|Other Abstract||Depression, which is one of the most prevalent psychological symptoms among adolescents, has a wide and far-reaching influence on adolescents’ development. Depression is the product of biology, psychology and social environment. Peer relationship is the most important factor in adolescents’ depression symptoms. Most of the studies on adolescents’ peer relationship and depression symptoms focused on perception of the relationships. Both dependent variables and undependent variables coming from self-report leads to greater risk of common method bias. The causality and mechanism are not. In this study, from the perspective of social network, class network was constructed by nominate method, and the relationship between peer relationship and depressive symptoms of senior high school students was explored by longitudinal follow-up experiment design.
This study selected a total of 751 students from Class 1, Grade 10 of Maotanchang Middle School in Lu 'an City, Anhui Province. The questionnaire survey was completed at 4 weeks, 11 weeks, 20 weeks and 31 weeks after the school started. The contents of the questionnaire included demographic information, depression scale and social network questionnaire. The social network questionnaire was prepared by nomination method, and each person could nominate up to 5 friends in the same class. After verifying the nomination data through the class list, the class network is constructed using the nomination data. Since the school re-divided classes in the fourth questionnaire survey, this study only analyzed the data of the first three times.
The first study focused on the development of adolescent social networks and depressive symptoms during the formative stage of class networks. The results showed that the overall number of peer relationships of adolescents increased first and then decreased. The per capita spotting degree in T2 was significantly higher than that in T1 and T3, and the per capita spotting degree in T3 was significantly higher than that in T1.However, the number of strong relationships showed a downward trend, and the number of reciprocal relationships per capita in T3 was significantly lower than that in T1 and T2.The depressive symptoms of adolescents increased first and then remained unchanged, and the depression scores at T1 were significantly lower than those at T2 and T3.Moreover, parental relationship quality can consistently predict children's depressive symptom scores at T1, T2 and T3.
In the second study, the stochastic actor-based model was used to analyze the co-evolutionary model of peer relationship and depression. Using the multi-group model of Rsiena package in R, the class network of ten classes is included into the model simultaneously. The results showed that individuals with high scores of depressive symptoms in the early stage of class network formation had lower social activity, while peer relationship had no significant effect on depressive symptoms. In the later stage of the formation of class network, individuals with high scores of depression symptoms have lower social activity, while individuals with low popularity and more reciprocal relationships are more likely to have higher scores of depression symptoms.
Conclusion: 1) In the formation stage of class network, the level of depression of adolescents is increasing; 2) The total number of peer relationships showed an inverted U-shaped trend, while the number of strong relationships showed a decreasing trend; 3) The higher the level of depressive symptoms, the lower the social activity.|