Work

Online Comments, Offline Crisis: Examining Sleep Duration and Sleep Quality as Potential Mediators in the Relationship Between Cyberbullying Victimization and Mental Health Among an Adolescent Cohort

Year: 2026
Type: dissertation
Abstract: Formal cyberbullying intervention initiatives have yet to evaluate sleep as an avenue to support affected youth. Sleep is a potential point of intervention, as inadequate sleep may be involved in the temporal pathway linking cybervictimization to poor mental health. More expansive mediation research is needed to inform the utility of cyberbullying sleep-based intervention studies. Survey data from a three-year linked sample of adolescents (N = 5,123) participating in the COMPASS study from Year 10 (Y10: 2021/22) to Year 12 (Y12: 2023/24) were analyzed. Two generalized linear regression models were tested to explore the prospective association between cybervictimization and each sleep metric (i.e., duration and quality) one-year later. Four parallel mediation models were then run to evaluate both sleep duration and quality as mediators in the prospective relationship between cybervictimization and each mental health outcome (i.e., anxiety symptoms, depressive symptoms, flourishing and complete mental health [a composite measure combining all three mental health varaibles together]). Cyberbullied youth reported 10-minutes shorter sleep duration one-year later (ß = -10.47, 95% CI: [-20.11; -0.84]), but similar sleep quality (ß = -0.03, 95% CI: [-0.13; 0.08]) relative to their peers, adjusting for sociodemographic characteristics. The association between cybervictimization and each sleep dimension (duration: ß = -9.47, 95% CI: [-19.11; 0.16]; quality: ß = -0.01, 95% CI: [-0.11; 0.09]) was attenuated slightly once non-sociodemographic confounders were added (i.e., parental support and screen time). Both sleep duration and quality functioned as partial mediators in the prospective relationship between cybervictimization at baseline and mental health two-years later. In total, 21.6%, 19.6%, 15.5%, and 14.1% of total effect in flourishing, depressive symptoms, complete mental health, and anxiety symptoms were mediated by sleep duration and quality, respectively. When adjusting for baseline sleep and mental health, sleep no longer mediated the relationship as the total, indirect, and direct effects became non-significant. Sleep is therefore not a robust mediator in the relationship between cybervictimization and poor mental health. Mediation effects were highly sensitive to baseline differences in sleep and mental health, requiring further examination of the short-term associations between cybervictimization, sleep, and mental health.
Language: English
Field: Psychology
Open Access status: green
OpenAlex ID: W7213728803
Cites: 145
Cited by:
Related to: 10