Abstract

In this paper, we propose a model for estimating population proportion of a personality possessing two related sensitive attributes in survey sampling by extending Warner’s Randomized Response Technique (RRT). The proposed estimators are more efficient than Lee’s simple model estimators as the population proportion escalates. Our proposed model performs better than Lee’s crossed model estimators when applied in a survey carried out on University students in Southwest Nigeria in a research to estimate the population proportion of those with the dual sensitive characters of examination misconduct and premarital sex noted as two rampant character traits among undergraduates.

Key words: Randomized response technique, two-character model estimator, examination misconduct, premarital sex.