On Mangat's improved randomized response strategy

Abstract

In this paper, an alternative to the Mangat (1994) randomized response sampling procedure for estimating the proportion of human population possessing a sensitive characteristic is suggested. The suggested estimator is unbiased and takes inadmissible values outside the range [0,1] less frequently as compared to the Mangat (1994) estimator. Two upper bounds of the exact variance formula of the suggested estimator are given.