| School bullying has become a worldwide social and public health problem,which has brought serious adverse effects on society,families and individuals,especially on the physical and mental development of the bullied.Depression and Internet addiction,as the most common and typical internalizing problems and externalizing problems of junior high school students,are closely related to bully victimization.However,previous studies on the relationship between school bullying and depression or Internet addiction were mostly cross-sectional designs,and limited longitudinal studies have mostly explored the relationship between variables at two time points,which can not explore the independent trajectory of single variables,the dynamic relationships between two variables and the joint trajectories between two variables.What’s more,there are a lack of discussion on the roles of demographic variables such as gender and parent-migrant.Therefore,this study further explores the relationships of bully victimization and depression(internalization)、 bully victimization and Internet addiction(externalization)among junior high school students based on three follow-up survey,and the roles of gender and parent-migrant.Objective: Question 1 investigated the independent developmental trajectories of bully victimization,depression and Internet addiction of the new junior school by establishing the latent growth modeling and the influence of gender and parent-migrant.Question 2 discussed the time-delay bidirectional relationships of bully victimization and depression,Internet addiction among the new junior school by establishing a series of competitive cross-lagged models,as well as the cross-lagged models to compare the differences between gender and left-behind groups.Question 3 investigated the optimal number of categories for the joint trajectories of bully victimization and depression,bully victimization and Internet addiction by establishing a parallel process latent class growth analysis(PP-LCGA).Methods: The questionnaire data were collected by group measurement,selecting a middle school freshmen in Pingdingshan as the participants.The study conducted three follow-up survey at an interval of six months,and the first questionnaire was collected from December 2020.In the three measurements,626 valid questionnaires were collected after eliminating the data missing twice and the questionnaires not answering seriously.For the first question,SPSS21.0 was used to analyze the data.The statistical methods mainly included descriptive statistics,repeated measurement ANOVA and independent sample T test.Meanwhile,Mplus8.3 was used to construct the unconditional linear latent growth modeling.For the second question,Mplus8.3 was used to establish a series of competitive cross-lagged models to explore the time-delay bidirectional relationships between bully victimization and depression,bully victimization and Internet addiction,and the cross-lagged models to compare the differences between gender and parent-migrant.For the third question,based on the individual-centered perspective,Mplus8.3 was used to conducted parallel process latent class growth analysis to explore the optimal number of categories for the joint developmental trajectories of bully victimization and depression,bully victimization and Internet addiction.Results: Question 1:(1)Verbal and relational bullying are more common types of bully victimization among junior high school students;(2)The time main effect and parent-migrant main effect of bully victimization were significant;depression were significant in time main effect,gender main effect,parent-migrant main effect,time and gender interaction effect;Internet addiction were significant in time main effect,gender main effect and interaction effect between time and gender.(3)Bully victimization,depression and Internet addiction showed a linear development trend with group heterogeneity;bully victimization showed a decreasing trend over time,and depression and Internet addiction showed an increasing trend over time.(4)Gender did not significantly predict the trajectory of bully victimization,and gender significantly predicted the initial level of depression and Internet addiction,but did not significantly predict the development rate of both.(5)Parent-migrant had a significant effect on the prediction of the initial level of bully victimization,but not on the prediction of the development rate,and not on the prediction of the initial level and development rate of depression and Internet addiction.Question 2:(1)There was a bidirectional predictive effect between bully victimization and depression,and the bidirectional predictive effect was stable across time;(2)Bully victimization and depression delay bidirectional prediction effect had gender differences.In particular,in the male group,depression had cross-temporal stability in the delayed prediction of bully victimization,but the time-delay prediction of bully victimization on depression was not stable across time;(3)After controlling for gender,the bidirectional prediction of bully victimization and depression was different between the left-behind and the non-left-behind.In the left-behind group,the bidirectional delay prediction of bully victimization on depression was stable across time,but the bidirectional delay prediction of depression on bully victimization was not stable across time.In the non-left-behind,the prediction effect of depression on bully victimization was stable across time,but the prediction effect of bully victimization on depression was not stable across time.(4)There was the delay bidirectional prediction effect between bully victimization and Internet addiction,but the delay bidirectional prediction effect between the two had no cross-time stability,(5)The delay bidirectional prediction effect model of bully victimization and Internet addiction was the same for male and female groups,that was,only bully victimization at T1 had a significant prediction effect on Internet addiction at T2,and other pathways were not significant.(6)After controlling for gender,the bidirectional effect of bully victimization and Internet addiction was also different.In left-behind group,bully victimization at T1 significantly predicted Internet addiction at T2,and Internet addiction at T2 significantly predicted bully victimization at T3,and other paths were not significant;In the non-left-behind group,the bidirectional effect of bully victimization and Internet addiction only existed at T1 and T2,while the relationship was not significant at T2 and T3.Question 3:(1)The joint developmental trajectories of bully victimization and depression showed group heterogeneity,which could be divided into three ideal categories: 6% students in the "danger group",70% students in the "safe group" and24% students in the "risk group".(2)The joint developmental trajectories of bully victimization and Internet addiction was not stable over time,and it was finally divided into three ideal categories: 5% students belonged to the "remission group",5%students belonged to the "danger group" and 90% students belonged to the "risk group".Conclusion:(1)Bully victimization,depression and Internet addiction of the new junior high school show a linear development trajectories,and there exist group heterogeneity.Gender only have significant effect on the initial level of depression and Internet addiction;left-behind only has significant effect on the initial level of bully victimization;gender and left-behind do not have significant effect on the development speed of bully victimization,depression and Internet addiction.(2)Bully victimization and depression have bidirectional predictive relationship,and the relationship between the two variables shows cross-time stability and cross-gender instability;after controlling for gender,the time-delay bidirectional prediction effect of bully victimization and depression is also different in the left-behind and the non-left-behind.There is also a bidirectional predictive relationship between bully victimization and Internet addiction,and the relationship between the two variables shows cross-time instability and cross-gender pattern stability;after controlling for gender,the bidirectional predictive relationship between bully victimization and Internet addiction is also different in the left-behind and the non-left-behind.(3)There is a relatively stable covariant relationship between bully victimization and depression,but the joint covariant trajectory between bully victimization and Internet addiction is unstable. |
