Please use this identifier to cite or link to this item: https://ah.lib.nccu.edu.tw/handle/140.119/124395
題名: 選舉競爭與資訊對投票率的影響:以2005-2014年台灣直轄市與縣市議員選舉為例
The Effects of Electoral Competition and Information on Voter Turnout: The Case of the Local Council Election in Taiwan, 2005-2014
作者: Tsai, 陳宏哲Hung-che Chen蔡宗漢Tsung-han
貢獻者: 選舉研究
關鍵詞: 選舉競爭;選舉資訊;投票率;直轄市/縣市議員選舉;多層次Beta迴歸模型
electoral competition ; electoral information ; voter turnout ; local councilor elections ; multilevel beta regression model
日期: Nov-2018
上傳時間: 22-Jul-2019
摘要: 本文主要在探究選舉制度對於投票率的影響,與傳統文獻不同的是,本文不是從選舉制度的不同特徵進行討論,而是從選舉制度影響投票率的機制著手。本文採取理性選擇理論的觀點,認為選舉競爭程度對於投票率的影響取決於選民掌握的選舉資訊,並且主張同黨候選人與現任者參選有助於選民獲取選舉資訊,進而影響選舉競爭提升投票率的程度。本文以2005-2014年的台灣直轄市與縣市議員選舉為分析案例,資料性質屬於縱橫資料,分析單位是選區-年。因為該地方選舉採用複數選區單記非讓渡投票制(Single Non-Transferable Vote with Multi-Member District Systems, SNTV-MMD),所以相當適合檢驗本文的研究假設。本文採用多層次Beta迴歸模型分析前述資料,分析結果顯示當議員選區中的同黨候選人參選程度或現任者參選比例較高的情況下,若選舉競爭程度愈強,則投票率會愈高。本文也針對這些研究發現提出其政治意涵。
This article examines the effects of electoral competition on voter turnout conditional on the number of co-partisan candidates and incumbent reelection rates. Utilizing the advantages of SNTV-MMD systems such as the variations in the number of co-partisan candidates and incumbent reelection rates across districts, we conduct a data analysis of 486 districts of local councilors in Taiwan from 2005 to 2014, which are time-series cross-sectional data. We analyze the data by employing multilevel beta regression models and the results show that, when the number of co-partisan candidates or incumbent reelection rates are relatively high, strong electoral competition significantly fosters voter turnout. The results have important implications for the effects of SNTV-MMD systems on voter turnout.
關聯: 選舉研究, 25(2), 55-88
資料類型: article
DOI: https://doi.org/ 10.6612/tjes.201811_25(2).0003
Appears in Collections:期刊論文

Files in This Item:
File Description SizeFormat
42.pdf1.82 MBAdobe PDF2View/Open
Show full item record

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.