Estimation of population mean for a sensitive variable in the presence of measurement error

Abstract

Survey researchers are well aware that non-sampling errors are likely to occur if individuals are asked to reveal information about personal or socially undesirable behavior. Getting accurate responses to the so-called sensitive questions is an age-old problem in survey research. The idea of randomized response technique (RRT) was given by Warner (1965), to provide privacy to respondents in such surveys. In this paper, a generalized estimator of population mean for sensitive study variable is proposed. A simulation study is conducted to assess the performance of the proposed estimator in the presence and absence of measurement error.