부산대학교 도서관

Results: According to factor analysis results, the behavior SE scales comprised 15 items categorized into five factors: extracurricular activities, class or school events, relationships with friends, academic work, and discipline. The emotion SE scales included 18 items of five factors: academic work, class, extracurricular activities, relationships with friends, and school. The recognition SE scales had 13 items of two factors: fostering of qualities and knowledge acquisition. Collectively, they were 46 items. In terms of the goodness of fit of each SE scale measurement model, RMSEA was less than or equal to 0.08; CFI was greater than or equal to 0.95, meaning that the construct validity was good. In terms of the criterion validity, the SE scales showed significant relations to academic achievement and health-promoting behaviors, by which the findings of earlier studies were able to be represented to a certain degree. The internal consistency of the scales was good, with α coefficients greater than or equal to 0.9. The considerations above allowed us to judge the reliability and validity of the SE scales as generally good. After adjusting for confounding variables, we examined the association between the SE scales and health indicators. Results show that the behavior SE was significantly and negatively associated with physical symptom scores and depression scale.