人口普查质量评估中的比对误差研究
发布时间：2019-05-27 15:16
【摘要】：针对人口普查质量评估中对普查登记人口名单及其质量评估调查人口名单进行比对,以找出二者匹配人数构造双系统估计量或单系统估计量可能出现的比对误差问题,提出测度比对误差及对人口普查覆盖误差估计影响的研究目标。为实现上述目标,采用文献解读法,以及吸纳国际相关前沿研究成果的研究思路,研究比对误差的总差异率、净差异率和相对误差率等问题。研究认为:内含比对误差的人口普查覆盖误差估计值不能真实反映人口普查登记工作的质量,影响普查数据用户的决策;欲测度比对误差对人口普查质量评估结论的影响,可从人口普查质量评估调查样本中抽取子样本再次比对(评估比对),然后比较评估比对与原始比对结果的差异;欲测度比对误差对人口普查净误差的影响,可分别计算原始比对和评估比对的双系统估计量,然后求二者之差;欲测度比对误差对人口普查错误登记人数(或遗漏人数)的影响,可分别构造原始比对和评估比对的错误登记人数(或遗漏人数)估计量,然后求二者之差;在中国开展比对误差测量工作势在必行,急需进行比对工作的基础建设。
[Abstract]:In view of the comparison between the census registered population list and its quality assessment survey population list in the census quality assessment, in order to find out the possible comparison error problem between the two matching numbers, the double system estimation or the single system estimation, The research objectives of the influence of measure comparison on error and population census coverage error estimation are put forward. In order to achieve the above objectives, the literature interpretation method and the research ideas of absorbing the relevant international research results are used to study the total difference rate, net difference rate and relative error rate of the comparison error. It is considered that the estimated value of the coverage error of the census with the error can not truly reflect the quality of the registration work of the census and affect the decision of the users of the census data. In order to measure the influence of error on the conclusion of census quality evaluation, we can take sub-samples from the survey samples of census quality assessment and compare them again (evaluation and comparison), and then compare the differences between the evaluation comparison and the original comparison results. In order to measure the influence of the error on the net error of the census, the two-system Estimators of the original comparison and the evaluation comparison can be calculated respectively, and then the difference between the two can be obtained. In order to measure the influence of error on the number of false registration (or missing number) in census, the estimation of error registration number (or missing number) of original comparison and evaluation comparison can be constructed respectively, and then the difference between the two can be obtained. It is imperative to carry out comparison error measurement in China, and it is urgent to carry out the basic construction of comparison work.
【作者单位】： 重庆工商大学数学与统计学院;明尼苏达大学人口中心;贺州学院经济与管理学院;
【基金】：重庆市社会科学规划重大委托项目《人口普查质量评估前沿理论研究》(2016WT03) 国家社会科学基金一般项目《人口普查净误差估计中的三系统估计量研究》(15BTJ011) 全国统计科学研究重点项目《我国户籍登记系统覆盖评估研究》(2016LZ29) 重庆高校人文社会科学研究重点项目《人口普查多报及漏报估计研究》(17SKG088) 美国健康研究院及人口统计和行为科学项目《亚欧人口普查样本数据整合》(R01HD047283)
【分类号】：C921.2
，
本文编号：2486253
[Abstract]:In view of the comparison between the census registered population list and its quality assessment survey population list in the census quality assessment, in order to find out the possible comparison error problem between the two matching numbers, the double system estimation or the single system estimation, The research objectives of the influence of measure comparison on error and population census coverage error estimation are put forward. In order to achieve the above objectives, the literature interpretation method and the research ideas of absorbing the relevant international research results are used to study the total difference rate, net difference rate and relative error rate of the comparison error. It is considered that the estimated value of the coverage error of the census with the error can not truly reflect the quality of the registration work of the census and affect the decision of the users of the census data. In order to measure the influence of error on the conclusion of census quality evaluation, we can take sub-samples from the survey samples of census quality assessment and compare them again (evaluation and comparison), and then compare the differences between the evaluation comparison and the original comparison results. In order to measure the influence of the error on the net error of the census, the two-system Estimators of the original comparison and the evaluation comparison can be calculated respectively, and then the difference between the two can be obtained. In order to measure the influence of error on the number of false registration (or missing number) in census, the estimation of error registration number (or missing number) of original comparison and evaluation comparison can be constructed respectively, and then the difference between the two can be obtained. It is imperative to carry out comparison error measurement in China, and it is urgent to carry out the basic construction of comparison work.
【作者单位】： 重庆工商大学数学与统计学院;明尼苏达大学人口中心;贺州学院经济与管理学院;
【基金】：重庆市社会科学规划重大委托项目《人口普查质量评估前沿理论研究》(2016WT03) 国家社会科学基金一般项目《人口普查净误差估计中的三系统估计量研究》(15BTJ011) 全国统计科学研究重点项目《我国户籍登记系统覆盖评估研究》(2016LZ29) 重庆高校人文社会科学研究重点项目《人口普查多报及漏报估计研究》(17SKG088) 美国健康研究院及人口统计和行为科学项目《亚欧人口普查样本数据整合》(R01HD047283)
【分类号】：C921.2
，
本文编号：2486253
本文链接：https://www.wllwen.com/shekelunwen/renkou/2486253.html
最近更新
教材专著
