敏感问题Simmons模型分层二阶段整群抽样的统计方法

目的为二分类敏感性问题提供科学的较复杂抽样调查方法及其统计量的计算公式。方法使用Simmons模型、分层两阶段整群抽样调查方法、全概率公式、方差的基本性质等理论和方法,进行了相关公式的推导;并且使用内容效度的评价方法对该统计方法进行效度评价。结果推导出二分类敏感问题Simmons模型在分层二阶段整群抽样下总体比例的估计量及其估计方差的计算公式,其效度评价的＂优良＂一致率高达100%。并成功应用该调查方法及相关公式调查得苏州大学本科生和研究生婚前性行为发生比例分别为14.71%和27.25%,总体比例为18.37%。结论敏感性问题Simmons模型的分层两阶段整群抽样调查方法及相应的统计量计算公式科学可行。

Stratified Two-stage Cluster Sampling on the Simmons Model for Sensitive Question Survey

Objective To explore scientific sampling methods and corresponding formulas for sen- sitive questions survey with the sample selected by stratified two-stage cluster sampling. Methods Stratified two-stage cluster sampling, Simmons model, total probality formulas and properties of variance were used to deduce corresponding formulas; in addition, agreement validity was applied to evaluate the statistical methods. Results Formulas for the estimation of the population proportion and its variance for dichotomous sensitive questions on Simmons model in stratified two-stage cluster sampling were de- duced, with high validity （the eoneordance rate of ＂excellence＂ and ＂goodness＂ was 100%）. The survey methods and formulas were employed successfully in a survey of premarital sexual behavior among stu- dents at Soochow University, and the estimated populated proportion was 14.71% for undergraduate, 27. 25% for graduate and 18.37% for overall, respectively. Conclusion The methods and corresponding formulas on Simmons model with the sample selected by stratified two-stage cluster sampling are feasi- ble.