Individual, Family, School, and Neighborhood Characteristics Associated with Bullying Perpetration: A Multilevel Analysis of Middle School Students in Taiwan

Background: School bullying is a common type of aggression among adolescents. Numerous studies in Western countries have examined its risk factors in multiple contexts including individual, family, and school, yet relatively little research has been conducted in Taiwan. Even less is known about the potential influence of neighborhood characteristics (e.g., poverty and safety issues). The present study examines whether the risk factors identified in Western literature are applicable to a Taiwanese context. It further extends the current literature by exploring the potential neighborhood effects on bullying among Taiwanese middle school students.

Methods: A sample of 1,176 seventh to ninth-grade students was randomly selected from 43 middle schools in New Taipei City, Taiwan. Written consents were obtained and self-reported questionnaires were group-administered in classrooms. Dependent variables included direct and indirect bullying, and independent variables consisted of risk factors at the individual- (e.g., delinquency), family-(e.g., lack of parental monitoring), school-(e.g., teacher maltreatment), and neighborhood-level (e.g., witnessing violence and poverty rate). Descriptive and basic inferential statistics were calculated. Given the nested nature of the data, hierarchical linear modeling (HLM) was used, and neighborhood poverty rate was specified as a level-2 predictor. A series of models including the null model (with no predictors), the fixed intercepts model (with level 1 predictors), and the intercepts-and-slopes-as-outcome model (with level 1 and 2 predictors) were constructed and tested stepwise. Separate analyses were conducted to examine the effects of the above factors on direct and indirect bullying respectively.

 Results: The results showed significant age difference, with 7^th and 8^th graders more likely to bully than 9^thgraders. Compared to girls, boys were more likely to engage in bullying. Delinquency (b=.29, p<.05), lack of parental monitoring (b=.03, p<.05), teacher maltreatment (b=.26, p<.05), and witnessing violence in neighborhood (b=.08, p<.05) were all significantly associated with direct bullying. In addition, direct bullying was lower in areas with higher poverty rates (level 2 factor: b=-.11, p<.05). Similar results were found with regard to indirect bullying, although the effect of parental monitoring became non-significant.

 Implications: The present findings were consistent with past research conducted in Western countries and confirmed the significant effects of individual, family, and school factors on Taiwanese adolescents’ direct and indirect bullying. In addition, neighborhood characteristics, such as community violence and poverty rates were found to be significantly associated with bullying perpetration. The importance of neighborhood context has been long suggested by criminologists and sociologists who have examined the relevance of neighborhood environment on youths’ adjustment and development. This study further highlights the impact of adverse neighborhood conditions on students’ peer relationships and interactions in school. To develop and implement effective youth violence prevention programs and policies, it is imperative that educators, school administrators, and practitioners in Taiwan consider not only school, but also other relevant social environments surrounding youth, such as home and neighborhood.