Efficiency comparison of unrelated question models based on same privacy protection degree

In this study, the problem of estimating the proportion πA of peoplebearing a sensitive attribute A is considered. Three dichotomous unrelated questionmechanisms which are alternative to the well-known Simmons’ model are discussedand their performance is evaluated taking into account both efficiency and respondentprivacy protection. The variance of the estimators of πA is compared under equal levelsof confidentiality measures introduced by Lanke (1976) and Leysieffer andWarner(1976).