Longitudinal associations between cybervictimization, cyberperpetration, and cyberbystanding: Cross-lagged panel analysis

In recent years, numerous studies have been carried out showing the longitudinal association between cybervictimization and cyberbullying perpetration, including studies with models that allow estimating the directional effects between variables at different times. However, very little attention has been paid to the dimension of cyberbystanding within this process, and it is key in the processes that legitimize and foment the dynamics of violence. The objectives are: 1) to analyze the autoregressive behavior over time of the cybervictimization, cyberperpetration, and cyberbystanding dimensions; 2) to explore the cross-prediction relationships between each dimension and time. For this purpose, an analytical and longitudinal study was conducted in three waves with approximately six months between each one, in which 1052 Spanish students (43.7 % boys, n = 460, range 11-17 years) participated. They were from 12 schools in seven different regions of Spain. A cross-lagged panel model was conducted to test the bidirectional and longitudinal relationships between cyberperpetration, cybervictimization, and cyberbystanding in adolescents. The results suggest that all autoregressive effects were positive and significant, implying that all three dimensions were stable over time. In addition, there were several significant cross-effects: between cybervictimization at T1 and cyberbystanding at T2 and between cybervictimization at T1 and cyberbullying perpetration at T2. However, the rest of the cross-effects were nonsignificant (p >= .05). The practical implications of these results in possible prevention programs are discussed, highlighting digital literacy to prevent risks on the Internet and reduce cybervictimization.