Abstract

In this paper, when we want to obtain information, which is a sensitive quantitative attribute, from a stratified population, we propose a stratified conditional quantitative attribute unrelated question model capable of stratified estimation for sensitive attributes by using the quantitative attribute unrelated question model of Greenberg et al. (1971). In proposed model, only persons who answered “yes” through a randomization device composed of less sensitive attributes and forced questions are required to use the Greenberg et al.'s (1971) quantitative attribute unrelated question model. In addition, the sample allocation problem of each stratum is dealt with by considering proportional allocation and optimum one. We also propose a stratified conditional two sample quantitative attribute unrelated question model that can estimate sensitive quantitative attributes by stratification using two independent samples when there is no information about unrelated variables in proposed stratified conditional quantitative attribute unrelated question model. Lastly, the efficiency of the proposed model and the stratified quantitative attribute unrelated question model which applied the stratified sampling method to Greenberg et al.'s (1971) quantitative attribute unrelated question model are compared. And the efficiency of the conditional stratified two sample quantitative unrelated question model and the stratified two sample quantitative unrelated question model are compared.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call