Longitudinal patterns of involvement in cyberbullying: Results from a Latent Transition Analysis

In the present study, we used Latent Transition Analysis as an innovative approach in cyberbullying research in order to detect multi-facetted involvement patterns. Since developmental aspects of cyberbullying are still poorly understood, we analyzed the stabilities and transition probabilities of these involvement patterns across time using longitudinal survey data. Based on a three-wave panel survey among 1723 pupils (12-15 years old), we identified a five-latent status model to best fit the data. Apart from a large group of non-involved pupils, there were four moderately to heavily involved cyberbullying classes, all characterized by a co-occurrence of perpetration and victimization experiences. We found two moderate and content-specific classes of cyberbullying: gossiping patterns that were predominant among girls and insulting patterns that rather appeared among male and lower-educated adolescents. Moreover, we revealed a heavily victimized group (with mild perpetration) and a very small class of heavy perpetrator-victims. Transition probabilities showed that cyberbullying behavior was quite stable over time.