| Summary: | 碩士 === 中華大學 === 資訊管理學系 === 105 === According to the previous studies, school bullying may cause short-term or long-term physical and mental problems to the bullied students, leaving them with a sense of powerless toward life. Moreover, their confidence and emotions are also hurt, which further influences their performance at school. In the worst case scenario, the bullied students would drop out of school or even hurt themselves. The present study aims to realize the relationship between the changes in class social network indicators of the seventh graders in junior high schools and students' gender as well as the experience of being bullied on campus. By analyzing the social networks of the bullied students, early detection and prevention of negative effects resulting from being bullied could be provided. It is hoped that the present study could serve as a reference material for educational administrations, schools, and teachers.
Eight classes of seventh graders in a junior high school in New Taipei City participated in this study. Two hundred and ten valid subjects were obtained, including 119 male participants and 91 female participants. A questionnaire survey on the experience of being bullied on campus and class social networks was conducted to the participants on the first, fifth, ninth, thirteenth, and seventeenth week respectively in a semester. Each indicator was analyzed by the social network analysis software NodeXL, and the integrated data was computed and analyzed by SPSS18.
The major findings of the present study were summarized as follows. Firstly, in the year before the participants entered junior high school, 31 percent of them already had the experiences of being bullied. Secondly, concerning the experience of being bullied, a higher percentage of the female participants had such experiences than that of their male counterparts. Thirdly, in terms of both network indicators of Clustering Coefficient and Reciprocated Vertex Pair Ratio, the female participants are higher than their male counterparts. Fourthly, whether the participants had the expereince of being bullied or not, there was no significant differences in the indicatores of class social network and happiness. Fifthly, after week seventeen, the participants who had the experience of being bullied had lower In-Degree in the indicator of playmates and confidence, and had higher Clustering Coefficient in the indicator of bosom friends and playmates. They also had higher Reciprocated Vertex Pair Ratio in the indicator of bosom friends and playmates. Sixthly, after week seventeen, the participants who did not have the experience of being bullied had lower In-Degree in the indicator of confidence, and had higher Betweenness Centrality in the indicator of bosom friends and confidence, higher Closeness Centrality in the indicator of confidence, higher Clustering Coefficient in the indicator of bosom friends and playmates, and higher Reciprocated Vertex Pair Ratio in the indicator of bosom friends and playmates. Seventhly, the indicator of class happiness did not vary because of gender and the experience of being builled. However, the indicator of happiness became lower in week five, and yet it showed no significant differences in week seventeen. Eightly, a low correlation was revealed between the indicatores of class happiness and partial class social networks.
|
|---|
