查詢結果分析
來源資料
頁籤選單縮合
題名 | 臺北市選民分裂投票之研究民國八十七年市長市議員選舉之分析=Split-Ticket Voting in 1998 Taipei Mayoral and City Council Elections |
---|---|
作者 | 吳怡銘; Wu, Yi-ming; |
期刊 | 選舉研究 |
出版日期 | 20010500 |
卷期 | 8:1 2001.05[民90.05] |
頁次 | 頁159-209 |
分類號 | 573.36 |
語文 | chi |
關鍵詞 | 區位謬誤; 分裂投票; 臺北市選舉; 策略投票; Ecological fallacy; Split-ticket voting; Taipei election strategic voting; |
中文摘要 | 選民的投票意向一直是政治學界注意的焦點,但是在秘密投票的限制下,能得知「個別」選民的投票態度與決定,唯有依靠民意調查,不過所得之答案的真實性則令人懷疑。對選舉唯一能確定的是,選舉後選民「集體」決定後的結果,但若是希望藉由得票結果資料作為個人投票態度的推論,恐怕會有「區位謬誤」(ecological fallacy)的風險。本文以八十七年台北市選舉各投票所之得票記錄分析選民對各政黨市長和市議員候選人分裂投票的現象,嘗試運用King的EI模型,跨層次推論分析選民在三黨市議員候選人與三黨市長候選人之間的抉擇,並初探性利用統計方法降低區位謬誤的風險。結果發現,與外界預期新黨市議員支持者大多採取分裂投票,投票支持國民黨市長候選人馬英九,其分裂投票率高達98.29%。另外,國民黨市議員支持者亦有分裂投票的情形,主要分裂投票對象為民進黨市長候選人陳水扁,分裂投票率為48.28%。 |
英文摘要 | Voting tendencies has been the focus of political field. Due to the limitations of the secretive voting system, the only way to learn each indi-vidual's voting preferences is to take polls. However, the reliability of the poll results is questionable. One thing that we can be sure of is that the voting results reflect every citizen's decision. It would be a risk of e-cological fallacy if one hypothesizes the voting attitude based on the voting result. Based on the voting result in the Taipei election of 1998, first of all, this thesis gives an analysis of the phenomenon of split-ticket voting between political parties in City Mayor and City congress. Applying King's EI model analyzes how citizens chose their desired candidates a-mong the three political parties. It also experimentally applies the statis-tics methods to lower the risk of ecological fallacy. So it turns out that as expected, the supporters of the New Party adapt split-ticket voting by voting KMT's candidate Mao Ying-jo. The rate of split-ticket voting reaches 98.29%. In addition, split-ticket voting also happens to the sup-porters of KMT's congress. The voters in split-ticket voting vote for DPP's City Mayor candidate Chen Shuei-Bien. The split-ticket voting rate registers 48.28%. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。