Studying the Quality of Environment Variables Using a Randomized Response Procedure for the Estimation of a Proportion Through Ranked Set Sampling

Abstract

Abstract Ranked set sampling is considered as a design for selecting a simple from a finite population. Generally it performs better than simple random sampling. Chen et al. (2006) proposed a procedure for ranking categorical variables. This paper is concerned with the study of a sensitive variable. The randomized responses procedure due to Warner (1965) is used for eliminating answer biases. The variance of the proposed estimator is derived and it is larger than the proposal of Chen et al. (2006). A study of the quality of the environment is developed. Direct responses and randomized response estimators are compared. Ranking procedures used information provided by measurements of the contaminants in the air and water.