Personal and Situational Factors in Cyberbullying Perpetration and Victimization: Links with Mental Health Outcomes in Adolescence

Introduction:Cyberbullying is a concern among many school-age children (Juvonen & Gross, 2008). Researchers have begun to explore the predictors and outcomes of cyberbullying, such as characteristics of the school environment. For example, a trusting and positive school climate has been shown to have a negative relationship with cyberbullying perpetration (Williams & Guerra, 2007). Negative mental health outcomes such as depression and delinquency have also been linked to cyberbullying perpetration and victimization (Mitchell, Ybarra, & Finkerhor, 2007). The proposed study will apply the General Aggression Model to better understand potential risk factors for cyberbullying, and the subsequent impacts on mental health. We hypothesized that contextual characteristics such as perceived safety, healthy school climate, and appropriate behavioral modeling in school would be associated with a decreased risk for involvement in cybervictimization.

Method:The data come from 28,104 students in 58 Mid-Atlantic high schools. The sample was approximately half of the sample was White/Caucasian, and a third was Black/African American.

Results: Approximately 12.5% of the youth reported being cyber-bullied in the past 3 months.

Among those who were bullied, around 8% experienced cyberbullying once or twice, whereas 4.5% experienced it two or more times in the previous 3-month period. In contrast, 5,848 (20.8) of the students reported being traditionally victimized (i.e., physical, social, verbal bullying only) within the past 30 days. A series of three-level hierarchical linear models (HLM) was fit to assess the personal and situational factors that are associated with cyber victimization prevalence and mental health outcomes. Preliminary results suggest that females (β = -0.05, p<.001), under class (9^th and 10^th grade) students (β = 0.02, p<.001), and traditionally victimized students (β = 0.36, p<.001) were at an increased risk of cyber victimization. At the school-level, a higher concentration of students feeling safe (β = 0.77, p<.05), a higher student mobility (β = 0.01, p<.05), and a higher percentage of minority students (β = 0.003, p<.05) led to increased risk of cybervictimization. We also explored some interactions and found that schools with lower levels of student mobility may help buffer the association between cyber victimization and internalizing symptoms (β = -0.02, p<.01). 

Conclusion: These findings have important implications for identifying risk and protective factors related to cyberbullying and mental health among adolescents. This line of work informs the small body of research focused specifically on cyberbullying prevention.