Peer support approaches, where pupils offer formal support to others, are used in schools as an additional source of pastoral care. Previous research shows benefits for whole school environments, pupils who receive support, and pupil peer supporters, but is largely limited to Western case studies, which are often short-term and/or limited to qualitative data. This thesis addresses these issues through cross-national work on peer support use, and longitudinal case studies. Peer support use in Japan was investigated through a qualitative study. A range of approaches was seen and major themes identified, including the evolution of Japanese peer support, conflicts in approaches, and an emphasis upon community. A qualitative study in South Korea explored peer support and other anti-bullying initiatives, both proactive and reactive. Thematic analysis showed the importance of collectivist values, and a gap between policy and practice. Peer support was little used and, despite positive attitudes, barriers were perceived. A comparison of the UK, Japan and South Korea considered the impact of cultural values, education systems and the nature of bullying upon peer support use. Two mixed methodology case studies of peer listening schemes in UK secondary schools were conducted, one over 18 months and another over six months. Both schemes had the general aim of providing additional pupil support, without focus on particular outcomes. Impact upon the domains typically benefited by peer support was evaluated; findings were broadly similar. Pupil awareness of the service was high, but in-depth knowledge was weaker and use was low. Attitudes were mixed, impact upon bullying and school climate was limited, but users and peer supporters perceived benefits. Both schemes evolved to include multiple approaches, and practical factors affecting development and effectiveness were identified. Finally, implications for peer support practice are drawn from the findings as a whole, and future research directions suggested.

Building on a small body of research, the present study explores the perceptions of teaching staff about their work with Gypsy, Roma, Traveller (GRT) children and young people. Specifically the research is concerned with participants’ views about the relationship between GRT children and schools in relation to attainment, social inclusion, the GRT culture and lifestyle and wider systemic factors. Existing literature and research about the educational experiences of GRT children and young people is explored. Situated within a critical realist epistemology, the present study utilises semi-structured interviews with 13 members of teaching staff across five schools in Greenshire County Council (pseudonym). Transcripts were subjected to thematic analysis following the model proposed by Braun and Clarke (2006). Six superordinate themes (educational outcomes; barriers to education; GRT parents; social inclusion; cultural dissonance and inclusive practices) were identified. Findings are explored in relation to previous research. One key finding concerns the views participants expressed about GRT children’s reports of bullying or racism. Implications for practice and future research are considered.

Background. Chronic exposure to stressors in childhood has been linked with heightened risk of developing symptoms of psychosis in both clinical and non-clinical populations. The association has been explicated with reference to developmental alterations in biological and psychological systems. One such stressor, being bullied in childhood, has been the focus of recent investigations. The current study endeavoured to systematically review the available evidence from studies purporting to investigate the association between childhood bullying and psychosis symptomatology. Method. A search of PubMed, Medline, PsycInfo, Embase, Scopus and Web of Science electronic databases, alongside manual searching and cross-referencing, was carried out. The quality of available evidence for and against the association was assessed using quality assessment tools found in the literature. Results. Meeting the study’s inclusion criteria were 30 studies (eighteen cross-sectional, twelve cohort). Longitudinal studies, by design, provided higher quality evidence – particularly those which examined specifically the association between the variables of interest. However, preselected key confounding variables were not always taken into account, highlighting that the association is not unequivocal and that further research is warranted. Cross-sectional studies provide lower quality of evidence (of greater variability) where severe limitations regarding the validity and generalisability of findings must be taken into account. Conclusions. Evidence suggests that the association between experience of bullying in childhood and onset of psychotic or psychotic-like symptoms of clinical and non-clinical severity is tenable. However, future research needs to re-examine the association while minimising methodological limitations including confounding variables and definitional issues. Establishment of an association warrants investigation of the mechanisms which potentially underlie it; the pre-existing, small research base on mediators and moderators of the relationships requires further attention.

