Ranked Set Sampling

Ranked set sampling (RSS) gives a new approach to dealing with sample selection. It was proposed
in the seminal paper of McIntyre (1952. A method for unbiased selective sampling using ranked
sets. Australian Journal of Agricultural Research 3, 385À390). His experience in agricultural appli-
cation provoked a challenge to the usual simple random sampling (SRS) design introducing a previ-
ous ordering of the units. The practical studies suggested that it produces more accurate estimators
of the mean. This proposal was taken into account by other practitioners dealing with agricultural
studies. They also obtained better results using RSS. The mathematical validity of the claim was
sustained by the work of Takahasi and Wakimoto (1968. On unbiased estimates of the population
mean based on the sample stratified by means of ordering. Annals of the Institute of Statistical
Mathematics 20, 1À31).
That fact also remained unnoticed by the majority of the statistical community but some inter-
esting results were developed for establishing the mathematical reasons sustaining having better
results when using RSS.