講題： A Comparison of Logistic Regression Estimates for Randomized Response Crossed Model
時間：2019 / 10 / 24（週四）下午 2:00 – 4:00
Randomized response (RR) theory is an ingenious data-collection survey mode conceived to reduce nonresponse rates and untruthful responding when sensitive topics are surveyed. In this study, I intend to propose some theoretical and empirical advances by furnishing the methodology for analyzing the factors that influence two sensitive characteristics under the RR crossed model under logistic regression model. I first provide the theoretical framework for obtaining the maximum likelihood estimates of the regression coefficients, then I carry out a simulation study to assess the performance of the estimation procedure. Finally, logistic regression analysis is illustrated by analyzing real data on the use of cannabis for recreational purposes and its legalization, as well as induced abortion and illegal immigration. Empirical results bring out a number of considerations about the effect on the estimates of the RR and direct questioning (DQ) survey modes, the “more-is-better assumption” and the validity of DQ analysis, inference about the sign and significance of the regression coefficients across data types, and the power analysis. All these considerations can contribute to the debate on whether or not the RRT is an effective survey method to reduce misreporting and improve the validity of analyses.
聯絡人：email@example.com (02)2789-8132 謝小姐