Development and validation of a scale of cyberbullying and online aggressive conduct in Brazilian adolescents

AI Summary

SCOAC developed as a 17-item, three-factor scale (Threat; Defamation/Exposure; Emotional Problems) with excellent CFA fit indices.
High reliability (total α=0.915; ω=0.919; subscales α=0.837-0.873) and strong convergent and criterion validity including ρ=0.649 with FCBVS.
Good screening accuracy (AUC 0.85; cutoff ≥8); 29.9% flagged with cyberbullying showed greater distress, impulsivity, aggression, and substance involvement.

Front Psychiatry. 2026 May 5;17:1759871. doi: 10.3389/fpsyt.2026.1759871. eCollection 2026.

ABSTRACT

BACKGROUND: Cyberbullying (CYB) is an increasing problem, especially among young people, requiring new ways to measure this construct. Currently, there are gaps in the development of a scale to validate the identification of CYB. Therefore, this study aimed to develop and validate the Scale of Cyberbullying and Online Aggressive Conduct (SCOAC), considering a sample of 642 adolescents (11-17 years old).

METHODS: For the internal structure, we conducted Exploratory Factor Analysis (EFA) and Network Analysis (NA) on 60% of the sample, followed by Confirmatory Factor Analysis (CFA) on the remaining 40%, with multigroup invariance analyses by gender. We also evaluated external evidence, including criterion validity (using the Florence Cybervictimization Inventory – FCBVS), convergent validity (SAS-SV, SPAI-SF, STDS, IAT, PIUQ-9), and nomological validity (DASS-21, DERS-18, UPPS-P, aggression, quality of life, CRAFFT/CESARE). Predictive validity was examined with an ROC curve based on risk classes from Latent Profile Analysis (LPA).

RESULTS: EFA indicated KMO = 0.890 and explained 72.4% of the variance; the final 3-factor solution with 17 items (Threat; Defamation/Exposure; Emotional Problems) was confirmed in the CFA (CFI = 0.991; TLI = 0.989; RMSEA≈0.030; SRMR = 0.048). The reliability of the SCOAC was high (total α=0.915; total ω=0.919; subscales α=0.837-0.873). In criterion validity, the total SCOAC score correlated with the FCBVSs instrument (ρ = 0.649). In convergent validity, we found higher correlations with indicators of smartphone use/messaging. In the nomological validity analysis, correlations were moderate for anxiety (ρ=0.437), stress (ρ=0.404), and depression (ρ=0.387). The ROC analysis showed an AUC of 0.85 (95% CI 0.82-0.88) and a cutoff point of ≥8. 29.9% were classified as having signs of CYB and demonstrated greater emotional distress, impulsivity, emotional dysregulation, aggression, poorer quality of life, and increased substance involvement compared to others.

CONCLUSIONS: The SCOAC has a stable structure, high reliability, strong evidence of validity, and good screening accuracy for CYB, making it a brief, practical tool for identifying cyber risks among adolescents.

PMID:42165063 | PMC:PMC13183630 | DOI:10.3389/fpsyt.2026.1759871

Document this CPD