数量特征敏感问题问卷设计  

A Questionnaire Survey Technique of Quantitative Sensitive Questions

作　　者：刘建平[1] 梁敏[1] 

机构地区：[1]暨南大学经济学院,广东广州510632

出　　处：《统计与信息论坛》2016年第4期13-16,共4页Journal of Statistics and Information

摘　　要：目前,数量特征敏感问题调查主要采用随机化策略,该策略需使用随机化装置,从而需要在现场实施。提出一种问卷设计技术,该技术用无关问题替代随机化装置,因而不需要调查者亲临现场,不受调查规模及调查单位聚散的限制,使得调查更加方便、实用、经济。给出了相应的无偏估计量,推算出估计量的方差和方差的估计量并举例说明。The randomized response technique（RRT）is a commonly used survey method when collecting quantitative sensitive information.In this technique,a randomizing device and face-to-face personal interviews are required. We propose a questionnaire survey technique in which we use an unrelated question instead of randomizing devices.Therefore,it is not limited in application to on-site and small-scale surveys and thus is more convenient,more practical and cost effective.Unbiased estimates and their variances are obtained.For completeness we present estimators for the variance of our estimator.A numerical example is also reported.

关 键 词：敏感性问题 随机化回答技术 问卷设计 

分 类 号：C811[社会学—统计学] O212.2[理学—概率论与数理统计]