On ranked set sampling variation and its applications to public health research

The foundation of any statistical inference depends on the collection of required data through some formal mechanism that should be able to capture the distinct characteristics of the population. One of the most common mechanisms to obtain such data is the simple random sample (SRS). In practice, a more structured sampling mechanism, such as stratified sampling, cluster sampling or systematic sampling, may be obtained to achieve a representative sample of the population of interest. A cost effective alternative approach to the aforementioned sampling techniques is the ranked set sampling (RSS). This approach to data collection was first proposed by McIntyre (Aust. J. Agr. Res. 3:385-390, 1952) as a method to improve the precision of estimated pasture yield. In RSS the desired information is obtained from a small fraction of the available units.