دخالت قومیت و زورگویی در یک نمونه ملی جوانان بریتانیا

Abstract This study investigated ethnic differences in bullying involvement (as victim and bully) among a UK wide sample of adolescents, controlling for potential confounders, including age, gender, economic situation, family structure and parent–adolescent relationships. 4668 youths, aged 10 to 15, who participate in the UK Household Longitudinal Study were assessed for bullying involvement. Binary logistic regression models were used to estimate ethnic differences across bullying roles while controlling for potential confounders. Overall, ethnic minority youths were not more likely to be victims; African boys and girls were significantly less likely to be victimised than same sex White youths. Pakistani and Caribbean girls were significantly more likely to have bullied others compared to White girls. Further research is necessary to explore why Pakistani and Caribbean girls may be more often perpetrators of bullying than girls in other ethnic groups.

Introduction Bullying is characterized by aggressive behaviour, engaged in repeatedly, by an individual or group of peers with more, actual or perceived, power than the victim (Olweus, 1993). The aggressive behaviour may be overtly physical, verbal or relational (Nansel, 2001). Peer victimisation or bullying perpetration in childhood is associated with, and a precursor of, a range of psychosomatic (Gini & Pozzoli, 2009) and mental health problems (Arseneault, Bowes, & Shakoor, 2010) including suicide ideations and behaviour (Fisher et al., 2012; Winsper, Lereya, Zanarini, & Wolke, 2012). Decreased school performance (Woods & Wolke, 2004) or involvement in crime (Ttofi, Farrington, Losel, & Loeber, 2011) have also been reported as consequences of bullying. To reduce and limit the negative impact of bullying, research must identify the factors most strongly associated with youths being bullied or engaging in bullying behaviours. Demographic characteristics such as age and gender have emerged as significant risk factors; among adolescents, bullying victimisation steadily declines with age, while bullying perpetration slightly increases (Smith, Madsen, & Moody, 1999). Furthermore, boys are more often victims and perpetrators of bullying than girls (Nansel, 2001). Ethnicity is another key demographic factor that may contribute to exposure to peer victimisation; however, there is continuing debate over whether rates of bullying differ between ethnic groups. While there has been substantial discussion in both academic and policy literatures relating to the prevalence of racist bullying and stereotyping in schools (Abrams, 2010; Eslea & Mukhtar, 2000; House of Commons Education and Skills Committee, 2007), small sample studies in the UK which compared single or mixed ethnic minority groups to majority White children have found no difference in the prevalence of bullying among ethnic groups (Durkin et al., 2012; Eslea & Mukhtar, 2000; Moran, Smith, Thompson, & Whitney, 1993); although ethnic minority children are more likely to identify their race or culture as the reason for them being bullied, they appear no more likely than white majority children to be victimised or to bully others (Boulton, 1995; Monks, Ortega-Ruiz, & Rodríguez-Hidalgo, 2008). Outside of the UK, findings are more mixed. Several European studies comparing immigrant and native-born children find immigrant children are more likely to report being bullies or victims (Fandrem, Strohmeier, & Roland, 2009; von Grunigen, Perren, Nagele, & Alsaker, 2010; Verkuyten & Thijs, 2002), although this appears somewhat dependent on children's language competence (von Grunigen et al., 2010) or the ethnic mix of the school they attend (Verkuyten & Thijs, 2002). In contrast, other European studies find no difference in bullying between ethnic groups (Monks et al., 2008), or show native born children to be more often bullies than immigrant children (Strohmeier, Spiel, & Gradinger, 2008). More consistent findings have been reported in the United States, where large datasets have been used to compare rates of bullying among White, African American and Hispanic children. The results suggest African American children are less likely to be victimised than those from other ethnic groups (Hanish & Guerra, 2000; Sawyer, Bradshaw, & O'Brennan, 2008; Spriggs, Iannotti, Nansel, & Haynie, 2007), however children from ethnic minority groups appear more likely to participate in bullying others (Carlyle & Steinman, 2007; Nansel, 2001; Wang, Iannotti, & Nansel, 2009). The consistency of the findings notably differs between Europe and the US, and this may result from use of differing sampling methods; research in the US mostly employs large-scale representative surveys, whereas European studies tend to rely on smaller classroom or school-based convenience samples (Durkin et al., 2012). In addition, the US represents a different context, with higher levels of overall group segregation than in the UK (Johnston, Wilson, & Burgess, 2004). Bullies and victims are therefore more likely to come from the same ethnic group, rather than bullying crossing ethnic divides. Where differences have been observed between ethnic groups, these have mostly been explained as a result of differing parenting practices, such as parental communication (Spriggs et al., 2007), discipline (Lansford, Deater-Deckard, Dodge, Bates, & Pettit, 2004), and supervision (Peeples & Loeber, 1994). Bullies are more likely to report greater physical discipline, poorer family cohesion (Espelage, Bosworth, & Simon, 2000) and less secure caregiver attachment (Walden & Beran, 2010) than youths not involved in bullying, while victims more often experience maltreatment (Holt, Kaufman Kantor, & Finkelhor, 2008) and poorer or inconsistent supervision at home (Bowers, Smith, & Binney, 1994). Thus, differences in parenting behaviours across ethnic groups may partly account for the variations in bullying involvement by different ethnic groups. Similarly, economic factors merit consideration, as ethnic minorities tend to experience greater poverty and deprivation than ethnic majorities (Platt, 2007). Victims of bullying show greater deprivation at home (Wolke & Skew, 2012), and come from families of lower affluence (Due et al., 2009). Given the economic disparities between ethnic groups, and the relationship this may have with bullying, such factors should be controlled for when examining ethnic differences in bullying.

