Chu's crisis or chance? Spatial Analysis of New Taipei City Mayoral Election in 2014

本研究利用空間分析方法，探討103年新北市長選舉朱立倫得票率和與99年相比之得票率變化的影響因素，並指出朱立倫在新北市的危機與轉機。 空間分析方法強調將變數資料以視覺化處理，並利用空間探索分析和空間迴歸模型，證實空間因素對社會現象的影響。本研究以新北市的1032個里為分析單位，從人口結構、經濟社會背景、施政表現、選舉參與和空間因素等面向，分析103年得票率和得票率變化的影響因素。 研究發現，新北市選舉存在鄰近效應和空間異質，以未考量空間因素的傳統迴歸模型對選舉結果進行解釋將有偏誤。103年得票率主要受人口結構和經濟社會背景影響，得票率變化則主要受人口結構影響，而朱立倫在第一任新北市長任期內的施政表現，則是未對選舉結果產生顯著影響。此外，朱立倫在首投族較多的地區，103年得票率較低，高所得和高教育程度的地區，對他的支持出現鬆動，在客家人、原住民和三級產業人口較多的地區，朱立倫的得票也有退步的情形，以上為朱立倫在新北市的危機。相反的，公部門人口較多的地區則沒有鬆動的跡象，仍對朱立倫表現高度支持，而103年選舉藍營選民投票率降低，確實使得朱立倫得票率退步，研究者懷疑此乃對選情過於自信的結果，未來只要適當催票應仍有希望，以上為朱立倫在新北市的轉機。In this research, spatial analysis method was used to examine the factors that determine the change of voting rate for Mr. Chu during the New Taipei City Mayoral Election in 2010 and 2014. Furthermore, the research also points out the crisis and opportunity for Mr. Chu in New Taipei City. Spatial analysis method emphasis on processing the data via visualization. By utilizing exploratory spatial data analysis as well as spatial regression, the impact of social phenomenon due to spatial factor can be proven. The analysis unit in this research is based on 1032 villages in New Taipei City. To analyze the voting rate and change of voting rate in year 2014, population structure, socio-economic backgrounds, government performance, electoral participation and spatial factors were used in the research. The research found that the election in New Taipei City will be biased in terms of interpreting result due to neighborhood effect and spatial heterogeneity, as well as, not considering the ordinary least squares regression for spatial factor. The voting rate in 2014 election was influenced by population structure and socio-economic background. The main reason for the change of voting rate was due to population structure. The performance of Mr. Chu in his term of office had minor effects on the result of the election. Furthermore, the voting rate was lower in area with more first-time voters in the 2014 election for Mr. Chu. There was fluctuation in the voting rate for area with higher income and education. The voting rate had dropped for Mr. Chu in areas with more Hakka, indigenous peoples and the third-level industry. The aforementioned were the crisis for Mr. Chu. On the contrary, areas with more public sector employees had no sign of voting change and had remained high in support for Mr. Chu. More ever, the turnout rate had dropped for KMT supporter. Hence, a drop in voting rate for Mr. Chu. The researcher believed this is due to over confidence in the election. Mr. Chu is still hopeful with an encouragement to vote in the next election