多级评分多策略认知诊断方法
|
摘要
多级评分比2级评分能提供更全面更丰富的诊断信息,更好地反映被试对知识的掌握状况.考虑到存在不同认知诊断策略对应不同属性层级关系的情形,甚至在某种评分规则下,存在同一个项目使用不同策略,对应的满分值也不同的复杂情况,该文利用项目反应理论等值的做法,处理这种复杂情况.Monte Carlo模拟实验表明:对含有多策略的多级评分反应数据,这种处理方法有助于提高认知诊断模型分类的准确性.
Cognitive diagnosis(CD) is the product of the combination of modern psychometrics and cognitive psychology.Polytomous scoring items can provide more comprehensive information,and better reflect the KS of the examinees.Taking into account the existence of different cognitive strategies corresponding to different attribute hierarchies,even under some scoring rules,there is a complex situation in which the same item uses different problem-sovling strategies maybe corresponding to different values,borrowing the equivalent approach of project response theory to deal with this complex situation.Monte Carlo simulation experiments show that in the polytomous scoring test when the response data of examinees contained multiple-strategy responses:The diagnosis results using multiple-strategy CD method pattern match ration(PMR) better than the single-strategy CD method.The method,the PMR in various slip all higher than other methods,their method the score conversion caused the lossing of information.
Cognitive diagnosis(CD) is the product of the combination of modern psychometrics and cognitive psychology.Polytomous scoring items can provide more comprehensive information,and better reflect the KS of the examinees.Taking into account the existence of different cognitive strategies corresponding to different attribute hierarchies,even under some scoring rules,there is a complex situation in which the same item uses different problem-sovling strategies maybe corresponding to different values,borrowing the equivalent approach of project response theory to deal with this complex situation.Monte Carlo simulation experiments show that in the polytomous scoring test when the response data of examinees contained multiple-strategy responses:The diagnosis results using multiple-strategy CD method pattern match ration(PMR) better than the single-strategy CD method.The method,the PMR in various slip all higher than other methods,their method the score conversion caused the lossing of information.
引文
[1] Rupp A A,Mislevy R J.Cognitive foundations of structured item response models [C].New York:Cambridge University Press,2007.
[2] 丁晓,吕娜,杨雅琳,等.工作记忆成分的年龄相关差异对算术策略运用的预测效应 [J].心理学报,2017,49(6):759-770.
[3] Geary D C,Hoard M K,Byrd-Craven J,et al.Cognitive mechanisms underlying achievement deficits in children with mathematical learning disability [J].Child Development,2007,78(4):1343-1359.
[4] Zhou Xinlin,Chen Chuansheng,Dong Qi,et al.Event-related potentials of single-digit addition,subtraction,and multiplication [J].Neuropsychologia,2006,44(12):2500-2507.
[5] de la Torre J,Douglas J.Model evaluation and multiple strategies in cognitive diagnosis:an analysis of fraction subtraction data [J].Psychometrika,2008,73(4):595-624.
[6] 涂冬波,蔡艳,戴海琦,等.一种多策略认知诊断方法:MSCD方法的开发 [J].心理学报,2012,44(11):1547-1553.
[7] 孙佳楠,张淑梅,辛涛,等.基于Q矩阵和广义距离的认知诊断方法 [J].心理学报,2011,43(9):1095-1102.
[8] 丁树良,祝玉芳,林海菁,等.Tatsuoka Q矩阵理论的修正 [J].心理学报,2009,41(2):175-181.
[9] 刘铁川.Mix-DINA模型功能开发及其与DINA、MS-DINA模型的模拟与实证比较 [D].南昌:江西师范大学,2012.
[10] 戴步云,张敏强,焦璨,等.基于CD-CAT的多策略RRUM模型及其选题方法开发 [J].心理学报,2015,47(12):1511-1519.
[11] Hartz S M.A bayesian framework for the unified model for assessing cognitive abilities:blendding theory with practicality [D].Urbana:University of Illinois at Urbana-Champaign,2002.
[12] 祝玉芳,王黎华,丁树良,等.多策略的多级评分认知诊断方法的开发 [J].江西师范大学学报:自然科学版,2015,39(4):371-376.
[13] Sun Jianan,Xin Tao,Zhang Shumei,et al.A polytomous extension of the generalized distance discriminating method [J].Applied Psychological Measurement,2013,37(7):503-521.
[14] 丁树良,罗芬,汪文义.多级评分认知诊断测验蓝图的设计:独立型和收敛型结构 [J].江西师范大学学报:自然科学版,2014,38(3):265-269.
[15] 丁树良,汪文义,罗芬.多级评分认知诊断测验蓝图的设计:根树型结构 [J].江西师范大学学报:自然科学版,2014,38(2):111-118.
[16] 田伟,辛涛.基于等级反应模型的规则空间方法 [J].心理学报,2012,44(1):249-262.
