【摘要】 目的　通过抽样设计 ,确定浙江省公共卫生监测样本地区。方法　选择社会经济、人口、教育和健康等多个指标的资料 ,采用相关分析、因子分析寻找分层标识 ,用K means聚类分析进行分层 ,采用整群随机抽样方法抽取多种组合样本 ,通过样本统计量与总体参数的比较确定样本容量。结果　全省分为 5类地区 ,城市 2类 ,分别有 17和 11个区 ;农村 3类 ,分别有 19、13、3 1个县。给定样本量进行样本统计量与总体参数比较显示 ,样本量为 2 0 ,2 5时 ,样本各指标均数与总体参数的平均误差为 1%左右 ,2 5以上的样本量 ,其精度 99%以上。样本量 3 0以上时 ,样本各指标均数的平均变异系数 5 %左右。结论　选择 3 0个县 (区 )作为全省公共卫生监测区是比较合适的更多还原

【Abstract】 Objective To select a most economic and efficient area sample for public health surveillance. Methods Data was obtained from national population census and the principle component factor was been as the stratification criterion using the principle component analysis. All units were stratified into strata by the score of principle component factor using K-means cluster analysis. Systematic random sampling was used to select a sample allowing each unit having equal probability of being selected regardless of its size. Results All units were stratified into 5 strata, 2 urban and 3 rural. Comprising sample statistics with population parameters for seven samples of 10 15 20 25 30 35 40 counties/districts, the average errors of sample statistics was less than 1%, and the average coefficient of varies of sample statistics was less than 5.78% when the sample size was up to 30. Conclusion We found that 30 sample counties/districts may be adequate to present the whole province.更多还原