Accuracy of cyber victimization, peer bullying, dark triad, risk behaviors and prosocial behaviors in classifying cyberbully and non-cyberbully adolescents

The aim of this study is to determine to what extent adolescents' cyber victimization, peer bullying, dark triad personality traits, risk behavior and prosocial behaviors predict cyberbully and non-cyberbully adolescents. This research is a descriptive quantitative research conducted in a cross-sectional design and designed in an exploratory correlational model. A total of 952 students studying at 9 faculties and 1 vocational school at İnönü University in the 2021-2022 academic year participated in the research. Revised Cyberbullying/Victimization Inventory 2, Peer Bullying Scale, Short Dark Triad Scale Turkish Form, Risk Behaviors Scale, Prosocial Tendencies Scale and personal information form were used as data collection tools in the study. The data of the study were analyzed by percent-frequency analysis and logistic regression analysis. Before performing the logistic regression analysis, it was determined that the sample was large enough, outlier data were not included in the analysis, and the correlation coefficients between the variables were examined and it was seen that there was no multicollinearity problem. As a result of the research, it was determined that 47% of the adolescents in the sample were involved in the cyberbullying process as a bully, victim or bully/victim (both bully and victim); 53% of them were not involved in cyberbullying. It was determined that 30% of the adolescents are cyberbullies, 40% are cyber victims, 7% are only cyberbullies, 17% are only cyber victims, and 23% are bullies/victims. In the logistic regression model obtained, it was determined that there was a relationship between cyberbullying and the predictive variables in the study, that the goodness of fit of the model was sufficient and it explained 25% of the variance in cyberbullying. It was determined that the model accurately classified cyberbully and non-cyberbully adolescents at the rate of 79%, and cyber victimization, peer bullying, alcohol use, antisocial behaviors and nutrition habits among the predictive variables in the model made significant contributions to the prediction of cyberbully and non-cyberbully adolescents.