{"title":"属性分层结构下认知诊断建模中的估计方法。","authors":"Lokman Akbay, Jimmy de la Torre","doi":"10.7334/psicothema2019.182","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Although research in cognitive psychology suggests refraining from investigating cognitive skills inisolation, many cognitive diagnosis model (CDM) examples do not take hierarchical attribute structures into account. When hierarchical relationships among the attributes are not considered, CDM estimates may be biased.</p><p><strong>Method: </strong>The current study, through simulation and real data analyses, examines the impact of different MMLE-EM approaches on the item and person parameter estimates of the G-DINA, DINA and DINO models when attributes have a hierarchical structure. A number of estimation approaches that can result from modifying either the Q-matrix or prior distribution are proposed. Impact of the proposed approaches on item parameter estimation accuracy and attribute classification are investigated.</p><p><strong>Results: </strong>For the G-DINA model estimation, the Q-matrix type (i.e, explicit vs. implicit) has greater impact than structuring the prior distribution. Specifically, explicit Q-matrices result in better item parameter recovery and higher correct classification rates. In contrast, structuring the prior distribution is more influential on item and person parameter estimates for the reduced models. When prior distribution is structured, the Q-matrix type has almost no influence on item and person parameter estimates of the DINA and DINO models.</p><p><strong>Conclusion: </strong>We can conclude that the Q-matrix type has a significant impact on CDM estimation, especially when the estimating model is G-DINA.</p>","PeriodicalId":48179,"journal":{"name":"Psicothema","volume":null,"pages":null},"PeriodicalIF":3.2000,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.7334/psicothema2019.182","citationCount":"9","resultStr":"{\"title\":\"Estimation approaches in cognitive diagnosis modeling when attributes are hierarchically structured.\",\"authors\":\"Lokman Akbay, Jimmy de la Torre\",\"doi\":\"10.7334/psicothema2019.182\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Although research in cognitive psychology suggests refraining from investigating cognitive skills inisolation, many cognitive diagnosis model (CDM) examples do not take hierarchical attribute structures into account. When hierarchical relationships among the attributes are not considered, CDM estimates may be biased.</p><p><strong>Method: </strong>The current study, through simulation and real data analyses, examines the impact of different MMLE-EM approaches on the item and person parameter estimates of the G-DINA, DINA and DINO models when attributes have a hierarchical structure. A number of estimation approaches that can result from modifying either the Q-matrix or prior distribution are proposed. Impact of the proposed approaches on item parameter estimation accuracy and attribute classification are investigated.</p><p><strong>Results: </strong>For the G-DINA model estimation, the Q-matrix type (i.e, explicit vs. implicit) has greater impact than structuring the prior distribution. Specifically, explicit Q-matrices result in better item parameter recovery and higher correct classification rates. In contrast, structuring the prior distribution is more influential on item and person parameter estimates for the reduced models. When prior distribution is structured, the Q-matrix type has almost no influence on item and person parameter estimates of the DINA and DINO models.</p><p><strong>Conclusion: </strong>We can conclude that the Q-matrix type has a significant impact on CDM estimation, especially when the estimating model is G-DINA.</p>\",\"PeriodicalId\":48179,\"journal\":{\"name\":\"Psicothema\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2020-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.7334/psicothema2019.182\",\"citationCount\":\"9\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Psicothema\",\"FirstCategoryId\":\"102\",\"ListUrlMain\":\"https://doi.org/10.7334/psicothema2019.182\",\"RegionNum\":2,\"RegionCategory\":\"心理学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"PSYCHOLOGY, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Psicothema","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.7334/psicothema2019.182","RegionNum":2,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PSYCHOLOGY, MULTIDISCIPLINARY","Score":null,"Total":0}
Estimation approaches in cognitive diagnosis modeling when attributes are hierarchically structured.
Background: Although research in cognitive psychology suggests refraining from investigating cognitive skills inisolation, many cognitive diagnosis model (CDM) examples do not take hierarchical attribute structures into account. When hierarchical relationships among the attributes are not considered, CDM estimates may be biased.
Method: The current study, through simulation and real data analyses, examines the impact of different MMLE-EM approaches on the item and person parameter estimates of the G-DINA, DINA and DINO models when attributes have a hierarchical structure. A number of estimation approaches that can result from modifying either the Q-matrix or prior distribution are proposed. Impact of the proposed approaches on item parameter estimation accuracy and attribute classification are investigated.
Results: For the G-DINA model estimation, the Q-matrix type (i.e, explicit vs. implicit) has greater impact than structuring the prior distribution. Specifically, explicit Q-matrices result in better item parameter recovery and higher correct classification rates. In contrast, structuring the prior distribution is more influential on item and person parameter estimates for the reduced models. When prior distribution is structured, the Q-matrix type has almost no influence on item and person parameter estimates of the DINA and DINO models.
Conclusion: We can conclude that the Q-matrix type has a significant impact on CDM estimation, especially when the estimating model is G-DINA.
期刊介绍:
La revista Psicothema fue fundada en Asturias en 1989 y está editada conjuntamente por la Facultad y el Departamento de Psicología de la Universidad de Oviedo y el Colegio Oficial de Psicólogos del Principado de Asturias. Publica cuatro números al año. Se admiten trabajos tanto de investigación básica como aplicada, pertenecientes a cualquier ámbito de la Psicología, que previamente a su publicación son evaluados anónimamente por revisores externos. Psicothema está incluida en las bases de datos nacionales e internacionales más relevantes, entre las que cabe destacar Psychological Abstracts, Current Contents y MEDLINE/Index Medicus, entre otras. Además, figura en las listas de Factor de Impacto del Journal Citation Reports. Psicothema es una revista abierta a cualquier enfoque u orientación psicológica que venga avalada por la fuerza de los datos y los argumentos, y en la que encuentran acomodo todos los autores que sean capaces de convencer a los revisores de que sus manuscritos tienen la calidad para ser publicados. Psicothema es una revista de acceso abierto lo que significa que todo el contenido está a disposición de cualquier usuario o institución sin cargo alguno. Los usuarios pueden leer, descargar, copiar, distribuir, imprimir, buscar, o realizar enlaces a los textos completos de esta revista sin pedir permiso previo al editor o al autor, siempre y cuando la fuente original sea referenciada. Para acervos y repositorios, se prefiere que la cobertura se realice mediante enlaces a la propia web de Psicothema. Nos parece que una apuesta decidida por la calidad es el mejor modo de servir a nuestros lectores, cuyas sugerencias siempre serán bienvenidas.