第八屆立法委員質詢行為與臉書經營關係之研究

Abstract

本研究具有延續、延伸以及探索之意義。在延續部分，接續前人對我國立法委員的立法問政行為影響因素研究，而取其有關立法委員於院會質詢的分析結構，因應時代背景與制度不同，作進一步的驗證與分析；在延伸部分，由於受到2008年美國總統大選候選人成功利用臉書造勢之鼓舞，我國立法委員在加入以及經營社群網站上表現顯得積極，本研究亦試圖分析立法委員臉書經營之關聯因素；最後，在探索的部分，本研究實驗性地採用文本分類技術中的TFIDF方法，連結立法委員質詢行為與臉書經營行為，以探究其影響因素。 研究結果發現，前人所分析之質詢行為的影響要素已不足於說明第八屆立法委員第一會期與第二會期的質詢行為模式；而研究中發現越資深及未擔任委員會召集委員的立法委員，及前次選舉中與主要對手之選票差距越小之區域立委，其質詢次數越多。 在立法委員臉書經營行為方面，因臉書經營行為層次的不同，也有不同的影響要素。其中，在不同層次延續其影響力的因素有「政黨」、「代表性質」、「教育程度」、「年齡」以及「性別」。 最後，在影響立法委員臉書經營行為連結其質詢行為之要素方面，以是否擔任委員會召集委員之因素最為顯著；而在較寬鬆的認定上，區域立委在前次選舉與主要對手之選票差距，亦有顯著性。 透過本研究，不僅能初步了解第八屆立法委員質詢行為、臉書經營行為，及立法委員質詢行為透過臉書傳遞的影響因素，亦為未來的研究者提供在政治學研究領域運用自動文本分類的實例。This paper has significance of continuation, extension and exploration. In continuation part, it follows the research on the factors influencing legislators’ representative behavior in Taiwan conducted by predecessors,to take analytical structure of inquiry made by relative legislators in the meeting of legislative Yuan, so that it makes further validation and analysis under different time backgrounds and systems. In extension part, inspired by the U.S. presidential election candidates who successfully made sensation by using Facebook in 2008, Taiwan’s legislators perform actively on joining and managing community websites. Moreover, the study attempts to analyze the correlated factors of legislators managing Facebook. Finally, in exploration part, the study adopts the TFIDF method of text classification technique experimentally, with the combination of the legislator inquiry behavior and the Facebook business behavior, to explore its influencing factors. It is found from the study results that the influencing factors of the inquiry behavior analyzed by predecessors have been not enough to explain inquiry behavior modes of the legislators in the eighth term during the first and the second sessions. It is also found that the more senior legislators who have not been appointed as the committee chairperson and regional legislators who had little vote gap with their main rivals in the last election have more inquiries. In the aspect of Facebook operation behavior of legislators, the influencing factors vary with different levels of Facebook operation behaviors. The factors which continue its influence in different levels include: political party, nature of representative, education level, age and gender. Finally, in the aspect of factors combining the Facebook operation behavior of legislators and its inquiry behavior, it is the most significant factor whether they are appointed as committee chairperson. While regarding looser authentication, the vote gap between the regional legislators and their main rivals in the last election also shows significant influence. Through this study, it can not only make us know about inquiry behavior and Facebook operation behavior of the legislators in the eighth term, as well as the influencing factors of legislators’ inquiry behavior transferred via Facebook, but also provide embodiments of using automatic text classification in political research field for researchers in the future.