Griz, Carolina
(2021)
BULLYING IN SCHOOLS: A Measurement Validation Study in Brazilian Children and Longitudinal Prediction of Childhood Bullying Behaviour in the UK.
PhD thesis, University of Liverpool.
|
Text
201061022_Nov2021.pdf - Unspecified Download (5MB) | Preview |
Abstract
Bullying is defined as intentional, power imbalanced and repetitive use of aggressive behaviours. Research shows bullying is a global issue, where roughly two in every ten pupils are directly involved in bullying. Furthermore, bullying involvement poses a high risk for developing emotional and psychological problems as well as educational problems. Though bullying studies date back to the 1970s, higher prevalence rates have progressively been reported despite international intervention polices being introduced. Although many previous studies have investigated causes of antisocial behaviour more broadly, fewer have been designed to examine risk and protective factors for engaging in bullying behaviours in particular. Most of these studies have examined predictors of bullying involvement during adolescence with the lower age for samples typically being around 12 years of age. Comparatively few have investigated predictors of earlier bullying involvement and validated measures of bullying have seldom been used. The ability to assess bullying involvement reliably is essential for assessment of outcomes in high quality longitudinal research and it is a key foundation for the identification of children who may benefit from early intervention to prevent behaviours becoming entrenched. In Brazil, unfortunately there is both a lack of robust validated bullying measures, and prevention and intervention initiatives are still incipient. In this context, the present doctoral research aimed to: (i) evaluate the reliability and validity of two bullying measures in Brazil: the Bullying Prevalence Questionnaire (BPQ; Rigby & Slee, 1993) and the University of Illinois Bully Scale (UIBS; Espelage & Holt, 2001); (ii) systematically review the international literature available on childhood factors that contribute to later bullying behaviours; (iii) validate the Forms of Bullying Scale (FBS; Shaw, Dooley, Cross, Zubrick & Waters, 2013) in childhood (ages 9-10) in the UK; and (iv) use the FBS to examine the role of a range of early socio-demographic variables, maternal relationship circumstances, maternal mental health, child psychological and interpersonal functioning, and parenting environment and practices as possible predictors of bullying behaviours at 9-10 years of age in a representative UK birth cohort, using measures completed at the time of school entry, aged 4-5 years. The University of Illinois Bullying Scale and the Bullying Prevalence Questionnaire were translated into Portuguese and administered to a group of Brazilian adolescents alongside indices of psychopathology and empathy. Exploratory factor analysis replicated the original structure of the UIBS, and construct validity and convergent validity were partially supported. Less encouraging results were attained for the BPQ. These study findings are encouraging and suggest its suitability for use in Brazil, over the BPQ, however a further large-scale study is required to confirm the findings and support its future use in Brazil. The Forms of Bullying Scale (FBS; Shaw et al., 2013) was used in a UK sample of 640 children aged 9-10 years taking part in the Wirral Child Health and Development Study. The results of Exploratory Factor Analysis mirrored the original factor structure of the FBS in adolescents (aged 12-15), being both statistically as well as conceptually robust. Furthermore, concurrent validity results for the FBS were confirmed in relation to traditional bullying, whilst convergent validity was tentatively supported though associations were small. The measure can now be cautiously recommended for future use in this younger age group but convergent validity in this young age group needs future replication. Following validation of the FBS in the WCHADS sample at age 9, the predictive independent effects of early socio- demographic, maternal relationship circumstances, maternal mental health, child psychological and interpersonal functioning, and parenting environment and practices variables assessed at age 5 on later bullying perpetration at age 9 were examined using a hierarchical regression analysis. Male gender, lower family income, financial problems, higher maternal anxiety, lower parental involvement, and higher inconsistent discipline in early childhood significantly predicted later bullying behaviour. A high proportion (over 90%) of children reporting bullying others at age 9 also reported experiencing victimisation. These findings make a novel contribution to the relatively scarce literature on early childhood predictors of emerging bullying behaviour in middle childhood and, if replicated, may serve to inform the focus of early interventions.
| Item Type: | Thesis (PhD) |
|---|---|
| Uncontrolled Keywords: | school bullying; early prospective predictors; childhood; longitudinal studies; psychometric testing |
| Divisions: | Faculty of Health and Life Sciences |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 18 Nov 2021 13:26 |
| Last Modified: | 18 Jan 2023 21:24 |
| DOI: | 10.17638/03143225 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3143225 |
Dimensions
Dimensions
