選樣偏誤模型在調查研究中項目無反應問題的應用

The Use of Selection Bias Models in Item Non-Response Surveys

10.6683/TPSR.200806.12(1).147-183

盛杏湲(Shing-Yuan Sheng)；周應龍(Ying-Lung Chou)

無反應 ； 選樣偏誤模型 ； 投票對象 ； 政治功效意識 ； non-response ； selection bias ； vote choice ； political efficacy

台灣政治學刊

12卷1期（2008 / 06 / 01）

147 - 183

繁體中文

在調查研究中常會面臨項目無反應的問題，亦即某些受訪者雖然接受了訪問，但卻在部份問題上沒有提供研究者有效的分析答案，而回答不知道、無意見、很難說或甚至拒答。然而，僅以有反應的樣本來推估母體參數，可能會造成母體參數估計的偏誤。本研究藉助選樣偏誤模型，試著解答究竟無反應問題在甚麼樣的情況下會造成估計的偏誤，以及如何校正估計的偏誤。 本研究以「台灣選舉與民主化2004年總統選舉」的調查資料為分析的基礎，以投票對象、政治功效意識與統獨立場三個研究主題說明無反應的問題與選樣偏誤模型的應用。研究發現顯示：在投票對象的模型中，由於泛藍認同者有較多偏向無反應，且政黨認同又決定性地影響投票對象，因此，無反應問題導致的估計偏誤相當嚴重。同時，在政治功效意識的模型中，泛綠認同者有較多偏向無反應，且他們又傾向有比較高的政治功效意識，因此無反應問題也會造成估計值的偏誤。但是在統獨立場的模型中，影響受訪者反應與否的因素並沒有同時影響受訪者的偏統或偏獨，因此並沒有造成估計值的偏誤。此一研究結論說明：無反應問題是否造成估計偏誤，與無反應受訪者的比例多寡無關，而是決定於影響受訪者無反應的因素，是否同時影響受訪者在研究主題的答案方向；如果是，則將會造成估計的偏誤，此時，以選樣偏誤模型來校正估計值的偏誤可以發揮相當不錯的效果。

Survey researchers inevitably face the problem of item non-response; that is, some respondents provide valid answers to some questions, but provide invalid answers to others, such as ”don't know,” ”no opinion,” or ”hard to say.” However, including only those respondents who provide valid answers for each question may bias the estimates, because non-response does not occur randomly. This study applies a selection bias model to the problem of item non-response and tries to answer the following questions: in what situations might item non-response cause estimation bias, and in which ways might we correct for this bias? By employing survey data collected by Taiwan's Elections and Democratization Studies Project on the 2004 Presidential Election (TEDS 2004P), this research applies selection bias models to three topics of interest in the survey-vote choice, political efficacy, and position on the independence-unification issue. The research findings show that the vote choice model estimates may suffer from serious selection bias because non-respondents tended to be pan-blue voters and to vote for the pan-blue candidate. In the political efficacy model, non-respondents tended to be pan-green voters and also to have higher levels of political efficacy. Estimates for that model may therefore also suffer from selection bias. In the independence-unification issue model, by contrast, the factors affecting rates of item non-response were not systematically related to the respondents' positions on the issue. Therefore, the model estimates do not suffer from selection bias. The research findings demonstrate that the seriousness of selection bias is not determined by the non-response rate per se, but rather by the degree to which response rates are systematically related to particular answers to a given survey question. The selection bias model can provide a satisfactory way to correct for estimation bias introduced by item non-response.

1. 杜素豪(2004)。投票意向問題不同類型項目無反應之分析。選舉研究，11(2)，111-131。
   連結：
2. 周應龍、盛杏湲(2005)。選樣偏誤模型在選舉預測上的應用。選舉研究，12(2)，1-44。
   連結：
3. 盛杏湲(2002)。統獨議題與台灣選民的投票行為：一九九０年代的分析。選舉研究，9(1)，41-80。
   連結：
4. 黃信豪(2005)。台灣民眾政治功效意識的持續與變遷：政黨輪替前後的分析。選舉研究，12(2)，111-147。
   連結：
5. 鄭夙芬、陳陸輝(2000)。台灣地區民眾參與調查研究態度的變遷。選舉研究，7(1)，115-138。
   連結：
6. Achen, Christopher H.(1986).The Statistical Analysis of Quasi-Experiments.Berkeley, CA:The University of California Press.
7. Berinsky, Adam J.(2002).Silent Voices: Social Welfare Policy Opinions and Political Equality in America.American Journal of Political Science,46,276-287.
8. Berinsky, Adam J.(1999).The Two Faces of Public Opinion.American Journal of Political Science,43,1209-1230.
9. Brehm, John(1993).The Phantom Respondents.Ann Arbor, MI:The University of Michigan Press.
10. Dubin, Jeffrey A.,Douglas Rivers(1989).Selection Bias in Linear Regression, Logit and Probit Models.Sociological Methods and Research,18,360-390.
11. Heckman, James J.(1979).Sample Selection Bias as a Specification Error.Econometrica,47,153-161.
12. Huang, Chi,Tsai-Hsing Liao(2004).To Respond or Not to Respond? A Multilevel Analysis of Item Nonresponse in the TEDS 2003 Survey.Paper presented at the International Conference on Taiwan`s Election and Democratization Study,Taipei:
13. Long, J. Scott(1997).Regression Models for Categorical and Limited Dependent Variables.London, England:Sage.
14. Sheng, Shing-Yuan(1994).Selection Bias in Vote Choice Models.Journal of Electoral Studies,1(2),221-250.
15. Verba, Sidney(1996).The Citizens as Respondent: Sample Surveys and American Democracy.American Political Science Review,90,1-7.
16. Verba, Sidney,Kay Lehman Schlozman,Henry E. Brady(1995).Voice and Equality: Civic Voluntarism in American Public.Cambridge, MA:Harvard University Press.
17. 劉義周(1985)。調查研究中「不知道」選項問題之分析。國立政治大學學報，52，65-90。

1. 涂志揚、俞振華(2017)。選舉預測：利用全國性調查推估區域立委選情。東吳政治學報，35(3)，71-120。
2. 蕭怡靖、游清鑫(2012)。檢測台灣民眾六分類統獨立場：一個測量改進的提出。臺灣政治學刊，16(2)，65-116。
3. 謝忠安、黃東益、施佳良(2011)。台灣文官調查研究資料蒐集方法與調查品質：「訪員遞送與受訪者自填問卷」方法之探討。調查研究：方法與應用，25，141-179。