[1]康春花,孙金玲,孙小坚,等.缺失数据比率和处理方法对非随机缺失数据能力参数估计准确性的影响[J].江西师范大学学报(自然科学版),2017,(03):302-307.
 KANG Chunhua,SUN Jinling,SUN Xiaojian,et al.The Effects of Missing not at Random Data to the Accuracy of Ability Parameter Estimation in IRT[J].Journal of Jiangxi Normal University:Natural Science Edition,2017,(03):302-307.
点击复制

缺失数据比率和处理方法对非随机缺失数据能力参数估计准确性的影响()
分享到:

《江西师范大学学报》(自然科学版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2017年03期
页码:
302-307
栏目:
出版日期:
2017-05-01

文章信息/Info

Title:
The Effects of Missing not at Random Data to the Accuracy of Ability Parameter Estimation in IRT
作者:
康春花孙金玲孙小坚曾平飞
浙江师范大学教师教育学院,浙江 金华 321004
Author(s):
KANG ChunhuaSUN JinlingSUN XiaojianZENG Pingfei
College of Teacher Education,Zhejiang Normal University,Jinhua Zhejiang 321004,China
关键词:
缺失数据比例 缺失数据处理方法 IRT参数估计
Keywords:
the proportion of missing data methods to deal with missing data parameter estimation in IRT
分类号:
B 841.7
文献标志码:
A
摘要:
探讨了IRT背景下非随机缺失数据的合适处理方法.采用IRTLAB模拟产生50批500个被试在20个0-1记分项目上的反应数据,产生了不同比率的MNAR; 再用IN、NP、FR、CM、MI和EM共6种方法分别处理MNAR,使用BILOG-MG软件估计被试的能力参数,并计算在不同条件下各种方法的BIAS、BIASabs、R(θ,θ)和RMSE.研究发现:随着缺失比率的增加,参数误差越来越大; FR会导致IRT参数估计产生较大的误差,且不稳定,而MI与EM算法则相对稳定; 综合BIAS和RMAE等几个指标,NP在处理MNAR时产生的误差较小也更稳定.因此,在IRT背景下估计被试能力参数时,应选择NP、MI或EM方法处理缺失数据
Abstract:
A simulation study is conducted to explore proper methods of handling missing not at random data in IRT context.First,generate response data of 500 subjects on 20 items,every item is scored by 0 or 1.Then,different percent of missing data were simulated,next,6 kinds of methods were used to deal with the missing data.Estimate the subjects’ ability via the BILOG-MG software,comparing different methods with the following four criterias:BIAS、BIASabs and RMSEabs and is shown that these methods exhibit varying degrees of effectiveness in dealing with MNAR.It is advisable for us to us the NP,MI,EM methods to handle with MNAR in IRT context.

参考文献/References:

[1] 游晓锋,丁树良,刘红云.缺失数据的估计方法及应用[J].江西师范大学学报:自然科学版,2011,35(3):325-330.
[2] Holman R,Glas C A W.Modeling non-ignorable missing-data mechanisms with item response theory models [J].The Brithish Psychological Society,2005,58(1):1-17.
[3] Little R J A,Rubin D B.Statistical analysis with missing data [M].Hoboken,NJ:John Wiley and Sons,Inc,2002.
[4] Peugh J L,Enders C K.Missing data in educational research:a review of reporting practices and suggestion for improvement [J].Review of Educational Research,2004,74(4):525-556.
[5] Roth P L.Missing data:a conceptual review for applied psychologists [J].Personal Psychology,1994,47(3):537-560.
[6] Rose N,von Davier M,Xu Xueli.Modeling nonignorable missing data with item response theory [EB/OL].
[2015-12-11].http://www.ets.org/Media/Research/pdf/RR-10-11.pdf.
[7] Newman D A.Missing data:five practical guidelines [J].Organizational Research Methods,2014,17(4):372-411.
[8] Kristin L,Sainani.Dealing with missing data and suggestion for improvement [J].Review of Educational Research,2004,74(4):525-556.
[9] 李斌,李晓毅,付志慧.IRT框架下不可忽视缺失数据的Bayes估计 [J].沈阳师范大学学报:自然科学版,2015,33(2):216-220.
[10] Pohl S,Gr?fe L,Rose N.Dealing with omitted and not-reached items incompetence tests:evaluating approaches accounting for missing responses in item response theory models [J].Educational and Psychological Measurement,2014,74(3):423-452.
[11] K?hler C,Pohl S,Carstensen C H.Taking the missing propensity into account when estimating competence scores:evaluation of item response theory models for nonignorable omissions [J].Educational and Psychological Measurement,2015,75(5):850-874.
[12] Finch H.Estimation of item theory parameters in the presence of missing data [J].Journalof Educational Measurement,2008,45(3):225-245.
[13] 张淑梅,辛涛,曾莉,等.2PL模型的EM缺失数据处理方法研究 [J].应用概率统计,2011,27(3):241-255.
[14] 汪金晖,张淑梅,辛涛.缺失数据下等级反应模型参数MCMC估计 [J].北京师范大学学报:自然科学版,2011,47(3):229-234.
[15] LordF M.Maximum likelihood estimation of item response parameters when some responses are omitted [J].Psychometrika,1983,48(3):477-482.
[16] Rubin,Donald B.Multiple imputation for nonresponse in surveys [M].New York:John Wiley Press,1987.
[17] Kadengye D T,Ceulemans E,Noortgate W V D.Direct likelihood analysis and multiple imputation for missing item scores in multilevel cross-classifacation educational data [J].Applied Psychological measurement,2013,31(8):61-80.
[18] Margot Peeters.How to handle missing data:a comparison of different approaches [J].European Journal of Developmental Psychology,2015,12(4):377-394.
[19] Schafer J L.Analysis of incomplete multivariate data [M].Wiley-VCH:Godin Lyttle Press,1997.

备注/Memo

备注/Memo:
收稿日期:2016-09-22基金项目:浙江省自然科学基金(LY15C090003)和教育部人文社会科学基金(16YJA190002)资助项目.通信作者:曾平飞(1963-),男,广西荔浦人,教授,博士,主要从事心理测量与评价方面的研究.E-mail:zpf@zjnu.edu.cn
更新日期/Last Update: 1900-01-01