台灣民眾民主態度及相關因素路徑關係之研究

Abstract

本研究係以「民主滿意度」、「民主支持」兩項指標探討台灣民眾的民主 態度，以了解台灣民眾對於民主政治實際運作的感受。本研究的研究目的有三： 第一是探討不同背景變項民眾的民主滿意度、民主支持之差異。第二是建構並 驗證本研究建構的結構模式，歸納台灣民眾的國家經濟評價、執政評價對政治 信任感、民主滿意度、民主支持是否存在直接或間接的影響。第三則是探討本 研究建構的結構模式能否通過多群組分析層層的檢驗，分析各群組建構系統的 平均數有無顯著差異。本研究以文獻分析、驗證性因素分析、結構方程模式與 多群組分析做為主要之研究方法。並以「2012 年總統選舉面訪案」(TEDS2012P) 所調查的資料加以實徵。 首先，研究結果發現，在差異分析中，台灣民眾民主滿意度、民主支持兩 向度的之整體情況為中上程度，顯示整體而言，台灣民眾對民主政治是持肯定 的態度。經施以平均數和變異數的檢定，不同性別、社團參與、政治知識的台 灣民眾在民主滿意度、民主支持的平均數並無顯著差異。而不同教育程度、年 齡的台灣民眾在民主滿意度的平均數達到顯著差異，但在民主支持的平均數並 無顯著差異。 其次，經驗證性因素分析，本研究測量模式無論信度、效度、估計參數， 各觀察變項之標準化殘差常態分配檢定，均達可接受水準。進一步經結構方程 模式整體適配度檢定及競爭模式分析，本研究的修正模式，無論在模式基本適 配度、內在品質與整體適配度都非常理想，且為最佳的模式。研究結果顯示， 國家經濟評價、執政評價是影響民眾政治信任感的主要因素；國家經濟評價、 執政評價、政治信任感是影響民眾民主滿意度、民主支持的重要變項。符合理 性選擇理論、公民資質理論的觀點。此外，國家經濟評價透過政治信任感間接 影響民眾民主滿意度、民主支持，因此，「政治信任感」中介變項的角色被驗證 確實成立。 此外，本研究以性別、社團參與、政治知識等三個變項，對修正模式進行 群組分析，經驗證結果顯示，即使加入這三個干擾因素後該模式依然維持可接 受的適配度。本研究也發現，不同群組在五個潛在變項的平均數也無顯著差異， 不同群組的平均數可視為相等，顯示台灣民眾在國家經濟評價、執政評價、政 治信任感、民主滿意度、民主支持五個潛在變項的狀況是相似的，也顯示本修 正模式並不會因為性別、社團參與、政治知識的不同，而出現推論與適用性的 限制。This study explored popular attitudes toward democracy by examining the people’s degree of satisfaction and support for democracy. The purpose of the study was to compare how satisfied people from different backgrounds were with democracy and to what degree they supported democracy. A structural equation model of the popular attitudes toward democracy in Taiwan was constructed and examined, and the relationships among the sociotropic evaluation of the economy, satisfaction with the current incumbent, political trust, satisfaction with democracy, and support for democracy of the Taiwanese people were determined. This structural equation model was verified by conducting a multiple-group analysis. The methods applied in this study were literature analysis, confirmatory factor analysis, structural equation modeling, and multiple-group analysis, and was tested using survey data collected in Taiwan’s Election and Democratization Study Project of 2012 (TEDS 2012P). The main findings of this study were the following: First, it was determined that the degree of people’s satisfaction and support for democracy was at a high-intermediate level. The degree of people’s satisfaction with democracy differed significantly depending on their age and education level. Second, the item and construct reliabilities as well as the differentiated validity of the measurement systems were satisfactory for constructing the structural equation model. Regarding the goodness-of-fit indices of the measurement models and analysis with 4 competition patterns, the results indicated that this modified model fit well and was supported. The analysis results showed that the sociotropic evaluation of the economy and satisfaction with the current incumbent considerably and directly influenced political trust. It revealed that the people’s satisfaction with democracy and support for democracy were derived from the influence of the sociotropic evaluation of the economy, satisfaction with the current incumbent, and political trust. The analysis results corresponded to the explanations of rational choice theory and citizenship theory. Moreover, an analysis conducted using the modified model revealed that a sociotropic evaluation of the economy substantially and indirectly influenced people’s satisfaction and support for democracy through political trust. In other words, political trust played a mediational role in explaining popular attitudes toward democracy. Finally, multigroup analyses of this modified model was also verified by considering the gender, association participation, and political knowledge of the participants, and it was proven that the fitness of this model was still acceptable and that the means of the 5 constructs were equal. These findings show that the model is acceptable for explaining the democratic attitudes of Taiwanese people.