广东工业大学学报 ›› 2018, Vol. 35 ›› Issue (04): 51-55.doi: 10.12052/gdutxb.180042

• 综合研究 • 上一篇    下一篇

抽样调查中计量误差的分解及应用

罗薇1,2   

  1. 1. 广东工业大学 管理学院, 广东 广州 510520;
    2. 广东工业大学 大数据战略研究院, 广东 广州 510006
  • 收稿日期:2018-03-05 出版日期:2018-07-09 发布日期:2018-05-24
  • 作者简介:罗薇(1980-),女,讲师,博士,主要研究方向为统计调查与数据分析.
  • 基金资助:
    国家社会科学基金资助项目（17BTJ037）；广东省哲学社会科学“十二五”规划项目（GD15XGL09）

Decomposition and Application of Measurement Errors in Survey Sampling

Luo Wei1,2   

  1. 1. School of Management, Guangdong University of Technology, Guangzhou, 510520;
    2. Institute of Big Data Strategic Research, Guangdong University of Technology, Guangzhou, 510006
  • Received:2018-03-05 Online:2018-07-09 Published:2018-05-24

HTML

PDF (PC)

2747

摘要： 建立了以计量误差作为随机变量的一般性计量误差模型，从方差和计量偏差两类误差效应来对误差进行分解，进一步从方差中分离出计量方差，包括简单计量方差和相关计量方差，以及抽样方差，并得出上述误差的计算公式和估计方法.针对抽样实践中最常见的调查员误差，结合调查员分配方式，提出使得重复调查中计量结果的相关系数尽可能小是控制相关计量方差的关键，而增加子样本中调查员的数量可以有效减少调查员方差.

关键词: 计量误差, 相关计量方差, 相关系数, 调查员误差

Abstract: With measurement error modeled as the random variables, a general measurement error model is presented. It decomposes the errors into variance and measurement bias. The variance can be further decomposed into measurement variance and sampling variance. The measurement variance is composed with simple measurement variance and correlated measurement variance. The calculation formula and estimation method of the above errors are obtained. According to the most common interviewer effects in survey practice, combined with investigator assignments, it is proposed that the low correlation coefficient of measurement results in repeated surveys is the key to control the correlation measurement variance, and that increasing the sub-sample interviewers can effectively reduce investigator variance.

Key words: measurement error, correlated measurement variance, correlation coefficient, interviewer variance

中图分类号: 

  • C811
[1] SARNDL C E, SWENSSON B, WRETMAN J H. Model assisted survey sampling[M]. New York:Springer, 2003:601-614.
[2] BUZAS J S, STEFANSKI L A, TOSTESON T D. Measurement error[M]. New York:Springer, 2014:1241-1282.
[3] LOKEN E, GELMAN A. Measurement Error and the Replication Crisis[J]. Science, 2017, 355(6325):584-585.
[4] 王华, 金勇进. 统计数据准确性评估的误差效应分析方法[J]. 统计与信息论坛, 2009, 24(9):10-16.WANG H, JIN Y J. Error effects analysis approach for statistical data accuracy evaluation[J]. Statistics & Information Forum, 2009, 24(9):10-16.
[5] FERRO C A T, FRICKER T E. A bias-corrected decomposition of the brier score[J]. Quarterly Journal of the Royal Meteorological Society, 2012, 138(668):1954-1960.
[6] SCHOOT R V D, SCHMIDT P, BEUCKELAER A D, et al. Editorial:measurement invariance[J]. Front Psychol, 2015, 6:1064
[7] ROOVER D, TIMMERMAN K, MARIEKE E, et al. What's hampering measurement invariance[J]. Front Psychol, 2015, 5:604
[8] YU C, ZHANG S, FRIEDENREICH C, et al. Using repeated measures to correct correlated measurement errors through orthogonal decomposition[J]. Communication in Statistics-Theory and Methods, 2017, 46(23):11604-11611.
[9] BLATTMAN C, JAMISON J, KOROKNAY-PALICZ T, et al. Measuring the measurement error:a method to qualitatively validate survey data[J]. Journal of Development Economics, 2016, 120:99-112.
[10] ABOWD J M, STINSON M H. Estimating measurement error in annual job earnings:a comparison of survey and administrative data[J]. Review of Economics & Statistics, 2013, 95(5):1451-1467.
[11] BIEMER, PAUL P, GROVES R M, et al. Interviewer, respondent, and regional office effects on response variance:a statistical decomposition[J]. Applied Physics Letters, 2016, 86(7):074104-074104.
[12] 王克林. 调查员误差的计量模型与测定方法[J]. 统计与决策, 2009, 298(22):11-12.WANG K L. Measurement models and measurement methods for interviewer errors[J]. Statistics & Decision, 2009, 298(22):11-12.
[13] HANSON R H, MARKS E S. Influence of the interviewer on the accuracy of survey results[J]. Journal of the American Statistical Association, 1958, 53(283):635-655.
[14] ELLIOTT M R, WEST B T. "Clustering by Interviewer":a source of variance that is unaccounted for in single-stage health surveys[J]. American Journal of Epidemiology, 2015, 182(2):118-126.
[15] DIJKSTRA W. How interviewer variance can bias the results of research on interviewer effects[J]. Quality and Quantity, 1983, 17(3):179-187.
[1] 徐维超, 马如豹. 相关系数在脉冲噪声环境下的稳健性综述[J]. 广东工业大学学报, 2015, 32(3): 1-4.
[2] 徐维超. 相关系数研究综述[J]. 广东工业大学学报, 2012, 29(3): 12-17.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发