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

Exploring the Use of Telephone Surveys and Propensity Score Adjustments to Correct Web Survey Biases

10.6166/TJPS.73(81-126)

俞振華(Eric Chen-hua Yu)；涂志揚(Chih-yang Tu)

網路調查 ； 非隨機樣本 ； Webographic變數（「網路使用者特徵」變數） ； 入選機率調整法 ； Web Survey ； Non-random Samples ； Webographic Variables ； Propensity Score Adjustment

政治科學論叢

73期（2017 / 09 / 01）

81 - 125

繁體中文

雖然網路調查已在我們日常生活中隨處可見，但民眾對於網路調查的結果往往半信半疑。對於網路調查最直接的批評即為，網路調查的樣本非來自隨機抽樣程序，為非隨機樣本（Non-random sample）。因此，網路調查的結果往往無法用來推論母體特徵。本研究分析多波「臺灣選舉與民主化研究」（Taiwan's Election and Democratization Studies, TEDS）計畫中的網路及電話調查平行測試資料，利用同時置於兩種不同調查方式中的所謂「Webographic」問題（或稱「網路使用者特徵」變數），來辨識網路使用者與非使用者的差異處，並藉入選機率調整法（propensity score adjustment, PSA）調整網路調查結果。換言之，本研究將電訪調查結果視為代表「正確結果」的「對照組」（參考樣本），並透過統計模型及結合電訪資料來調整網路調查結果（觀察樣本）。總之，本研究旨在從多個角度驗證PSA結合電訪與網路調查資料的可行性，並提出此項作法的優勢與限制。

While web surveys are observed quite frequently in our day-to-day lives, scholars rarely take their results seriously. This is mainly due to the fact that the results of web surveys are the product of non-random sampling and cannot be used to draw inferences about the target populations. This paper utilizes numerous parallel waves of telephone and web survey data, from Taiwan's Election and Democratization Studies (TEDS), to explore whether or not the possible biases produced by web surveys can be corrected with a ＂reference group＂ dataset drawn from telephone surveys. Specifically, it uses the so-called ＂Webographic variables＂ and combines the two types of data to produce proper estimates for web surveys via the propensity score adjustment (PSA) method. Overall, this analysis examines the effectiveness of using data from telephone surveys and the PSA method when dealing with possible biases of web survey results. Additionally, this paper highlights some key pros and cons of using such methods and suggests various venues for future research.

