本論文以民國九十年台北縣縣長選舉為研究議題，應用「沉默螺旋理論」探討政治上泛藍與泛綠陣營兩股勢力的選民意見分佈情形，並瞭解不同統獨立場選民的表達意願與害怕孤立的可能因素。此外，研究具有四種意見氣候認知的選民，照鏡知覺（Looking-Glass Perception）、多數無知（Pluralistic Ignorance）、錯誤的共識（False Consensus）、死硬派（Hardcore），在「表達意願」上的差異。 研究結果發現此次選舉選民的民意形成過程的確契合「沉默螺旋理論」的基本假設：當個人意見與預估未來意見一致者，其表達意願會比個人意見與預估未來意見不一致者來的高。在統獨議題上，傾向「支持獨立」的少數選民符合「死硬派」的特徵，積極在各種政治議題上發表己見。另外，當選民支持的候選人與心中認定會當選的候選人是同一人時，認為他人比自己更容易受媒體影響的選民（有第三人效果者），表達意願較高。選民個人支持的候選人與親友支持的候選人一致時，其在公開場合表達己見的意願較高；在控制個人意見與預估未來意見的一致程度下，個人意見與親友意見一致程度並不影響表達意願。因此，在這次的選舉議題上，選民並非透過參考團體來觀察大環境的意見氛圍，而是藉由大眾媒介來察知大環境的意見氣候。具有照鏡知覺（Looking-Glass Perception）認知的選民與具有錯誤的共識（False Consensus）認知的選民，在「表達意願」上沒有太大差異；具有多數無知（Pluralistic Ignorance）認知的選民較具有死硬派（Hardcore）認知的選民在「表達意願」的程度上較低。 This study applied “Spiral of Silence Theory” to testing voters’ willingness to express opinions publicly and to detecting the factors affecting their fear of isolation, if any, during the Taipei County Magistrate’s Election in 2001. The other vital research question is whether voters were willing to voice their standpoint about Taiwan’s future, independence or unification with China or keeping status quo, while facing people with contradictory political maintenance. In addition, based on voters’ perceptions of public opinion climate that were categorized as “Looking-Glass Perception”, “Pluralistic Ignorance”, “False Consensus”, and “Hardcore,” this study attempted to examine if it would reinforce voters’ willingness to express opinions on condition that individual opinion was consistent with perceived public opinion climate. The findings supported the hypothesis of “Spiral of Silence Theory”. Voters were more willing to express opinions publicly when their perceptions of the distribution of public opinion were the same as their own. A small part of interviewers who strongly supported “Taiwan Independence” seemed to behave like so-called “hardcores.” They were very active to voice out their opinions about any political issues. Taking “third-person effect” as the mediating factor, one who believed others would be more influenced by mass media than him/herself was more likely to express his/her own opinion given that the candidate he/she supported was the same as considered to be elected in the election. Regarding the impact of reference groups on voters’ willingness to express their opinions, only when voters and their relatives (or friends) supported the same candidate, did the former tend to openly express opinions. However, the effect of reference groups was cancelled out when voters’ opinion was identical with their perceptions of the distribution of public opinion. In this election, voters perceived the public opinion climate through the mass media instead of through reference groups such as relatives or friends.
