Identifying bullying victimization among Chinese children and adolescents: a psychosocial risk modeling study
Year: 2026
Type: article
Study design: Observational Study
Abstract: Bullying victimization threatens adolescent mental health, yet validated risk-identification tools for Chinese students remain scarce. This study aimed to develop and internally validate a psychosocial model for identifying students likely to report victimization. A school-based cross-sectional survey was conducted among 8,145 students in eastern China through class-based cluster recruitment. Standardized self-report measures assessed individual-, family-, and school-level factors, as well as demographic variables. Participants were randomly divided into a training set (70%) and a test set (30%). Chi-squared and Mann-Whitney U tests were conducted for preliminary screening, followed by binary logistic regression to identify variables independently associated with victimization. After constructing a nomogram based on significant variables, the model’s performance was evaluated using receiver operating characteristic (ROC) analyses and calibration assessments, while decision curve analysis (DCA) was conducted to examine the model’s potential decision value in both datasets. A total of 21.36% of students reported being bullied. Eight significant predictors were identified: boy, primary school, higher depressive and anxiety symptoms, poorer family function, heavier academic burden, poorer classmate relationships, and lower social self-esteem. The model demonstrated good discrimination, with an area under the curve (AUC) of 0.778 (95% CI: 0.763–0.792) in the training set and 0.782 (95% CI: 0.760–0.804) in the test set. Calibration was also satisfactory (training set: Hosmer-Lemeshow χ² ( P ) = 6.10 (0.636), slope = 1.00; test set: χ² ( P ) = 8.29 (0.406), slope = 1.01). DCA showed that the model provided greater net benefit than the “All” and “None” strategies across threshold probabilities of 0.10–0.60 in both datasets. These findings suggest that the ecologically grounded risk model may support the preliminary identification of Chinese students who are more likely to report victimization and may benefit from further assessment in compulsory education settings.
Source: BMC Public Health
Institutions: Chulalongkorn University, Wenzhou Medical University
Language: English
Subfield: Social Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $3,290
OpenAlex ID: W7235619880
