应用相对应分析法对客家庄在投票结构之探索性研究
【外文题名】Correspondence Analyses of the Vote Structure in Hakka Townships: An Exploratory Study
【作者】 黄永政
【外文关键词】 correspondence analysis election study
【导师姓名】杨传凯
【学位名称】硕士
【学位年度】2009
【学位授予单位】暂无
【所属分类】暂无
【录入时间】2009-01-01
【全文挂接】 读秀挂接
【摘要】料分析的方式投影到歐式空間的平面圖，並利用三角圖示法與資料散佈圖目視總體分佈，來探索臺灣公職人員選舉的政治空間。使用資料為從1992-2008年六次立法委員選舉投開票資料，研究發現樣本點第一集群「高密度高客家人口鄉鎮」、第二集群「高密度中客家人口鄉鎮」與第三集群「高密度低客家人口鄉鎮」，在相對應分析結果圖中距離相近應屬於同類，變量點泛藍陣營六次得票數的距離亦相近，關聯性方向是一致的表示這些變量在選民的心目中具有某種程度的同義性。同時這三個樣本點群又與泛藍陣營六次得票數變量點群相近。所以，這三個同一類型的樣本點群將為鄰近的六變量點群所表徵。即表示在選民的心目中這三個集群的定位是類似的就是泛藍的票倉。...
【外文摘要】This paper aims to do an exploratory research on the voting pattern of the village of Hakka in order to assess the influence of the election results through grouping two variables, urban Hakka population density and Hakka population , in the units of village and town and doing a two-dimensional analysis of the composition of the cluster. The sample point is the location data of each cluster of townships, and the variable is the demography of votes of political parties or camps. The method is correspondence analysis, and the triangle graphic interpretation and the material dispersion pattern are used to perceive population distribution and to explore the political space.Using data from the six legislative elections between 1992 and 2008, this study finds that the first cluster, "high density and high population in the Hakka Township", the second cluster, "high density and medium population in the Hakka Township", and third third cluster, "high density and low population in the Hakka Township"; should belong to the same variable points for they are close to each other on the graph generated by the correspondence analysis. Moreover, the pan-blue camp s variable points are equally distant from each other in the six elections, and the direction of relevancy is consistent, which indicates that these variables in voters mind are synonymous to some degree. At the same time, these three sample point clusters are close to the pan-blue camp s six variable points of votes. Therefore, these three samples point groups of the same type will be characterized by the six point groups of the variable. That means that in voters mind the orientation of three clusters is equal to the pan-blue camp s ticket bunker....