t-tests were conducted to examine the influence of gender, age and the religious makeup of the school setting on trait anger, interparental conflicts, moral disengagement and school connectedness. A multinomial logistic regression analysis was conducted to explore the correlates of traditional bullying and cyberbullying perpetration. 3.2. Descriptive statistics The frequencies and percentages of perpetrator types by gender, grade and the religious makeup of the school setting are shown in Table 1. It was found that 17% of perpetrators fell into six bullying categories (traditional bully only; cyberbully only; both a traditional and cyber bully; traditional ‘bully-victim’; cyber ‘bully-victim’; and traditional and cyber ‘bully-victim’). There were another six low frequency categories which comprised 8.2% of bullies where the combination of bullying others was combined with other roles such as a traditional bully and cyber victim. While 28.9% of the boys surveyed reported bullying someone else, 22.6% of girls reported this bullying behavior. More high school students (26.0%) reported bullying someone with 20.9% of primary students saying that they had been a perpetrator since the last January. While more than a quarter of the students attending to an Islamic school (27.5%) engaged in bullying perpetration, a little less than a quarter of the students not attending a Muslim school (24.2%) reporting having bullied others. 3.3. Gender, grade and religious makeup of the school differences regarding the four associated variables Two independent sample t-tests were conducted to examine whether there were significant gender and age differences in terms of the associated variables, specifically trait anger, interparental conflict, moral disengagement and school connectedness. As seen in Table 3, boys and girls significantly differed in reported interparental conflict t(473) = − 2.79, p < .01 and moral disengagement t(481) = − 1.92, p < .05. Primary and high school students significantly differed in trait anger t(491) = 2.44, p < .05 and interparental conflict t(484) = − 2.53, p < .05. A further independent sample t-test was conducted to ascertain if there were any differences between the students who attended the Islamic schools and those who did not. It was found there were no significant differences between students at the Islamic school and other students in terms of trait anger t(497) = − .22, p > .05 and moral disengagement t(498) = − .30, p > .05, but there were significant differences in reported interparental conflict t(490) = − 1.97, p < .05 and school attachment t(496) = 2,42, p < .05. Table 3. T-test results comparing associated variables by gender, grade and school setting (for the five perpetrator categories). Gender Grade School setting Girls M(SD) Boys M(SD) t Primary Sc. M(SD) High Sc. M(SD) t Islamic Sc. M(SD) Other Sc. M(SD) t Trait anger 24.83(5.78) 23.21(5.87) 1.53 21.66(6.04) 23.58(5.61) 2.44⁎ 21.78(5.45) 21.67(4.99) − .22 Interparental conflict 12.20(3.86) 14.05(3.38) − 2.79⁎⁎ 13.35(3.59) 13.53(3.65) − 2.53⁎ 14.01(3.47) 13.38(3.41) − 1.97⁎ Moral disengagement 36.97(7.60) 39.86(8.99) − 1.92⁎ 40.06(8.93) 38.31(8.37) 0.77 35.25(7.70) 35.03(7.89) − .30 School connectedness 17.43(5.29) 16.61(5.51) 0.84 17.53(6.21) 17.09(5.28) 0.30 18.36(4.80) 19.43(4.69) 2.42⁎ Notes. M = mean. SD = standard deviation. Sc. = school. The lowest and the highest moral disengagement scores ranged from 18 to 72. The lowest and the highest interparental conflict scores ranged from 6 to 18. The lowest and the highest school connectedness scores ranged from 5 to 25. The lowest and the highest trait anger scores ranged from 12 to 36. ⁎ p < 0.05. ⁎⁎ p < 0.01. Table options 3.4. Correlates of traditional bullying and cyberbullying perpetration Prior to performing the multinomial logistic regression analysis, multicollinearity among the independent variables was inspected. No independent variables were strongly interrelated since estimates were within the limits between 0.09 and 0.30. Thus, all independent variables were added to the analysis. A multinomial logistic regression analysis was conducted to explore the correlates of traditional bullying and cyberbullying perpetration using SPSS software (version 21 for Windows). Due to the exploratory nature of the tested model (Field, 2009), stepwise forward entry method was used to estimate the contribution of each variable to the model. The dependent variable was being involved as a perpetrator of traditional bullying and cyberbullying. As the cyberbully-victim category in Table 2 was too small, it was discarded from further analyses. Thus, the dependent variable was composed of the four identified categories of traditional bully, traditional ‘bully-victim’, cyberbully, and traditional bully-cyberbully. Since some significant differences were found for gender, grade and the religious makeup of the school setting (Table 3), these variables were added as independent variables in the model. Therefore, gender, grade, trait anger, moral disengagement, interparental conflict, school