近年來大學生自傷案件頻傳,顯現大學生心理健康議題探討深具急切性。本研究以國立臺灣大學的大學生為研究對象,了解社會因素(社會支持)和心理因素(兩個人特質:自尊和心理韌性,以及情緒調節自我效能感)對個體心理健康(負向情緒反應、幸福感和生活滿意度)的影響機制,其中主要關注情緒調節自我效能感(regulatory emotional self-efficacy, RESE)的預測效果。此外,本研究也探索對個體幸福感經驗造成影響的幸福感認知結構,並以此結構之維度意涵呼應社會心理因素對心理健康的重要性。本研究分成三個子研究。研究一及研究二樣本依資料蒐集時間,分為預試階段和正式階段兩階段,兩階段有效樣本分別為1526筆和3164筆資料。研究一使用兩階段資料,進行社會心理因素量表和心理健康量表的描述統計分析和信效度檢驗,結果顯示所採用量表皆具良好心理計量特性。研究二使用正式階段資料,以結構方程模型先比較與檢驗兩社會心理因素影響心理健康機制之模型適配,選擇其中適配較佳之模型進行中介效果檢驗,並且再挑選其延伸模型確認模型適配性與進行序列中介效果檢驗。結果顯示模型一為適配較佳的模型,且RESE在社會支持和心理健康的關係之間具中介效果;模型一所延伸之模型三亦具良好適配,且個人特質及RESE在社會支持和心理健康的關係之間具序列中介效果。研究三為一幸福感研究,有效樣本為683筆資料,研究工具包含RESE之語義差異量尺、幸福感量表和幸福感相關語詞之認知判斷。幸福感相關語詞所使用語詞為美好生活、有意義的生活、快樂、幸福感、心理幸福感、主觀幸福感、生活滿意度、生活品質、身心健康及關係和諧。研究三根據迴歸分析結果,發現RESE之語義差異量尺分數對幸福感量表分數仍具預測效果;根據多元尺度分析結果,發現個體的幸福感認知結構是二維認知結構。本研究透過向量擬合結合多元尺度分析結果,更客觀地總結前述兩個維度由「物質層面的-精神層面的」和「個人的-團體的」所詮釋。本研究結果有助研究者和實務工作者了解社會心理因素對大學生心理健康的影響,並可以作為校方和實務工作者輔導與介入的參考。
In recent years, self-harm behaviors among undergraduate students have been reported more frequently, indicating the urgent need to address undergraduate students’ mental health issues. This research investigates the mechanism of social support as a social factor and personal traits (self-esteem and psychological resilience), regulatory emotional self-efficacy (RESE) as psychological factors on National Taiwan University undergraduate students’ mental health (negative emotional responses, well-being and life satisfaction). Additionally, the effect of RESE on individual well-being is tested. Finally, the research explores the cognitive structure of well-being, which influences individual perception of well-being, and echoes the importance of psychosocial factors in mental health through dimensions of the structure. This research is divided into three sub-studies. For Study 1 and Study 2, participants were recruited at two time points, labeled as pilot and formal stage, with 1526 and 3164 students in each respective stage; data were collected through online questionnaires. In study 1, the psychometric properties of the psychosocial factors and mental health scales were examined and ensured in both stage data. Analyzing the formal stage data, Study 2 used structural equation modeling (SEM) to examine and compare two models of psychosocial factors affecting mental health. The better-fitted model was selected to test the effects of mediator, and an extended version of this model was then chosen to assess both the model fit and the serial mediation effects. Study 2 shows that Model 1 was well-fitted, and RESE mediated the relationship between social support and mental health; Model 3, which was extended from Model 1, was also well-fitted, and personal traits and RESE mediated the relationship between social support and mental health serially. Study 3 was a well-being study involving a sample of 683 students. The study utilized the semantic differential scale of RESE, well-being scales, and cognitive judgments of well-being-related terms as the study instruments. The well-being-related terms include good life, meaningful life, happiness, well-being, psychological well-being, subjective well-being, life satisfaction, quality of life, health and harmony. Results of Study 3 indicate that, first, the semantic differential scale of RESE could still predict individual’s well-being score. Second, a two-dimensional cognitive structure of well-being was revealed by multidimensional scaling (MDS). These two dimensions could be interpreted as "material-spiritual" and "personal-general" by using the vector fitting method. The research results contribute to an understanding of the effects of psychosocial factors on the mental health of undergraduate students, serving as a reference for schools and practitioners in counseling and intervention.