An Optimum Stratification For Stratified Cluster Sampling Design When Clusters Are of Varying Sizes

Abstract-The paper considers the problem of determining optimum strata boundaries for cluster sampling design considering unequal sizes of clusters. The minimal equations giving optimum strata boundaries by minimising the variance of the estimator of the population mean. sampling in each stratum being carried out independently by simple random sampling without replacement SRSWOR. These minimal equations are difficult to solve exactly. Thus the approximate solutions to these minimal equations have been obtained for three allocation methods namely proportional equal and neyman allocation. The paper concludes with Numerical illustrations.