Results Prevalence of bullying and age/gender differences Among all youths, 11.3% (N = 536) were victims of bullying, 2.4% (N = 114) were bullies, and 0.9% (N = 43) bully-victims (incorporated into the victim group for subsequent analysis). Girls were less likely than boys to be victims (OR = 0.77, 95% CI = 0.64–0.92) or bullies (OR = 0.62, 95% CI = 0.43–0.91). Examining age differences for bullying, compared to the youngest group (aged 10), youths from all other age groups (from 11 to 15 years old) were significantly less likely to be victims of bullying. No significant differences were found in the likelihood of bullying others between age groups. Characteristics of the sample Table 2 describes differences between ethnic groups for each of the potential confounders. Youths from minority ethnic groups differed on all potential confounders but age. Minority youths, with the exception of those of Indian origin, were more likely to come from low-income households, to live below the poverty line, and to more often experience financial stress than White youths. Similarly, ethnic minorities showed greater levels of material deprivation than the ethnic majority, while also possessing fewer consumer items, and experiencing greater financial stress. Family structure differed between ethnic groups; Pakistani and Bangladeshi young people tended to have larger families than all other ethnic groups, while Indian, Pakistani, Bangladeshi and other Asian youths were more likely to live with both of their natural parents than White youths. Those from Caribbean and African families were the least likely to live with both natural parents. Differences were observed in parent–adolescent relationships: Indian and Pakistani youths reported more positive relationships with their parents than other ethnic groups, while White youths and those from the Other Asian category reported more negative relationships. Parent self-reports suggested that White and Caribbean parents had the highest levels of harsh parenting. Association between bullying and ethnicity Crude associations between bullying role and ethnic groups were first analysed (Model A), and then repeated controlling for potential confounding variables of age, gender, parental qualifications and economic situation (Model B) and family structure and parent–adolescent relationships (Model C) (Table 3). Associations were found between ethnicity and victim status across all models only for African youths (Model C: OR = 0.26, 95% CI = 0.13–0.52). Bangladeshi youths (OR = 0.44, 95% CI = 0.23–0.84) were at reduced risk of victimisation once confounding factors had been controlled for (Models B and C). Given their much greater poverty risks, they may have been expected to have higher absolute rates of victimisation; but as introducing the controls demonstrated, once poverty economic factors were controlled, they had lower rates than otherwise similar majority group children. Indian youths (OR = 0.53, 95% CI = 0.31–0.92) were found to be at reduced risk of victimisation at Model B but not at Model C, once family factors had been controlled for. Pairwise comparisons between all ethnic groups found only African and White youths differed significantly in their reports of victimisation (Bonferroni, p < 0.005). Fig. 2 shows gender differences in victimisation among ethnic groups. Significant differences compared to white youths of the same sex were observed for African boys (OR = 0.17, 95% CI = 0.05–0.54) and African girls (OR = 0.37, 95% CI = 0.16–0.89), who were both less likely to be victims of bullying than White youths. In contrast, the effect for less victimisation of Bangladeshi boys or girls was only found when both sexes were combined ( Table 3), and was no longer statistically significant when the analysis was broken down by sex ( Fig. 2). For those who bullied others, across all models both Pakistani (OR = 3.32, 95% CI = 1.65–6.68) and Caribbean (OR = 2.74, 95% CI = 1.25–6.02) youths showed a greater likelihood of bullying others (Table 3). Bangladeshi youths (OR = 2.89, 95% CI = 1.42–5.91) were associated with bullying in Model A, but this was no longer significant once confounding factors had been controlled for. Additional multiple comparisons using the Bonferroni method found no significant differences in bullying perpetration between ethnic groups. Gender differences between ethnic groups for bullying perpetrators are shown in Fig. 2. When broken down by sex, no significant differences in bullying perpetration were observed between ethnic minority and ethnic majority boys. However, Pakistani (OR = 6.56, 95% CI = 2.33–18.45) and Caribbean girls (OR = 4.82, 95% CI = 1.63–14.26) were significantly more likely to report that they had bullied others compared to white girls.