藍者恆藍？ 基隆政治版圖之變動與影響因素

Abstract

本研究認為基隆市的政治版圖有其特殊性，因此針對1995年到2009年期間內舉行總計13次的總統、立委、市長以及議員選舉進行分析，分析單位則是基隆市內157個里，透過傳統以及空間分析方法找尋影響基隆市政治版圖分布的要素。 在比較過基隆市各里藍營得票率的差異以及空間分布型態，本研究認為選舉制度、地區屬性、空間自相關以及空間異質會造成藍營得票率的變動。透過傳統以及空間迴歸模型的幫助，就能更精確的了解影響政治版圖的原因。 最後研究結果發現基隆市藍營高得票率的背後是充滿了特例與分歧，同時藍營得票率也會隨著選舉類型不同而呈現差異，能夠解釋總統、立委、市長選舉和議員選舉結果的自變數不盡相同，地區特色的影響力只出現在總統、立委、市長選舉，對議員選舉較無影響，而鄰近效應和空間異質則同時對總統、立委、市長選舉和議員選舉得票率具有影響力。

Because of the particularity of the political geography in Keelung City, this research is to determine the spatial distribution of it. By using the traditional and the spatial analysis, this study tries to find the factors that affect the spatial distribution of the percentage of vote. The percentage of vote comes from 13 elections are the presidential elections, legislator elections, mayor elections and city councilor elections during the period 1995 – 2009, and the data is collected from 157 villages in Keelung. After comparing the differences of the percentage of vote between pan-blue and pan- green parties, this thesis suggests electoral system, local attribute, spatial autocorrelation and heterogeneity of region influence the reason why the percentage of vote changes. Thanks for the help of the traditional and spatial regression model, this research can analyze why and how political geography of Keelung changes more precisely. This research found that behind the high percentage of vote in pan-blue party, there are many exceptional cases and differences; meanwhile, the percentage of vote in pan-blue party changes from different electoral systems. The significant independent variables are different between the presidential elections, legislator elections, mayor elections and local election model, and the inference of local characteristic doesn’t work in the city councilor elections model. Last but not least, the neighboring and heterogeneity of region both have an effect on the presidential elections, legislator elections, mayor elections model and city councilor elections model.