List Experiments with Measurement Error

被引:30
作者
Blair, Graeme [1 ]
Chou, Winston [2 ]
Imai, Kosuke [3 ]
机构
[1] Univ Calif Los Angeles, Polit Sci, Los Angeles, CA 90024 USA
[2] Princeton Univ, Dept Polit, Princeton, NJ 08544 USA
[3] Harvard Univ, Govt & Stat, Inst Quantitat Social Sci, 1737 Cambridge St, Cambridge, MA 02138 USA
关键词
auxiliary information; indirect questioning; item count technique; misspecification test; sensitive survey questions; unmatched count technique; STATISTICAL-ANALYSIS; DESIGN; POLITICS; SUPPORT;
D O I
10.1017/pan.2018.56
中图分类号
D0 [政治学、政治理论];
学科分类号
0302 ; 030201 ;
摘要
Measurement error threatens the validity of survey research, especially when studying sensitive questions. Although list experiments can help discourage deliberate misreporting, they may also suffer from nonstrategic measurement error due to flawed implementation and respondents' inattention. Such error runs against the assumptions of the standard maximum likelihood regression (MLreg) estimator for list experiments and can result in misleading inferences, especially when the underlying sensitive trait is rare. We address this problem by providing new tools for diagnosing and mitigating measurement error in list experiments. First, we demonstrate that the nonlinear least squares regression (NLSreg) estimator proposed in Imai (2011) is robust to nonstrategic measurement error. Second, we offer a general model misspecification test to gauge the divergence of the MLreg and NLSreg estimates. Third, we show how to model measurement error directly, proposing new estimators that preserve the statistical efficiency of MLreg while improving robustness. Last, we revisit empirical studies shown to exhibit nonstrategic measurement error, and demonstrate that our tools readily diagnose and mitigate the bias. We conclude this article with a number of practical recommendations for applied researchers. The proposed methods are implemented through an open-source software package.
引用
收藏
页码:455 / 480
页数:26
相关论文
共 25 条
[1]   List Experiment Design, Non-Strategic Respondent Error, and Item Count Technique Estimators [J].
Ahlquist, John S. .
POLITICAL ANALYSIS, 2018, 26 (01) :34-53
[2]   Alien Abduction and Voter Impersonation in the 2012 U.S. General Election: Evidence from a Survey List Experiment [J].
Ahlquist, John S. ;
Mayer, Kenneth R. ;
Jackman, Simon .
ELECTION LAW JOURNAL, 2014, 13 (04) :460-475
[3]  
[Anonymous], 1984, THESIS GEORGE WASHIN
[4]  
Aronow Peter M, 2015, J Surv Stat Methodol, V3, P43, DOI 10.1093/jssam/smu023
[5]   Design and Analysis of the Randomized Response Technique [J].
Blair, Graeme ;
Imai, Kosuke ;
Zhou, Yang-Yang .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2015, 110 (511) :1304-1319
[6]   Comparing and Combining List and Endorsement Experiments: Evidence from Afghanistan [J].
Blair, Graeme ;
Imai, Kosuke ;
Lyall, Jason .
AMERICAN JOURNAL OF POLITICAL SCIENCE, 2014, 58 (04) :1043-1063
[7]   Statistical Analysis of List Experiments [J].
Blair, Graeme ;
Imai, Kosuke .
POLITICAL ANALYSIS, 2012, 20 (01) :47-77
[8]  
Blair Graeme, 2019, REPLICATION DATA LIS, DOI [10.7910/DVN/L3GWNP, DOI 10.7910/DVN/L3GWNP]
[9]   Statistical Analysis of Endorsement Experiments: Measuring Support for Militant Groups in Pakistan [J].
Bullock, Will ;
Imai, Kosuke ;
Shapiro, Jacob N. .
POLITICAL ANALYSIS, 2011, 19 (04) :363-384
[10]  
Carroll J., 2006, MEASUREMENT ERROR NO, V2nd edn