Please use this identifier to cite or link to this item: https://ah.nccu.edu.tw/handle/140.119/70138


Title: 「2001年台灣選舉與民主化調查研究」抽樣設計:分層效果、抽樣誤差與設計效果之分析
Other Titles: Stratification, Sampling Error, and the Design Effect of the 2001 TEDS Sampling Design
Authors: 許勝懋;洪永泰
Hsu, Sheng-Mao;Hung, Yung-Tai
Contributors: 政治系
Keywords: 抽樣設計;分層效果;抽樣誤差;設計效果
sampling design;stratification;sampling error;design effect
Date: 2003.11
Issue Date: 2014-09-25 10:44:18 (UTC+8)
Abstract: 抽樣設計之優劣對於查結果有很大的影響,而分層設計更是考量研究目的和推論效果等諸多因素中的第一步。「2001年台灣選舉與民主化調查研究」應用多變量分析中的主成分分析、集群分析等方法,以台灣地區各鄉鎮市區的人文區位資料進行分層。本文運用已完成的調查資料透過變異數分析與卡方分析,探討這一項抽樣計畫的分層效果。另外配合原有的抽樣設計,以兩組獨立的半樣本,針對一群有代表性的變數計算抽樣誤差以及估計整個抽樣設計的設計效果(Design effect),作為統計推論與日後研究設計的參考。
Among many factors that accommodate research objectives and the efficiency of a survey design, stratification usually comes first. The 2001 TEDS Survey employed some multivariate techniques using ecological data from 359 townships in Taiwan for stratification. This article examines the stratification efficiency of the sampling design, estimates sampling errors for selected variables using independent half-sample method, and calculates design effect for inferential purposes and further usage.
Relation: 選舉研究 , 10(2) , 59-91
Data Type: article
Appears in Collections:[選舉研究 TSSCI] 期刊論文

Files in This Item:

File Description SizeFormat
59-91.pdf1500KbAdobe PDF839View/Open


All items in 學術集成 are protected by copyright, with all rights reserved.


社群 sharing