| In addition to the family and companions, school plays a core role in young teenagers'life. School belonging is an emotional attitude worthy of attention. The purpose of this study was to explore the communicative attribution, social support and school belonging in the grade, gender, and city of difference between middle students, and the relationship among the three variables, focus of the study due to social support as intermediary between communicative attribution and school belonging.In this study, collected 732 students through random sampling, and questionnaires were to examine the relationship of communicative attribution, social support and school belonging. The results shown that:(1) Whether success or not, middle students'causal attribution of communicative attribution had significant difference in grade and gender, but no significant difference in changes of expectations and emotional reactions. There was gender difference in emotional reactions. Grade difference of social support reflected in the utilization of support dimension, no significant in objective support and subjective support. Gender difference was only significant in total of support. Grade difference of school belonging was significant except for the dimension of school integration. Gender difference was only significant in school inputs.(2) Confirmed the one child difference of junior high school students was significant in ISU whether attribution success or not, and city of significant difference in some dimensions. One child and city differences were existed in subjective support of social support. There was on significant difference in one child of school belonging, but city difference in school environment.(3) In successful situation, there was a positive correlation between 8 causal attribution of communicative attribution and social support, then, regression analysis shown that ESC, IUC and EUC were to predict the impact of social support. Communicative attribution and social support were a significant negative correlation in failure situations, IUC, ISU and EUC were three reasons that the forecast variables of social support.(4) Social support and school belonging were positive correlation. Objective support and total score of support could predict the school belonging.(5) In successful situation, the ISC, IUC, IUU, ESC and EUC of the communicative attribution and the school belonging have a significant positive correlation. ISU only and the school environment, school inputs, peer relations and the total scores of school belonging were significantly correlated. ESU and school integration are not relevant. EUU was a negative correlation with school inputs, and were not relevant with other dimensions. Communicative attribution and the school belonging were most negatively related with each other in failure intuition. Regression analysis shown that communicative attribution could predict school belonging.(6) By hierarchical regression and structural model, found that students'communicative attribution could directly affect school belonging, also indirect effects to school belonging through intermediary of social support. |