2018年10月

以地址最末兩碼調整之戶中選樣表
Using the Last Two Numbers of an Address for Objective Respondent Selection within a Household

研究論文 Article

謝淑惠
Shu-Hui Hsieh

119-147

台灣於2012年10月1日開始施行「個人資料保護法」（以下簡稱個資法）。為了符合個資法規定，政府機關對於個人資訊的維護與管理更加嚴格，導致戶籍資料僅能提供給政府機構申請，而不再提供學術研究單位做為抽樣清冊。因此，面訪調查將會以門牌地址抽樣的方式代替過去30年來慣用的戶籍抽樣，進而開啟應用戶中選樣於面訪調查之研究。戶中選樣方法首先由Leslie Kish提出，用於從中選住戶地址挑選一位受訪者。本文所提出之戶中選樣表係以門牌地址最末兩碼數字為隨機機制來控制戶中選樣程序，令每一位受訪者被選取的機率均不同。為了比較戶中選樣方法之樣本差異性，我們以電腦模擬方式說明樣本配適性。最後，以2016年科技部補助之門牌地址抽樣計畫預試調查，來說明我們所提出之戶中選樣表的執行成效與結論。

The Personal Information Protection Act (PIP Act) became effective in Taiwan in October 2012. To comply with the stipulations of the PIP Act, the Ministry of the Interior started restricting the use of any personal information from the population register, even for academic research. Unsurprisingly, the large-scale face-to-face surveys with the newly designed address-based sampling (ABS) strategies, after 30 years of reliance on governmental register-based sampling, became less feasible and reliable. ABS often requires selection, in each household, of a single person or one respondent only as the designated (or target) respondent for the interview. Leslie Kish pioneered the random selection of members within a household. Since then, several approaches of selecting household members have been adopted in survey practices. A technique in which the interviewer used the last two digits of the address number to randomly select the designated respondent was designed in this study. To investigate the performance of the proposed method, we conducted a simulation to generate samples by using various sample selection methods. In addition, we used a pilot survey of ABS to explore the effectiveness of the proposed method.

門牌地址抽樣, 戶籍抽樣, 個人資料保護法, 面訪調查, 戶中選樣
address-based sampling, register-based sampling, Personal Information Protection Act, face-to-face survey, respondent selection within the household

社會學
Sociology

https://doi.org/10.7014/SRMA.2018100004

全文下載