堅毅理論之跨文化比較：青少年攻擊行為、物質使用及心理困擾之研究

Purposes: This project aims to build the resilience model among Taiwanese adolescents by investigating the cross-sectional and longitudinal associations between the promotive factors and problem behaviors (aggression, substance use, mental distress) while considering risk factors. In addition, the project will conduct cross-country comparisons on the resilience models between the Taiwanese and U.S. adolescents. Methods: Northern Taiwan sample was recruited in Taoyuan and New Taipei City. The systematic clustered sampling method is used to select schools. In the first year, 24 junior high schools and 8 senior high schools participate in the project. Totally, 1712 students were participated in the first semester of the 108 Academic year. In the second semester of the 108 Academic year, 1725 students participated. Only 1237 students who provided the same information for computing pseudo-ID of the first wave and the second wave. Miami sample comprised 849 students who were followed since their 6th grade in 2020. The instrument of the project consisted of three sections. First section is negative outcomes which included physical and verbal aggressive behaviors, smoking and drinking, as well as depressive symptoms and anxiety. Second section is multi-level risk factors. Third section is multi-level promotive factors. Multi-levels referred to individual, familial, peer, and school/community levels. To construct the time order between related factors and dependent outcomes, we built multiple linear regression models and multiple logistic regression models where the problem behaviors measured in the second wave served as dependent variables while the risk and promotive factors measured in the first wave were used as independent variables. To avoid collinearity, stepwise linear models and stepwise logistic models were used to filter the most significant factors. Results: In northern Taiwan sample, the results of the tests of goodness-of-fit showed that the sample with completed data of the two waves was not different with the sample in the first wave on background characteristics. In terms of negative outcomes, besides physically aggressive behaviors, all other five negative outcomes, namely verbally aggressive behaviors, smoking, drinking, depressive symptoms and anxiety, were more serious among high school students than middle school students. Physically and verbally aggressive behaviors both decreased during the surveyed year, and other four negative outcomes did not change. Besides smartphone addiction, other significant risk factors decreased. Besides organized activity participation, other significant promotive factors decreased. Results of multiple regression models indicated that sex was related to the six negative outcomes. Regarding the differences between the northern Taiwan and Miami samples, physically and verbally aggressive behaviors decreased significantly in Miami sample. Verbally aggressive behavior was higher in the northern Taiwan sample than in the Miami sample.