Estimation of Population Variance from Multi-ranker Ranked Set Sampling Designs

In this article, we develop an estimator for a population variance based on a multi-ranker ranked set sampling design. In a multi-ranker design, the units are ranked by more than one ranker allowing ties whenever the confidence level of the rankers is low. The ranking information of all rankers is then combined in a meaningful way to create a single measure. This measure is used to construct the sampling design and a new estimator for the population variance. The article investigates the bias and relative efficiency of the proposed variance estimator. It is shown that the new estimator performs as good as or better than its competitors in the literature.