The total number of clusters, drawn from a list of all sections and their population estimates, is divided by 30 to obtain the cluster interval K.
从所有的区域和它们的估计人口列表中得到总分组数，再将总分组数除以30得到分组抽样间隔k。
2
PLS partitions the population into a group of disjoint subpopulations by cluster analysis, then USES simplex method to search local optima in each subpopulation in parallel.
Methods Subjects were collected from community population by stratified multi-stage cluster sampling method. The data gathered from case investigation and serological detection.