Estimation of population proportion of a qualitative character using randomized response technique in stratified random sampling

ABSTRACT

This paper proposes an efficient stratified randomized response model based on Chang et al.'s (2004) model. We have obtained the variance of the proposed estimator of π_s, the proportion of the respondents in the population belonging to a sensitive group, under proportional and Neyman allocations. It is shown that the estimator based on the proposed model is more efficient than the Chang et al.'s (2004) estimator under both proportional as well as Neyman allocations, Hong et al.'s (1994) estimator and Kim and Warde's (2004) estimator. Numerical illustration and pictorial representation are given in support of the present study.

KEYWORDS:

• Population mean
• Randomized response technique
• Sensitive attribute
• Stratified random sampling
• Variance.

AMS SUBJECT CLASSIFICATION:

• 62D05

Acknowledgment

Authors are thankful to the erudite referees for their valuable suggestions regarding improvement of the paper.