總統執政信任度變化的實證分析：以2016至2019年蔡英文總統為例

本研究旨在探討民眾政府「施政滿意度」與「政治人物形象」，對於政治信任的影響。研究對象上，選擇現任蔡英文總統作為研究對象；分析資料範圍自2016年6月上任後到2019年6月，資料來源使用政大選研中心TEDS「總統滿意度電訪及手機/網路調查」中的四次調查資料，分別是甫上任的2016年6月、其次為2017年6月、接著是2018年12月、最後則是2019年6月，歷次資料約間隔半年至一年半。本研究利用多元線性迴歸模型進行分析，在施政滿意度變數上，有經濟發展、兩岸關係、外交滿意度三者，而政治人物形象變數則包含領導能力與品格兩者。實證研究發現：第一，在本文選定的四個不同時期中，民眾重視蔡總統之領導能力與品格；另一方面，隨著時間變化，民眾對形象變數重視的面向有所不同，似乎較沒有固定的規律可循；第二，在不同時期中，兩岸關係、外交方面兩者滿意度皆呈現顯著，而經濟發展則是甫上任沒有；同時，民眾重視施政的面向頗為全面，隨著時間變化，滿意度對政治信任的影響程度沒有顯著差異；第三，整體而言，在政治信任低的時期中，民眾可能期待多一點施政上的實際政策或產出。最後，政黨認同對於政治信任仍具有高度的影響性。

關鍵字

總統；政治信任；政黨認同；施政滿意度；政治人物形象

並列摘要

The purpose of this study is to investigate the impact of “citizen satisfaction” and “the images of politicians” on political trust. The incumbent President Tsai Ing-wen is selected as the research object and the four periods of data used in this analysis are taken from “Telephone and Mobile Phone Interview Survey of the Presidential Satisfaction” conducted by Taiwan’s Election and Democratization Study (TEDS) of Election Study Center, NCCU. There is an interval of 0.5 to 1.5 year between each period of data and they are June 2016 when President Tsai took office, June 2017, December 2018 and June 2019 respectively. The multiple regression analysis is conducted in the study. The variables representing citizen satisfaction include economic development, cross-strait relations and diplomacy while the variables standing for images of politicians comprise leadership and morality. Based on the analysis, several results are found in this study. First, people value both leadership and morality over the periods. However, people attach different level of importance on the two variables in each period. Second, over the periods, people’s valuation on government performance is quite comprehensive. All the hypotheses for citizen satisfaction are significant except the one on economic development in June 2016. Furthermore, economic development, cross-strait relations and diplomacy have equal influence on political trust. Third, overall, during a period of low political trust, people may expect more policies made or more output generated. Finally, party identification still has a strong influence on political trust.