1. 杜素豪、羅婉云、洪永泰(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. 洪永泰(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─方法與應用，18，9-44。
   連結：
3. 陳肇男(2001)。大型電訪之省思。調查研究，10，121-138。
   連結：
4. 廖培珊(2010)。態度量表之選項標示語：調查資料之潛藏類別分析。調查研究─方法與應用，24，91-134。
   連結：
5. 鄭天澤、楊亨利、陳麗霞，2015，《2015年台灣寬頻網路使用調查報告》，臺北：財團法人台灣網路資訊中心。
6. Terhanian, G. and J. Bremer. 2000. Confronting the Selection-bias and Learning Effects Problems Associated with Internet Research. Rochester, NY: Harris Interactive White Paper..
7. GfK, 2012. “GfK Bibliography: Articles and Presentations based on GfK's Collected Panel-Data Analysis, or Methodology.” in http://www.knowledgenetworks.com/ganp/docs/KN-Bibliography.pdf. Last update 10 April 2016.
8. Augurzky, B.,Schmidt, C. M.(2001).The Propensity Score: A Means to An End.Bonn:Institute of Labor Economics (IZA).
9. Austin, P. C.(2011).An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies.Multivariate Behavioral Research,46(3),399-424.
10. Baker, R.,Blumberg, S. J.,Brick, J. M.,Couper, M. P.,Courtright, M.,Dennis, J. M.,Dillman, D.,Frankel, M. R.,Garland, P.,Groves, R. M.,Kennedy, C.,Krosnick, J.,Lavrakas, P.J.,Lee, S.,Link, M.,Piekarski, L.,Rao, K.,Thomas, R. K.,Zahs, D.(2010).Research Synthesis: AAPOR Report on Online Panels.Public Opinion Quarterly,74(4),711-781.
11. Caliendo, M.,Kopeinig, S.(2008).Some Practical Guidance for the Implementation of Propensity Score Matching.Journal of Economic Surveys,22(1),31-72.
12. Cochran, W. G.(1968).The Effectiveness of Adjustment by Subclassification in Removing Bias in Observational Studies.Biometrics,24(2),295-313.
13. Couper, M. P.(2000).Web Surveys: A Review of Issues and Approaches.Public Opinion Quarterly,64(4),464-494.
14. Couper, M. P.,Kapetyn, A.,Schonlau, M.,Winter, J.(2007).Noncoverage and Nonresponse in an Internet Survey.Social Science Research,36(1),131-148.
15. Crump, R. K.,Hotz, V. J.,Imbens, G. W.,Mitnik, O. A.(2009).Dealing with Limited Overlap in Estimation of Average Treatment Effects.Biometrika,96(1),187-199.
16. Dever, J.,Rafferty, A.,Valliant, R.(2008).Internet Surveys: Can Statistical Adjustments Eliminate Coverage Bias?.Survey Research Methods,2,47-60.
17. Dillman, D. A.(2000).Mail and Internet Surveys: The Tailored Design Method.New York:Wiley.
18. Dillman, D. A.,Smyth, J. D.,Christian, L. M.(2009).Internet, Mail, and Mixed-Mode.Hobiken:John Wisley & Sons.
19. Drake, C.(1993).Effects of Misspecification of the Propensity Score on Estimators of Treatment Effect.Biometrics,49(4),1231-1236.
20. Dutwin, D.,Buskirk, T. D.(2017).Apples to Oranges or Gala Versus Golden Delicious? Comparing Data Quality of Nonprobability Internet Samples to Low Response Rate Probability Samples.Public Opinion Quarterly,81(s1),213-249.
21. Elliot, M. R.,Valliant, R.(2017).Inference for Nonprobability Samples.Statistical Science,32(2),249-264.
22. Fricker, S.,Galesic, M.,Tourangeau, R.,Yan, T.(2005).An Experimental Comparison of Web and Telephone Surveys.Public Opinion Quarterly,69(3),370-392.
23. Hewson, C.,Yule, P.,Laurent, D.,Vogel, C.(2003).Internet Research Methods: A Practical Guide for the Social and Behavioral Sciences.London:Sage.
24. Holbrook, A. L.,Cho, Y. IK,Johnson, T.(2006).The Impact of Question and Respondent Characteristics on Comprehension and Mapping Difficulties.Public Opinion Quarterly,70(4),565-595.
25. Huppler H. K.,Louis, T. A.(2002).Propensity Score Modeling Strategies for the Causal Analysis of Observational Data.Biostatistics,3,179-193.
26. Imai, K.,King, G.,Stuart, E.(2008).Misunderstandings between Experimentalists and Observationalists about Causal Inference.Journal of the Royal Statistical Society Series A,171(2),481-502.
27. Lee, S.(2004).College Park, MD,University of Maryland.
28. Lee, S.(2006).Propensity Score Adjustment as a Weighting Scheme for Volunteer Panel Web Surveys.Journal of Official Statistics,22(2),329-349.
29. Lee, S.,Valliant, R.(2009).Estimation for Volunteer Panel Web Survey Using Propensity Score Adjustment and Calibration Adjustment.Sociological Methods & Research,37(3),319-343.
30. Loosveldt, G.,Sonck, N.(2008).An Evaluation of the Weighting Procedures for an Online Access Panel Survey.Survey Research Methods,2(2),93-105.
31. Mercer, A. W.,Kreuter, F.,Keeter, S.,Staurt, E. A.(2017).Theory and Practice in Nonprobability Surveys: Parallels between Casual Inference and Survey Inference.Public Opinion Quarterly,81(s1),250-279.
32. Moors, G.(2007).Exploring the Effect of a Middle Response Category on Response Style in Attitude Measurement.Quality & Quantity,42(6),779-794.
33. Nie, N.,Erbring, L.(2000).,Stanford:Stanford Institute for the Quantitative Study of Society.
34. Norris, P.(2001).Digital Divide: Civic Engagement, Information Poverty, and the Internet Worldwide.Cambridge, MA:Cambridge University Press.
35. Rivers, D.(2007).Sampling for Web Surveys.the 2007 Joint Statistical Meeting,Salt Lake City, UT:
36. Rosenbaum, P. R.,Rubin, D. B.(1984).Reducing Bias in Observational Studies Using Subclassification on the Propensity Score.Journal of the American Statistical Association,79(387),516-524.
37. Rubin, D. B.,Thomas, N.(1996).Matching Using Estimated Propensity Score: Relating Theory to Practice.Biometrics,52(1),254-268.
38. Schonlau, M.,Couper, M. P.(2017).Options for Conducting Web Surveys.Statistical Science,32(2),279-292.
39. Schonlau, M.,van Soest, A.,Kapteyn, A.(2007).Are 'Webographic' or Attitudinal Questions Useful for Adjusting Estimates from Web Surveys Using Propensity Scoring?.Survey Research Methods,1(3),155-163.
40. Schonlau, M.,Zapert, K.,Simon, L. P.,Sansted, K. H.,Marcus, S. M.,Adams, J.,Kan, H.,Turner, R.,Berry, S. H.(2004).A Comparison between Responses from a Propensity-Weighted Web Survey and Identical RDD Survey.Social Science Computer Review,22,128-138.
41. Schuman, H.,Presser, S.(1981).Questions and Answers in Attitude Surveys: Experiments on Question Form, Wording, and Context.New York:Academic Press.
42. Valliant, R.,Dever, J. A.(2011).Estimating Propensity Adjustments for Volunteer Web Surveys.Sociological Methods and Research,40,105-137.
43. Yeager, D. S.,Krosnick, J. A.,Chang, L.,Javitz, H. S.,Levendusky, M. S.,Simpser, A.,Wang, R.(2011).Comparing the Accuracy of RDD Telephone Surveys and Internet Surveys Conducted with Probability and Non-probability Samples.Public Opinion Quarterly,75(4),709-747.
44. 杜素豪(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，56，115-150。
45. 洪永泰(1996)。抽樣調查中樣本代表性問題。調查研究，1，7-37。
46. 陳陸輝編(2013)。民意調查新論。臺北:五南。
47. 游清鑫編(2009)。民意調查新論。臺北:五南。
48. 游清鑫、黃紀、洪永泰、蔡佳泓(2007)。,臺北:行政院研究發展考核委員會。
49. 翟本瑞(2002)。台灣網路市場調查公司的終結。網路社會通訊期刊，27

1. 陳鴻嘉,柯秀憓(2019)。眾裡尋「輕」千百度：電訪戶中選Young之調整。調查研究-方法與應用，42，7-81。
2. (2019)。雖遠似近－親子關係維繫與溝通管道之分析。傳播與社會學刊，47，55-85。