随着社会经济的发展,人们越来越多地使用调查数据,尤其是利用复杂调查数据进行研究。复杂调查数据源于复杂抽样设计,是指在抽样调查过程中使用分层、多阶段、整群和不等概率等复杂抽样设计获得调查数据。在数据分析中,若忽视层、群等抽样设计的复杂性,直接利用调查数据按照传统数据分析方法,容易得出错误的结论,尤其是涉及标准误的估计。本文主要介绍复杂抽样设计和复杂调查数据的特征,对抽样信息在复杂抽样数据中的应用进行研究,并以中国家庭追踪调查(China Family Panel Studies,CFPS)为例详细说明抽样信息在复杂调查中的应用,说明在复杂抽样调查数据中使用抽样信息的重要性。
With the development of society and economy, more and more people do some research using survey data especially the complex survey data. Complex survey data refers to sample design in which samples have been sampled in a way that is muhi-stage, stratified, clustered and unequal probability sampling design. In data analysis, it will get the wrong conclusions if use the traditional data analysis method and neglect the complex sampling design, especially for the standard error estimators. This paper mainly introduces the characters in complex sampling design and complex survey data and studies the application of sampling information in complex sample survey data. Basing on the data of China family panel studies, it shows how to use the information in complex survey data and explains the importance of sampling information in complex sample survey data.