Generic Scrambled Response Model for Quantitative Sensitive Questions
Sensitive topics or highly personal questions are often faced in medical psychological and socio-economic survey. Warner (1965) was the first to propose an ingenious method to collectinformation on such questions without disclosing the privacy of the respondents, which is called randomized response (RR) technique. As a method for reducing evasive answer bias while estimating the proportion of people in a community bearing a sensitive attribute, has been studied extensively over the last four decades. And the literature on RR procedures also comprises works dealing with situations when the response to a sensitive question results in a quantitative variable. This paper proposes a generic scrambled response (SR) model for survey sampling with quantitative sensitive characteristics. Two classes of estimators for the mean of sensitive variable are proposed under arbitrary sampling design with and without replacement respectively. The most of existing research results about quantitative sensitive data may be considered as special cases of this generic method. Some special models of the two classes are discussed in detail.
Scrambled Response Sensitive Variable Quantitative Characteristic Privacy Protection
YAN Zaizai
Science College of Inner Mongolia university of technology, Hohhot, Inner Mongolia, P.R.China, 010051 Management College of Inner Mongolia university of technology, Hohhot, Inner Mongolia, P.R.China, 010051
国际会议
2009 International Institute of Applied Statistics Studies(2009 国际应用统计学术研讨会)
青岛
英文
1-9
2009-07-25（万方平台首次上网日期，不代表论文的发表时间）