重点阐述其设计要点，以及如何利用多重插补方法对缺失数据进行处理。
The topics enclose the important phases of designing a split questionnaire, and the methods of using the multiple imputation method to deal with the missing data.
本文介绍的插补方法有：演绎估计，均值插补，随机插补，回归插补和多重插补。
This paper introduces several imputation methods: those methods include: Deductive imputation, Mean value imputation, Randomized imputation Regression method and Multiple imputation.
重点讨论了单一插补的方差估计与多重插补的简化计算以及使用回答概率的单一插补等。
Then, variance estimates of single imputation, simplified calculation of multiple imputation and imputation using response probability are studied.
介绍分层随机抽样条件下多重插补法处理缺失数据的基本思想，分析可忽略无回答的分层随机抽样建立多重插补的常用方法，并通过实例加以说明。
The paper introduces multiple imputation (mi) for missing data in stratified random sampling, and discusses the ordinary method of mi with ignorable nonresponse, and illustrates the essential steps.
介绍分层随机抽样条件下多重插补法处理缺失数据的基本思想，分析可忽略无回答的分层随机抽样建立多重插补的常用方法，并通过实例加以说明。
The paper introduces multiple imputation (mi) for missing data in stratified random sampling, and discusses the ordinary method of mi with ignorable nonresponse, and illustrates the essential steps.
应用推荐