非抽样误差视角下普查涵盖误差及其模型研究  

Census Coverage Error and Model Study from the Perspective of Non-sampling Error

作　　者：陶然[1] 

机构地区：[1]国家统计局统计科学研究所

出　　处：《统计研究》2012年第12期81-87,共7页Statistical Research

基　　金：全国统计科学研究计划重点项目"基于普查涵盖误差测量技术的基本单位名录库维护与更新研究"(2011LX003);国家社科基金项目"普查数据质量的事后抽查理论及其应用研究"(11BTJ009);中国人民大学研究生科学研究基金重点项目(中央高校基本科研业务费专项资金资助)"经济普查数据质量评估与控制研究"(10XNG049)资助。

摘　　要：本文根据普查数据生成过程,将实际普查汇总结果与目标总体真值的净误差定义为普查涵盖误差;从非抽样误差的作用分析,提出涵盖误差来源影响的三个假设,并论证采用净误差表现普查涵盖误差的合理性。在此基础上,将涵盖误差的产生机制和普查数据汇总模型结合,构建不同普查类型下计数与内容涵盖误差的模型与误差分解过程;以此论述非抽样误差对涵盖误差的影响作用,以及计数涵盖误差和内容涵盖误差间的联系,为进一步研究普查数据质量评估与控制奠定理论基础。Based on aggregation process of census data, census coverage error is the net error between the actual census aggregated results and the true value of the target population. Through analyzing the role of non-sampling error, three errors theoretical assumptions are put out and further studies on the reasonableness of using coverage error to express the aggregated census data errors are done. On these basis, combining the generation mechanism of coverage error with the census data aggregated model, the paper studies how to establish the count and content coverage error models with the different census types, discusses the manifestation impact of non-sampling error on the census coverage error and the relationship between the count and content coverage errors and lays the theoretical foundation for the further study of census data quality assessment and control.

关 键 词：非抽样误差 普查数据汇总 普查涵盖误差模型 

分 类 号：C811[社会学—统计学]