Objective: To identify the latent classes of cyberbullying in college students, and to analyze its relationship with depressive symptoms and suicidal ideation, so as to provide a theoretical reference for effectively intervention of cyberbullying. Methods: Cluster random sampling was used to select 1 094 college students from Liaoning Province and Jilin Province from April to May, 2021. The Cyberbullying Victims Behaviors Scale, the Center for Epidemiological Studies Depression Scale and BSI CV were used in this study and data was analyzed by using latent class analysis, one way analysis of variance and Logistic regression to analyze the data. Results: Cyberbullying of college students could be divided into three latent classes: low incidence group ( 72.40 %), general incidence group (20.20%) and high incidence group (7.40%). There were significant differences in depressive symptoms and suicidal ideation among different classes of cyberbullying ( P <0.01). Logistic regression analysis showed that after controlling for demographic variables, general incidence group and high incidence group significantly and positively predicted depressive symptoms ( β general incidence group =0.88, β high incidence group =2.44, P <0.01) and suicidal ideation ( β general incidence group =0.50, β high incidence group =1.20, P <0.01). Conclusion Families, schools and society should conduct different interventions for college students who suffer from different classes of cyberbullying to reduce depressive symptoms and suicidal ideation and promote mental health of college students.

Objective: To identify potential categories of perceived social support for middle school students and further analyze the association of different categories with bullying defender and depressive symptoms, so as to provide a reference for promoting protection for primary school bullying and reducing depression. Methods: From September to November 2021, a total of 1 021 junior high school students and senior high school students from there middle schools in Liaoning Province by convenient whole group sampling method were taken as the object of the research. Students completed the Multidimensional Scale of Perceived Social Support (MSPSS), Participant Role Questionnaire (PRQ) and Self rating Depression Scale (SDS). Latent Profile Analysis, the R3STEP method and the DU3STEP method and Logistic regression analysis were used to analyze the data. Results: Middle school students perceived social support was classified into four potential categories: low perceived social support group (8.03%, 82), perceived friendship support group(15.38%, 157), moderate perceived social support group (35.26%, 360) and high perceived social support group(41.33%, 422). Logistic regression analysis showed that girls and junior high school students had greater proportions in the perceived friendship support group( OR=0.65, 0.52, P <0.05). There were significant differences in bullying defender and depression symptom among middle school students among different potential categories of perceived social support ( χ 2=47.99, 48.80, P <0.01). The score of bullying defending the low perceived social support group was the lowest(33.75±0.82), but the highest in depressive symptoms(49.82±0.72). Conclusion There is heterogeneity in perceived social support among middle school students. By consciously improving the level of perceived social support for middle school students, families and schools can effectively promote bullying defender and reduce depressive symptoms.