Sibling relationships have a great impact on children’s social and psychological development. This thesis provides an all-encompassing examination of the precursors and outcomes of sibling bullying through three quantitative studies: the first study, a meta-analysis, provides a foundational schema of the factors associated with sibling conflicts; the second study, a short-term longitudinal study, examines the individual and proximal precursors of sibling bullying and its short-term outcomes (one and two years later); the third study, a long-term longitudinal study, examines the distal precursors of sibling bullying and its long-term outcomes (five years later). The first study assessed the strongest effect sizes associated with sibling conflicts. It examined the link between parent-child relationships, familial factors and sibling conflicts. Studies were identified through a systematic search, coded, and selected based on criteria relevant for this study resulting in 60 studies (178 effect sizes), which in total involved 43,270 participating children and adolescents. Studies were categorised as proximal and distal factors. Those involved in sibling conflicts were significantly less likely to have authoritative, and warm and affectionate parents, and less likely to come from families with affluent socioeconomic-status, positive family climate and good marital quality. Conversely, more sibling conflicts were significantly related to abusive and neglectful parents, and parent-child conflicts; and more likely to come from families with poor mental health, low SES, adverse family atmosphere and parental conflict. The factors were moderated by assessment methods, study design, direction and form conflict, gender constellation, and continent. This study served as a building block for the two following studies, as it highlighted key factors to focus on in further assessing the precursors and outcomes of sibling bullying. The second study, which was based on the Edinburgh Study of Youth Transition and Crime (ESYTC, 2014) found that parenting factors were crucial to sibling bullying. Parental involvement, parent-child conflict and parent-child leisure time were precursors and outcomes of sibling bullying, so that more parental involvement and parent-child leisure time were associated with less sibling bullying perpetration and victimisation. Further, sibling bullying perpetration and sibling victimisation were precursors of peer bullying perpetration and victimisation one and two years later. However, the strength of the association declined over the course of two years. Impulsive behaviour and social alienation seem to be fundamental influencing factors in the development of sibling bullying and sibling victimisation, respectively. Additionally, children who were involved in peer bullying were more likely to have been involved in sibling bullying, compared to peer neutrals one and two years later. The third study, which was based on the Avon Longitudinal Study of Parents and Children (ALSPAC, Boyd et al., 2012) found that maternal somaticism was the strongest predictor of sibling bullying. Further, the strongest predictor of sibling victimisation was partner-to-mother verbal violence. Symptoms of depression at 16.5 years of age was the strongest outcome of sibling bullying perpetration and victimisation at 12.5 years of age. Children who were peer bully-victims when they were 17.5 years old were more likely to have been sibling pure bullies and sibling bully-victims, compared to children who were peer neutrals. The results suggest that familial factors significantly influence the quality of sibling relationships. Additionally, the findings show that sibling bullying is related to peer bullying, so that children mirror bullying behaviours across social contexts (i.e. family environment and school environment). The findings of this thesis are important for clinical practitioners, social workers, parents and schools. Based on these findings practitioners could tailor family and parenting intervention programs that prevent siblings from establishing conflictual relationships with one another. Particularly, it is suggested that bullying intervention programs should integrate three aspects: family members should play an integrated and active role in their plans to reduce bullying and victimisation; bullying intervention and prevention studies should commence at preschool ages; positive family climate should actively be nurtured, in addition to lowering hostility.

There is an increasing academic and policy interest in subjective well-being (SWB). However, the questions of whether and how public policy can promote children’s SWB remain understudied. This thesis aims to reduce this gap by studying the association between education policy and students’ SWB, with a focus on life satisfaction (LS). To quantitatively study this question, this thesis analyses data on 15-year-old students in 33 countries that participated in the 2015 Programme for International Student Assessment (PISA) study. The analysis draws on Bronfenbrenner’s ecological model of child development and it is conducted by using a range of advanced quantitative methods, mainly multilevel regression. This thesis presents analysis demonstrating an association between several education policy-relevant factors and students ́ LS, which is particularly prominent –and observed in almost all countries- for schoolwork-related anxiety, bullying and parents’ emotional support in relation to school. Results also indicate that schools may play an important role in shaping students’ LS. This is supported by evidence that these associations tend to vary by school, by evidence on the existence of school effects in almost all countries, and by the finding that a proportion – substantial in some countries- of the variation in students’ LS is explained by differences between schools. Moreover, findings suggest that school type and school peers’ characteristics can be important to students’ LS too. In addition, in many countries, the links between schools and education policy and students’ LS differ for girls and boys and for students of different socio-economic status. Finally, in all the analyses described above, there are significant differences across countries. Overall, this thesis makes key contributions to our understanding of whether and how children’s SWB can be influenced by schools and education policy, supporting calls that education policy should also be assessed in terms of its impact in children’s SWB.