Varaince Estimation Under Balanced Sampling Plans Excluding Adjacent Units

Sampling plans that exclude the selection of adjacent units within a given sample, while maintaining a constant second-order inclusion probability for non-adjacent units, have been proposed as a means of collecting information from populations where neighboring units provide similar responses. Although significant advancements have been achieved concerning the generalization and existence of such sampling plans for finite, one-dimensional populations, many other aspects of these plans warrant further investigation. Results of an investigation of three biased variance estimators of the Horvitz-Thompson estimator of the population mean under such plans are presented Key Words: variance estimation; balanced sampling plans excluding adjacent units; polygonal designs; circularly and linearly ordered populations; finite population samplin