connectedness and the religious makeup of the school setting were the independent variables. All independent variables were simultaneously included in the analysis. Non-perpetrators who were victims or not-involved in a bullying/cyberbullying incident were specified as the reference group to examine how traditional bullying/cyberbullying perpetrators differed from the non-perpetrators. Table 4 presents the results of the multinomial logistic regression analysis. Compared to the non-perpetrator group, trait anger (b = 0.13, Wald x2(1) = 9.78, p < .01) and moral disengagement (b = 0.06, Wald x2(1) = 4.05, p < .05) were significantly correlated with being in the traditional bully group. Odds ratio values showed that if trait anger and moral disengagement increase one more unit, the changes of the odds of belonging to traditional bully group are 1.14 and 1.06, respectively. In short, as trait anger and moral disengagement increase, participants are more likely to be in the traditional bully group than the non-perpetrator group. The other variables were not statistically significant. Table 4. Multinomial logistic regression analysis. Trad. bully Trad. bully-victim Cyberbully Trad. and cyberbully B (SE) Wald OR [95% CI] B (SE) Wald OR [95% CI] B (SE) Wald OR [95% CI] B (SE) Wald OR [95% CI] Gender Girls − 0.12 (0.44) 0.07 0.89 [0.37, 2.10] − 0.04 (0.34) 0.02 0.96 [0.49, 1.88] 0.17 (0.60) 0.08 1.19 [0.37, 3.86] − 1.08 (0.54)⁎ 4.04 0.34 [0.12, 0.97] Boys (ref.) Grade Primary S. − 0.48 (0.62) 0.62 0.62 [0.18, 2.06] 0.07 (0.44) 0.02 1.07 [0.45, 2.56] − 20.09 (0.00) 0.00 1.89 [1.89, 1.89] − 19.91 (0.00) 0.00 2.25 [2.25, 2.25] High S. (ref.) School setting Islamic 0.66 (0.43) 2.33 1.93 [0.83, 4.49] 0.21 (0.34) 0.38 1.23 [0.63, 2.41] − 0.58 (0.68) 0.73 0.56 [0.15, 2.13] − 0.08 (0.56) 0.02 0.92 [0.31, 2.75] Others (ref.) Trait anger 0.13 (0.04)⁎⁎ 9.78 1.14 [1.05, 1.23] 0.09 (0.03)⁎⁎ 8.11 1.09 [1.03, 1.16] 0.05 (0.06) 0.68 1.05 [0.94, 1.17] 0.10 (0.05)⁎ 4.08 1.11 [1.00, 1.22] Interparental conflict 0.10 (0.07) 2.35 1.11 [0.97, 1.26] − 0.09 (0.05)⁎ 3.89 0.91 [0.83, 1.00] 0.18 (0.10) 3.25 1.20 [0.98, 1.45] − 0.08 (0.07) 1.43 0.92 [0.80, 1.05] Moral disengagement 0.06 (0.03)⁎ 4.05 1.06 [1.00, 1.12] 0.05 (0.02)⁎ 6.16 1.05 [1.01, 1.10] 0.06 (0.04) 2.40 1.06 [0.98, 1.15] 0.07 (0.03)⁎ 4.10 1.07 [1.00, 1.14] School connectedness − 0.03 (0.04) 0.56 0.97 [0.89, 1.05] − 0.13 (0.03)⁎⁎⁎ 18.43 0.88 [0.83, 0.93] − 0.04 (0.06) 0.52 0.96 [0.85, 1.08] − 0.05 (0.05) 1.15 0.95 [0.86, 1.04] Notes: Reference group was the non-perpetrators. Trad. = traditional. B = regression weight. SE = standard error. OR = odds ratio. CI = confidence interval. R2 = .21(Cox & Snell), .25 (Nagelkerke). Model x2(28) = 107.85, p < .001. ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. Table options In comparing the traditional bully-victim group with the non-perpetrator group, the relative risk of being in the traditional bully-victim group was significantly related to trait anger (b = 0.09, Wald x2(1) = 8.11, p < .01), interparental conflict (b = − 0.09, Wald x2(1) = 3.89, p < .05), moral disengagement (b = 0.05, Wald x2(1) = 6.16, p < .05) and school connectedness (b = − 0.13, Wald x2(1) = 18.43, p < .001). Odds ratio values indicated that when trait anger, interparental conflict, moral disengagement and school connectedness increase one more unit, the changes of the odds of belonging to traditional bully-victim group are 1.09, 0.91, 1.05, and 0.88, respectively. In other words, as trait anger and moral disengagement increase, participants are more likely to be in the traditional bully-victim group than the non-perpetrator group. However, if interparental conflict and school connectedness increase, participants are less likely to be in the traditional bully-victim group. Gender, grade and the religious makeup of the school setting were not statistically significant for the traditional bully-victim group. None of the variables were associated with being in the cyberbully group over the non-perpetrator group. Gender (b = − 1.08, Wald x2(1) = 4.04, p < .05), trait anger (b = 0.10, Wald x2(1) = 4.08, p < .05) and moral disengagement (b = 0.07, Wald x2(1) = 4.10, p < .05) were significantly associated with a participant's being in the traditional-and-cyber bully group than in the non-perpetrator group. According to the odds ratio values, as gender changes from girl to boy, the change in the odds of belonging to the traditional-and-cyber bully group is 0.34. In other words, compared to the non-perpetrator group, the odds of a boy to be in the traditional-and-cyber bully group are 1/0.34 = 2.94 times more than a girl. In addition, the odds ratio values indicated that when trait anger and moral disengagement increase one more unit, the changes of the odds of belonging to traditional-and-cyber bully group are 1.11 and 1.07, respectively. In other words, as trait anger and moral disengagement increase, participants are more likely to be in the traditional-and-cyber bully group than the non-perpetrator group. No other variables were statistically significant.