| dc.contributor | 政治系 | |
| dc.creator (作者) | 黃紀 | |
| dc.creator (作者) | Huang, Chi | |
| dc.date (日期) | 2020-04 | |
| dc.date.accessioned | 23-Dec-2020 15:53:42 (UTC+8) | - |
| dc.date.available | 23-Dec-2020 15:53:42 (UTC+8) | - |
| dc.date.issued (上傳時間) | 23-Dec-2020 15:53:42 (UTC+8) | - |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/133375 | - |
| dc.description.abstract (摘要) | 本計畫之目的是建立一套通用之地理區編碼系統（geocoding system），打通微觀與宏觀數據之任督二脈，讓個體民調資料與集 體數據相互整合，跨越並超越層次、全盤掌握脈絡，洞悉政治現象 之全貌。本計畫首先以地理資訊系統（geographic information system, GIS）為基礎，建立一套地理區編碼系統，作為串連民意調 查資料中的個別（微觀）受訪者與其居住地的集體（宏觀）數據資 料的基準。其次，本計畫以前述開發之地理區系統對微觀與宏觀數 據進行地理編碼，透過地理空間作為串接民調個體資料與區位變數 的骨幹，串接微觀與宏觀數據，建構與選區空間單位相同的多層模 型。此種「整合空間」（spatially integrated）的民調資料不但 可以分析傳統上選民特徵、政黨偏好、政治態度等微觀因素對政治 行為的影響，也能檢證學理上社經環境、鄰近效應等空間因素的作 用。接著，本計畫以地理空間頗為多樣異質之新北市為例，於 2018年地方公職人員選舉後進行電話訪問，蒐集個體選民微觀資料 。並根據2018年地方選舉時的重要議題，蒐集以地理區塊為單位的 新北市社會經濟及選舉等集體資料。最後，本計畫整合上述個體民 調資料與集體資料（aggregate data），建立多層模型 （multilevel model）加以分析，將蒐集之資料應用至選舉投票研 究，分析新北市選民在2018年市長、市議員選舉中之一致與分裂投 票。 | |
| dc.description.abstract (摘要) | This research project plans to develop a unified geocoding system so as to spatially integrate micro-level survey data with macro-level area (or ecological) data. The purpose of this micro-macro data linkage through geocoded survey data is to emancipate investigators from given data constraints and stimulate context-rich and spatially integrated multilevel analyses. Survey research is undoubtedly a major approach to observe and measure individuals’ attitudes, opinions, and behavior. In his seminal article, Robinson (1950) warned the possibility of erroneously drawing conclusions about individuals solely from the data of groups. On the other hand, however, an overly strict limitation of analysis at individual level may well lead to another fallacy of ignoring contextual effects. Recent decades have witnessed the growing awareness of the importance merging individual- and aggregate-level data into contextual analyses. Yet the fact that most survey data sets remain limited to individual respondents’ records often hinders analysts from reaching out relevant higher-level data, which are typically scattered in different archives and based on cross-cutting geographical boundaries. Thus, this two-year research project has two objectives: 1. In terms of methods: This project will develop a unified geocoding system for Taiwan so as to spatially integrate micro-level survey data with macro-level area data based on the geographic information system (GIS). This geocoding system will serve as the key in linking respondents in survey data with the related area-based social, economic and electoral data released by the various government agencies (including the General Accounting Office, the Ministry of Interior, and the Central Election Commission) as well as academic data archives (such as the Taiwan’s Political Geographic Information System, TPGIS). 2. In terms of applied research: This project will apply multilevel model to the spatially integrated and geocoded survey data to incorporate local contextual effects into traditional analysis of voting behavior with political attitudes, party identification and demographic variables. Geocoding will first be applied to the planned TEDS 2017 face-to-face survey. The same geocoding system will then be extended to a telephone interview to be conducted after the 2018 local elections of the New Taipei City. Furthermore, small area estimation and spatial microsimulation methods will also be applied while constructing electoral districts’ social and economic indicators. | |
| dc.format.extent | 6411416 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.relation (關聯) | 科技部, 執行起迄：2018/08/01~2019/07/31, MOST 106-2410-H-004 -087 -MY2 | |
| dc.subject (關鍵詞) | 貫通微觀與宏觀資料、空間整合、地理區編碼系統、開放資料、多 層模型、分裂與一致投票 | |
| dc.subject (關鍵詞) | Micro and Macro Data Linkage, spatial Integration, geocoding system, open data, multilevel model, split-ticket voting | |
| dc.title (題名) | 貫通微觀與宏觀數據：民調與區位資料之空間整合與多層分析 2/2 | |
| dc.title (題名) | Micro and Macro Data Linkage: Spatially Integrating Surveys with Area-Based Open Data for Multilevel Analyses | |
| dc.type (資料類型) | report | |
