Cyberbullying-entrenched or cyberbully-free classrooms? A class network and class composition approach

Cyberbullying threatens young people's well-being. This study is one of the first to examine cyberbullying from a social network perspective. The class-based friendship networks of 103 classes were reconstructed using peer-nomination questionnaires. Closeness centralization and the global clustering coefficient were calculated using social network analysis and subsequently entered into a Poisson-regression. The outcomes of these analyses show that in classes featured by high closeness centralization in the offline and online friendship network, more cyberbullying happens. Additionally, our analyses reveal that in classes featuring a high global clustering coefficient in the online network, less cyberbullying occurs. The composition of the class in terms of gender and ethnicity were not significantly associated with the number of cyberbullying incidents in class.