Estimation of the variance of a finite population for cluster samples
C. J. Skinner · ePrints Soton (University of Southampton) · 1981
A finite clustered population is assumed to be derived from a superpopulation in which both cluster means and cluster variances are random effects. For this superpopulation and for equal cluster sample sizes, the best quadratic unbiased estimator of the variance of the finite population is given by a standard estimator. The variance of this estimator is given.