[1]汪文义,郑娟娟,宋丽红,等.认知诊断模型参数估计算法比较[J].江西师范大学学报(自然科学版),2022,(03):291-299.[doi:10.16357/j.cnki.issn1000-5862.2022.03.12]
　WANG Wenyi,ZHENG Juanjuan,SONG Lihong,et al.The Comparison of Parameter Estimation of Cognitive Diagnosis Models[J].Journal of Jiangxi Normal University:Natural Science Edition,2022,(03):291-299.[doi:10.16357/j.cnki.issn1000-5862.2022.03.12]
点击复制

认知诊断模型参数估计算法比较()
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:
2022年03期
页码:
291-299
栏目:
心理与教育测量
出版日期:
2022-05-25

文章信息/Info

Title:
The Comparison of Parameter Estimation of Cognitive Diagnosis Models
文章编号:
1000-5862(2022)03-0291-09
作者:
汪文义1; 郑娟娟1; 宋丽红2; 胡海洋1
1.江西师范大学计算机信息工程学院,江西 南昌 330022; 2.江西师范大学教育学院,江西 南昌 330022
Author(s):
WANG Wenyi1; ZHENG Juanjuan1; SONG Lihong2; HU Haiyang1
1.School of Computer and Information Engineering,Jiangxi Normal University,Nanchang Jiangxi 330022,China; 2.School of Education,Jiangxi Normal University,Nanchang Jiangxi 330022,China
关键词:
认知诊断模型;  EM算法;  MCMC算法;  参数估计
Keywords:
cognitive diagnosis model;  EM algorithm;  MCMC algorithm;  parameter estimation
分类号:
B 841.7
DOI:
10.16357/j.cnki.issn1000-5862.2022.03.12
文献标志码:
A
摘要:
该文在不同条件的组合下考查了EM算法和MCMC算法对3种常用的认知诊断模型(DINA模型、DINO模型和G-DINA模型)的参数估计返真性问题.借助项目参数或作答概率分布的偏差、均方根误差、平均绝对离差以及被试的平均属性判准率等指标,评价这2类算法的表现.模拟研究结果表明:MCMC算法更适用于低质量题目、小样本、测验短的条件,而在其他条件下EM算法的表现与MCMC算法的表现相当.
Abstract:
The performance of the EM algorithm and MCMC algorithm under different combinations of conditions is investigated.Three cognitive diagnosis models are considered,such as the DINA(deterministic inputs,noisy "and" gate),DINO(deterministic inputs,noisy "or"gate),and the generalized DINA(G-DINA)model.The bias,root mean square error,average absolute deviation and average attribute correct classification rate are used to evaluate the performance of two parameter estimation methods applicable to the three models under different conditions.The simulation results show that the MCMC algorithm performs better than the EM algorithm for the condition combination of low-quality and short test with small sample size,while these two methods performs similarly in the other condidions.

参考文献/References:

