Peer led models to prevent bullying and cyberbullying: how and for whom they can be effective

This dissertation aims to improve our knowledge of the effectiveness of peer-led models within the literature on anti-bullying programs. Previous studies highlighted discordant positions on the effectiveness of involving peers in antibullying interventions (Ttofi and Farrington, 2011; Smith et al. 2012; Lee et al., 2015) and invited scholars to study this model more deeply in order to understand “what works, for whom and under what circumstances.” In order to answer to this issue, we hypothesized that recruitment of peer educators might influence the characteristics of the group of students, and, above all, the effectiveness of the entire program. The First chapter describes the theoretical framework of the research. The social nature of bullying, what is known about antibullying interventions, and the evaluation and the characteristics and strengths of peer-led models within health psychology and prevention of risk behaviours. In the Second Chapter we presented a pilot research on the characteristics of the peer educators in the NoTrap! anti-bullying program (edition 2011/2012). Specifically, we found that, compared with classmates (N= 406; males= 46%), peer educators (N= 118; males= 51%) have higher levels of victimization, perceived support from friends, and prosocial and defending behaviour. In Third chapter we presented two studies from a unique research design carried out within the 2015/2016 edition of the NoTrap! program. Specifically, we used two different recruitment strategies (volunteering vs peer nomination) with a sample of classes with voluntary peer educators (N=500) and a group of classes with peer educators nominated by classmates (N=466). The two studies aimed to understand how peer educators were different in the two groups and how the different methods can affect the results of the intervention. In study 1 we found that voluntary peer educators are more involved in victimization, whereas nominated ones are the most popular and accepted by classmates. In study II, we tried to answer the questions: “under which circumstances and for whom” is NoTrap! effective. A set of linear mixed-effect model (MIXED) procedures showed that the program was effective only in the “voluntary recruitment condition,” in which there was a decrease of bullying and victimization, and a concurrent increase of defending behavior for the whole class (peer educators and their classmates). In the final chapter results are discussed highlighting their implications for future studies.