[17] 丁树良,杨淑群,汪文义.可达矩阵在认知诊断测验编制中的重要作用 [J].江西师范大学学报:自然科学版,2010,34(5):490-494.
[18] 丁树良,汪文义,杨淑群.认知诊断测验蓝图的设计 [J].心理科学,2011,34(2):258-265.
[19] Leighton J P,Gierl M J,Hunka S M.The attribute hierarchy method for cognitive assessment:avariation on Tatsuoka′s rule-space approach [J].Journal of Educational Measurement,2004,41(3):205-237.
[20] Samejima F.Estimation of latent ability using a response pattern of graded scores [J].Psychometrika,1970,35(1):139.
[21] 陈青,丁树良,朱隆尹,等.3参数等级反应模型及其参数估计 [J].江西师范大学学报:自然科学版,2010,34(2):117-122.
[22] Leighton J P,Gierl M J.Cognitive diagnostic as-sessment for education:theory and applications [J].Journal of Qingdao Technical College,2007,45(4):407-411.
[23] 丁树良,罗芬,汪文义,等.Q矩阵探微 [J].江西师范大学学报:哲学社会科学版,2017,50(1):71-79.
[2] 丁晓,吕娜,杨雅琳,等.工作记忆成分的年龄相关差异对算术策略运用的预测效应 [J].心理学报,2017,49(6):759-770.
[3] Geary D C,Hoard M K,Byrd-Craven J,et al.Cognitive mechanisms underlying achievement deficits in children with mathematical learning disability [J].Child Development,2007,78(4):1343-1359.
[4] Zhou Xinlin,Chen Chuansheng,Dong Qi,et al.Event-related potentials of single-digit addition,subtraction,and multiplication [J].Neuropsychologia,2006,44(12):2500-2507.
[5] de la Torre J,Douglas J.Model evaluation and multiple strategies in cognitive diagnosis:an analysis of fraction subtraction data [J].Psychometrika,2008,73(4):595-624.
[6] 涂冬波,蔡艳,戴海琦,等.一种多策略认知诊断方法:MSCD方法的开发 [J].心理学报,2012,44(11):1547-1553.
[7] 孙佳楠,张淑梅,辛涛,等.基于Q矩阵和广义距离的认知诊断方法 [J].心理学报,2011,43(9):1095-1102.
[8] 丁树良,祝玉芳,林海菁,等.Tatsuoka Q矩阵理论的修正 [J].心理学报,2009,41(2):175-181.
[9] 刘铁川.Mix-DINA模型功能开发及其与DINA、MS-DINA模型的模拟与实证比较 [D].南昌:江西师范大学,2012.
[10] 戴步云,张敏强,焦璨,等.基于CD-CAT的多策略RRUM模型及其选题方法开发 [J].心理学报,2015,47(12):1511-1519.
[11] Hartz S M.A bayesian framework for the unified model for assessing cognitive abilities:blendding theory with practicality [D].Urbana:University of Illinois at Urbana-Champaign,2002.
[12] 祝玉芳,王黎华,丁树良,等.多策略的多级评分认知诊断方法的开发 [J].江西师范大学学报:自然科学版,2015,39(4):371-376.
[13] Sun Jianan,Xin Tao,Zhang Shumei,et al.A polytomous extension of the generalized distance discriminating method [J].Applied Psychological Measurement,2013,37(7):503-521.
[14] 丁树良,罗芬,汪文义.多级评分认知诊断测验蓝图的设计:独立型和收敛型结构 [J].江西师范大学学报:自然科学版,2014,38(3):265-269.
[15] 丁树良,汪文义,罗芬.多级评分认知诊断测验蓝图的设计:根树型结构 [J].江西师范大学学报:自然科学版,2014,38(2):111-118.
[16] 田伟,辛涛.基于等级反应模型的规则空间方法 [J].心理学报,2012,44(1):249-262.
[17] 丁树良,杨淑群,汪文义.可达矩阵在认知诊断测验编制中的重要作用 [J].江西师范大学学报:自然科学版,2010,34(5):490-494.
[18] 丁树良,汪文义,杨淑群.认知诊断测验蓝图的设计 [J].心理科学,2011,34(2):258-265.
[19] Leighton J P,Gierl M J,Hunka S M.The attribute hierarchy method for cognitive assessment:avariation on Tatsuoka′s rule-space approach [J].Journal of Educational Measurement,2004,41(3):205-237.
[20] Samejima F.Estimation of latent ability using a response pattern of graded scores [J].Psychometrika,1970,35(1):139.
[21] 陈青,丁树良,朱隆尹,等.3参数等级反应模型及其参数估计 [J].江西师范大学学报:自然科学版,2010,34(2):117-122.
[22] Leighton J P,Gierl M J.Cognitive diagnostic as-sessment for education:theory and applications [J].Journal of Qingdao Technical College,2007,45(4):407-411.
[23] 丁树良,罗芬,汪文义,等.Q矩阵探微 [J].江西师范大学学报:哲学社会科学版,2017,50(1):71-79.