Classes of bullying behaviour in Chilean preadolescents
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Bullying is defined as persistent, hostile actions intended to harm the recipient in a power imbalance relationship. While children and adolescents have been identified as either victims, bullies or bully-victims in past research, few studies have identified patterns of more specific forms of bullying and victimisation behaviour. In addition, questions about stability and changes in patterns of bullying behaviours still need to be answered. This study aimed to characterise classes of bullying and victimisation and estimate the latent transitions of bullying behaviour and its associations with psychosocial and emotional dimensions in students from schools serving socially vulnerable populations. The study utilized secondary self–report data from Chilean students (n = 5151) aged 9–13 years at baseline. The Olweus Bully-Victim Questionnaire was administered to identify victims and bullies. We included items related to diverse types of bullying behaviour, such as teasing, exclusion, fighting and rumours, among others. We used latent class analysis to identify students’ classes in 2016 and latent transition analysis to explore change or stability between 2016 and 2017. We used the Strengths and Difficulties Questionnaire (SDQ) self-reported student version to estimate the association of the classes with individual predictive factors, such as emotional symptoms and problematic behaviours. Overall, 54% of the 2016 sample was included in a class associated with bullying behaviour, either as a victim, a bully, or a bully-victim. The prevalences of the classes were non-involved (46%), teasing-victim (25%), teasing-bully (13%), only victim (10%) and bully-victim (6%).Regarding predictors, the emotional symptoms were associated with higher odds of belonging to the bully-victim, victim, and teasing-victim classes, in contrast to non-involved classes. In comparison, conduct problems were associated with higher odds of belonging to any class other than the non-involved class. The simultaneous modelling of data from 2016 and 2017 (n = 3,790) allowed us to identify four invariant statuses across the two time points: bully-victim, verbal psychological victim, teasing-bully, and non-involved. The students in the four statuses had a greater probability of remaining in the same status: the most stable status was the non-involved (.804), and the least stable was the bully-victim (.333), while the verbal psychological victim (.569), the teasing-bully (.450) had an almost equal probability of remaining and transitioning. Emotional and behavioural indicators significantly predicted the transition between statuses. Sex as a predictor was related to status membership. For example, students being female in 2016 increased the odds of transitioning to non-involved and verbal psychological status as opposed to remaining in teasing-bully status in 2017. The intervention variable and age were not predictive of status changes.Identifying heterogeneous classes of bullying and victimisation and specific correlates may allow for tailoring intervention to prevent specific forms of bullying behaviour. Measuring the processes of change and stability will allow us to understand bullying from a dynamic perspective and why students do not remain in the same role.
Description
Tesis (Doctor in Epidemiology.)--Pontificia Universidad Católica de Chile, 2025
Keywords
Bullying, Stability of bullying, Latent class analysis, Latent transition analysis