方差—协方差矩阵在认知诊断中的作用  

The Variance-covariance Matrix based Statistical Inferences in Cognitive Diagnostic Models

作　　者：吴琼琼 赵悦 刘彦楼[2] Wu Qiongqiong;Zhao Yue;Liu Yanlou(School of Psychology,Qufu Normal University,Jining 273165;Academy of Big Data for Education,Qufu Normal University,Jining 273165)

机构地区：[1]曲阜师范大学心理学院,济宁273165 [2]曲阜师范大学教育大数据研究院,济宁273165

出　　处：《心理学探新》2023年第3期262-268,共7页Psychological Exploration

摘　　要：认知诊断模型(cognitive diagnostic model,CDM)的统计检验目前主要包括模型参数的标准误估计、项目功能差异检验、项目水平模型比较、Q矩阵修正、属性层级关系探索等5个重要的研究领域。方差—协方差矩阵(信息矩阵的逆矩阵)在CDM的以上5种统计检验中具有基础和核心的作用。文章评述了方差—协方差矩阵在CDM的统计检验中的作用,梳理了以往研究者提出的信息矩阵估计方法的发展思路和脉络。最后对已有研究存在的重要问题进行讨论和展望。The variance-covariance matrix for the maximum likelihood estimates of model parameters in cognitive diagnostic models plays a key role in statistical inference.We(a)extensively analyze the applications of variance-covariance matrixin various fields of cognitive diagnosis,including the estimation of models parameter standard error,detection of differential item functioning,item-level comparison of saturated and reduced models,the Q-matrix estimation or validation and attribute hierarchy exploration,(b)provide possible explanations forprevious study resultsand suggests how to advance the application of variance-covariance matrix,(c)introduce 13 information matrix estimation methods proposed in literature,which are classified according to whether structural parameters are considered,(d)briefly comment the advantages or disadvantages of these methods.We conclude by discussing future directions for variance-covariance matrix research.

关 键 词：认知诊断模型 方差—协方差矩阵 信息矩阵 标准误 Q矩阵修正 

分 类 号：B841.2[哲学宗教—基础心理学]