[1] 汪文义,高朋,宋丽红,等.带噪音预处理的改进探索性Q矩阵标定方法[J].江西师范大学学报(自然科学版),2020,44(2):136-141.
[2] TEMPLIN J L,HENSON R A.Measurement of psychological disorders using cognitive diagnosis models[J].Psychological Methods,2006,11(3):287-305.
[3] DE LA TORRE J.DINA model and parameter estimation:a didactic[J].Journal of Educational and Behavioral Statistics,2009,34(1):115-130.
[4] DE LA TORRE J,DOUGLAS J A.Higher-order latent trait models for cognitive diagnosis[J].Psychometrika,2004,69(3):333-353.
[5] SEN S,TERZI R.A comparison of software packages available for DINA model estimation[J].Applied Psychological Measurement,2020,44(2):150-164.
[6] DE LA TORRE J.The generalized DINA model framework[J].Psychometrika,2011,76(2):179-199.
[7] YAMAGUCHI K,OKADA K.Variational Bayes inference for the DINA model[J].Journal of Educational and Behavioral Statistics,2020,45(5):569-597.
[8] YAMAGUCHI K,OKADA K.Variational Bayes inference algorithm for the saturated diagnostic classification model[J].Psychometrika,2020,85(4):973-995.
[9] ZHAN Peida,JIAO Hong,MAN Kaiwen,et al.Using JAGS for Bayesian cognitive diagnosis modeling:a tutorial[J].Journal of Educational and Behavioral Statistics,2019,44(4):473-503.
[10] CULPEPPER S A,HUDSON A.An improved strategy for Bayesian estimation of the reduced reparameterized unified model[J].Applied Psychological Measurement,2018,42(2):99-115.
[11] WANG Juntao,SHI Ningzhong,ZHANG Xue,et al.Sequential Gibbs sampling algorithm for cognitive diagnosis models with many attributes[J].Multivariate Behavioral Research,doi:10.1080/00273171.2021.1896352.
[12] JIANG Zhehan,CARTER R.Using Hamiltonian Monte Carlo to estimate the log-linear cognitive diagnosis model via Stan[J].Behavior Research Methods,2019,51(2):651-662.
[13] FOX C,CUI Tiangang,NEUMAYER M.Randomized reduced forward models for efficient Metropolis-Hastings MCMC,with application to subsurface fluid flow and capacitance tomography[J].GEM-International Journal on Geomathematics,2020,11(1):26.
[14] HASTINGS W K.Monte Carlo sampling methods using Markov chains and their applications[J].Biometrika,1970,57(1):97-109.
[15] 汪文义,宋丽红.教育认知诊断评估:理论与技术研究[M].北京:北京师范大学出版社,2015.
[16] WANG Wenyi,SONG Lihong,CHEN Ping,et al.Attribute-level and pattern-level classification consistency and accuracy indices for cognitive diagnostic assessment[J].Journal of Educational Measurement,2015,52(4):457-476.
[17] MA Wenchao,DE LA TORRE J.GDINA:an R package for cognitive diagnosis modeling[J].Journal of Statistical Software,2020,93(14):1-26.
[18] AKBAY L,DE LA TORRE J.Estimation approaches in cognitive diagnosis modeling when attributes are hierarchically structured[J].Psicothema,2020,32(1):122-129.
[19] TU Dongbo,WANG Shiyu,CAI Yan,et al.Cognitive diagnostic models with attribute hierarchies:model estimation with a restricted Q-matrix design[J].Applied Psychological Measurement,2019,43(4):255-271.
[20] LIU Yanlou,XIN Tao,ANDERSSON B,et al.Information matrix estimation procedures for cognitive diagnostic models[J].British Journal of Mathematical and Statistical Psychology,2019,72(1):18-37.
[21] YIGIT H D,DOUGLAS J A.First-order learning models with the GDINA:estimation with the EM algorithm and applications[J].Applied Psychological Measurement,2021,45(3):143-158.
[22] LIU Yanlou,XIN Tao,JIANG Yu.Structural parameter standard error estimation method in diagnostic classification models:estimation and application[J].Multivariate Behavioral Research,doi:10.1080/00273171.2021.1919048.

相似文献/References:

[1]宋丽红,汪文义,丁树良.测验Q矩阵的修正方法及其比较研究[J].江西师范大学学报(自然科学版),2015,(06):623.
　SONG Lihong,WANG Wenyi,DING Shuliang.The Q-Matrix Validation Methods and Comparison to Three Existing Methods[J].Journal of Jiangxi Normal University:Natural Science Edition,2015,(03):623.
[2]蔡艳,涂冬波.基于属性层级关系的rRUM模型优化——模型解释力及判准率的提升视角[J].江西师范大学学报(自然科学版),2016,40(01):47.
　CAI Yan,TU Dongbo.The Revision of the Reduced RUM Based on the Attribute Hierarchy[J].Journal of Jiangxi Normal University:Natural Science Edition,2016,40(03):47.
[3]龙 兵.逐步I型区间删失下逆威布尔分布的参数估计[J].江西师范大学学报(自然科学版),2018,(05):527.[doi:10.16357/j.cnki.issn1000-5862.2018.05.16]
　LONG Bing.The Parameter Estimation of Inverse Weibull Distribution under Progressive Type-I Interval Censoring[J].Journal of Jiangxi Normal University:Natural Science Edition,2018,(03):527.[doi:10.16357/j.cnki.issn1000-5862.2018.05.16]
[4]罗照盛,杭丹丹,秦春影,等.可以处理补偿作用的认知诊断模型:CDINA模型[J].江西师范大学学报(自然科学版),2020,(05):441.[doi:10.16357/j.cnki.issn1000-5862.2020.05.01]
　LUO Zhaosheng,HANG Dandan,QIN Chunying,et al.The Cognitive Diagnostic Model Which Can Deal with Compensation and Noncompensation Effects:CDINA[J].Journal of Jiangxi Normal University:Natural Science Edition,2020,(03):441.[doi:10.16357/j.cnki.issn1000-5862.2020.05.01]

备注/Memo

备注/Memo:
收稿日期:2021-12-16
基金项目:国家自然科学基金(62067005,61967009)资助项目.
作者简介:汪文义(1983—),男,湖南衡山人,教授,博士,主要从事教育测量与信息处理的研究.E-mail:wenyiwang@jxnu.edu.cn
更新日期/Last Update: 2022-05-25