A randomized response model based on blank card and geometric distribution to estimate sensitive proportion

Abstract

Using geometric probability (GP) distribution as a randomization device, a new randomized response (RR) model has been proposed which has an immense potential to estimate the human population proportion that possess a stigmatized character. Privacy protection measure of proposed model and some of its properties have been investigated. In addition, empirical experiments are conducted to validate the theoretical results, which demonstrate the better performance of the suggested estimators over their direct competitors. Finally, results are analyzed and appropriate suggestions are made available to survey practitioners when dealing with sensitive aspects.