眾裡尋「輕」千百度：電訪戶中選Young之調整

Selecting the Young Respondents: Adjustment of Random Selection within Household

10.7014/SRMA.2019040001

陳鴻嘉(Hung-chia Chen)；柯秀憓(Siou-huei Ke)

戶中選樣 ； 戶籍資料 ； 普查 ； 電話調查 ； census ； household registration ； telephone survey ； within-household sampling

調查研究-方法與應用

42期（2019 / 04 / 01）

7 - 81

繁體中文

現今住宅電話調查中年輕樣本不足的情形一般皆歸因於唯手機族的日漸增多。但事實上，在年輕人多半住在多人戶的臺灣家戶結構限制下，現行住宅電話調查先等機率抽戶，再等機率抽人的抽樣設計，本來就不容易接觸到年輕樣本。準此，本研究以2010普查資料為依據調整戶中選樣表，試圖改善年輕樣本不足的情況。結果顯示，抽樣模擬與實際撥打的成效皆不錯。因此，本研究建議每年根據戶籍或普查資料調整戶中選樣表。

The most important reason for the lack of young respondents in telephone surveys may not be the growth of cell-phone-only users among the young generation. Two-stage sampling and household structures are the major factors. Therefore, this study adjusts the method of random sampling within household based on the 2010 census. The results indicate that the adjustment improved participation among young people significantly. Thus, the current research suggests that it is necessary to adjust random selection within household on the basis of census or household registration data annually.

1. 杜素豪, Su-hao,羅婉云, Wan-Yun,洪永泰, Yung-Tai(2009)。以入選機率調整法修正調查推估偏差的成效評估。政治科學論叢，41，151-176。
   連結：
2. 俞振華, Chen-hua,涂志揚, Chih-yang(2017)。探討以電訪資料及「入選機率調整法」修正網路調查偏誤的可行性。政治科學論叢，73，81-125。
   連結：
3. 洪永泰, Yung-Tai(2005)。台灣地區抽樣調查各種母體定義、抽樣底冊和涵蓋率的比較。調查研究，18，9-44。
   連結：
4. 洪永泰, Yung-Tai,洪百薰, Baai-Shyun,林宇璇, Yu-Hsuan,呂孟穎, Ming-Ing,許勝懋, Shen-Mao,吳淑惠, Shu-Hui,卓仲彥, Chung-Yen,徐書儀, Shu-Yi(2014)。手機使用對臺灣地區電話調查涵蓋率之影響評估。調查研究，31，7-30。
   連結：
5. 張鐙文, Teng-wen,黃東益, Tong-yi,洪永泰, Yung-tai(2017)。住宅電話與手機雙底冊調查的組合估計：以 2016總統選舉預測為例。選舉研究，24(2)，65-96。
   連結：
6. 許勝懋, Shen-Mao(2015)。「唯手機族」對未來電話調查的挑戰與啟發。調查研究—方法與應用，34，33-65。
   連結：
7. 陳信木, Hsin-Mu,林佳瑩, Chia-Ying,邱泯科, Min-Ko(2006)。臺灣家戶結構現況下的戶中抽樣課題。調查研究—方法與應用，20，45-91。
   連結：
8. American Association for Public Opinion Research, 2016, “Response Rate Calculator V4.0.” In AAPOR (https://www.aapor.org/aapor_main/media/mainsitefiles/response-ratecalculator– 4– 0-clean– 18-may– 2016.xlsx, Date Visited: November 16, 2017).
9. Backstrom, Charles Herbert,Hursh-César, Gerald(1981).Survey Research.New York:Macmillan.
10. Kish, Leslie, 1949, “A Procedure for Objective Respondent Selection within the Household.” Journal of the American Statistical Association 44: 380– 387.
11. Lee, Sunghee(2006).Propensity Score Adjustment as a Weighting Scheme for Volunteer Panel Web Surveys.Journal of Official Statistics,22(2),329-349.
12. Richard, Jean-Baptiste,Andler, Raphaël,Gautier, Arnau,Guignard, Romain,Leon, Christophe,Beck, François(2017).Effects of Using an Overlapping Dual-frame Design on Estimates of Health Behaviors: A French General Population Telephone Survey.Journal of Survey Statistics and Methodology,5(2),254-274.
13. Salmon, Charles T.,Nichols, John Spicer(1983).The Next-birthday Method of Respondent Selection.The Public Opinion Quarterly,7(2),270-276.
14. Troldahl, Verling C.,Carter, Roy E., Jr(1964).Random Selection of Respondents within Households in Phone Surveys.Journal of Marketing Research,1(2),71-76.
15. Valliant, Richard,Dever, Jill A.(2011).Estimating Propensity Adjustments for Volunteer Web Surveys.Sociological Methods & Research,40(1),105-137.
16. 王昭正（譯）,朱瑞淵（譯）,Lavrakas, Paul(1999).電話調查方法.臺北:弘智.
17. 行政院主計處，2017，105 年家庭收支調查報告（http://win.dgbas.gov.tw/fies/doc/result/105.pdf，取用日期：2017 年 11 月 16 日）。（Directorate General of Budget, Accounting and Statistics, Executive Yuan, ROC (Taiwan), 2017, “Report on the Survey of Family Income and Expenditure, 2016.” (https://win.dgbas.gov.tw/fies/doc/result/105.pdf, Date Visited: November 16, 2017.)）
18. 杜素豪, Su-hao(2015)。比較入選機率分組與其他加權方法對電話調查樣本的調整：上網率的推估。臺灣社會學刊，56，105-150。
19. 洪永泰, Yung-Tai(2001).戶中選樣之研究.臺北=Taipei:五南=Wunan Publications.
20. 洪永泰, Yung-Tai(2015)。母體是什麼？台灣實證研究中加權與推估的困境。「2015調查研究方法與應用」學術研討會，臺北=Taipei, Taiwan:
21. 張文馨，2014，頂尖對決！洪永泰預測柯 P 得票只差 1%（http://www.storm.mg/article/25670，取用日期：2017 年 11 月 16 日）。（Chang, Wen-Hsin, 2014, “Face off! The Difference between the Prediction of Yung-Tai Hung and the Rate of Votes Wen-Je, Ko Had Is Just 1%.” (http://www.storm.mg/article/25670, Date Visited: November 16, 2017.)）
22. 楊雅惠，2016，電話民調老是不準？因為它對 39 歲以下族群涵蓋率不到六成（https://www.inside.com.tw/2016/11/15/telephone-polls，取用日期：2017 年 11 月 16 日）。Yang, Ya-Hui, 2016, “Why the Opinion Polls Were Always Wrong? The Coverage Rates of People Who Are under the Age of 39 is Less than 60%.” (https://www.inside.com.tw/2016/11/15/telephone-polls, Date Visited: November 16, 2017.)）
23. 瞿海源（編）, Hei-yan(2007).調查研究方法.臺北=Taipei:三民=Sanmin Publications.