Cognitive Diagnostic Models With Attribute Hierarchies

Model Estimation With a Restricted Q-Matrix Design

Dongbo Tu, Shiyu Wang, Yan Cai, Jeffrey A Douglas, Hua-hua Chang

Research output: Contribution to journalArticle

Abstract

Attribute hierarchy is a common assumption in the educational context, where the mastery of one attribute is assumed to be a prerequisite to the mastery of another one. The attribute hierarchy can be incorporated through a restricted Q matrix that implies the specified structure. The latent class–based cognitive diagnostic models (CDMs) usually do not assume a hierarchical structure among attributes, which means all profiles of attributes are possible in a population of interest. This study investigates different estimation methods to the classification accuracy for a family of CDMs when they are combined with a restricted Q-matrix design. A simulation study is used to explain the misclassification caused by an unrestricted estimation procedure. The advantages of the restricted estimation procedure utilizing attribute hierarchies for increased classification accuracy are also further illustrated through a real data analysis on a syllogistic reasoning diagnostic assessment. This research can provide guidelines for educational and psychological researchers and practitioners when they use CDMs to analyze the data with a restricted Q-matrix design and make them be aware of the potentially contaminated classification results if ignoring attribute hierarchies.

Original languageEnglish (US)
Pages (from-to)255-271
Number of pages17
JournalApplied Psychological Measurement
Volume43
Issue number4
DOIs
StatePublished - Jun 1 2019

Fingerprint

diagnostic
estimation procedure
Research Personnel
Guidelines
Psychology
data analysis
Research
Population
simulation

Keywords

  • attribute hierarchies
  • classification accuracy
  • cognitive diagnostic models
  • restricted Q matrix

ASJC Scopus subject areas

  • Social Sciences (miscellaneous)
  • Psychology (miscellaneous)

Cite this

Cognitive Diagnostic Models With Attribute Hierarchies : Model Estimation With a Restricted Q-Matrix Design. / Tu, Dongbo; Wang, Shiyu; Cai, Yan; Douglas, Jeffrey A; Chang, Hua-hua.

In: Applied Psychological Measurement, Vol. 43, No. 4, 01.06.2019, p. 255-271.

Research output: Contribution to journalArticle

@article{c2b967fe102148b7b1fae069010d54ef,
title = "Cognitive Diagnostic Models With Attribute Hierarchies: Model Estimation With a Restricted Q-Matrix Design",
abstract = "Attribute hierarchy is a common assumption in the educational context, where the mastery of one attribute is assumed to be a prerequisite to the mastery of another one. The attribute hierarchy can be incorporated through a restricted Q matrix that implies the specified structure. The latent class–based cognitive diagnostic models (CDMs) usually do not assume a hierarchical structure among attributes, which means all profiles of attributes are possible in a population of interest. This study investigates different estimation methods to the classification accuracy for a family of CDMs when they are combined with a restricted Q-matrix design. A simulation study is used to explain the misclassification caused by an unrestricted estimation procedure. The advantages of the restricted estimation procedure utilizing attribute hierarchies for increased classification accuracy are also further illustrated through a real data analysis on a syllogistic reasoning diagnostic assessment. This research can provide guidelines for educational and psychological researchers and practitioners when they use CDMs to analyze the data with a restricted Q-matrix design and make them be aware of the potentially contaminated classification results if ignoring attribute hierarchies.",
keywords = "attribute hierarchies, classification accuracy, cognitive diagnostic models, restricted Q matrix",
author = "Dongbo Tu and Shiyu Wang and Yan Cai and Douglas, {Jeffrey A} and Hua-hua Chang",
year = "2019",
month = "6",
day = "1",
doi = "10.1177/0146621618765721",
language = "English (US)",
volume = "43",
pages = "255--271",
journal = "Applied Psychological Measurement",
issn = "0146-6216",
publisher = "SAGE Publications Inc.",
number = "4",

}

TY - JOUR

T1 - Cognitive Diagnostic Models With Attribute Hierarchies

T2 - Model Estimation With a Restricted Q-Matrix Design

AU - Tu, Dongbo

AU - Wang, Shiyu

AU - Cai, Yan

AU - Douglas, Jeffrey A

AU - Chang, Hua-hua

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Attribute hierarchy is a common assumption in the educational context, where the mastery of one attribute is assumed to be a prerequisite to the mastery of another one. The attribute hierarchy can be incorporated through a restricted Q matrix that implies the specified structure. The latent class–based cognitive diagnostic models (CDMs) usually do not assume a hierarchical structure among attributes, which means all profiles of attributes are possible in a population of interest. This study investigates different estimation methods to the classification accuracy for a family of CDMs when they are combined with a restricted Q-matrix design. A simulation study is used to explain the misclassification caused by an unrestricted estimation procedure. The advantages of the restricted estimation procedure utilizing attribute hierarchies for increased classification accuracy are also further illustrated through a real data analysis on a syllogistic reasoning diagnostic assessment. This research can provide guidelines for educational and psychological researchers and practitioners when they use CDMs to analyze the data with a restricted Q-matrix design and make them be aware of the potentially contaminated classification results if ignoring attribute hierarchies.

AB - Attribute hierarchy is a common assumption in the educational context, where the mastery of one attribute is assumed to be a prerequisite to the mastery of another one. The attribute hierarchy can be incorporated through a restricted Q matrix that implies the specified structure. The latent class–based cognitive diagnostic models (CDMs) usually do not assume a hierarchical structure among attributes, which means all profiles of attributes are possible in a population of interest. This study investigates different estimation methods to the classification accuracy for a family of CDMs when they are combined with a restricted Q-matrix design. A simulation study is used to explain the misclassification caused by an unrestricted estimation procedure. The advantages of the restricted estimation procedure utilizing attribute hierarchies for increased classification accuracy are also further illustrated through a real data analysis on a syllogistic reasoning diagnostic assessment. This research can provide guidelines for educational and psychological researchers and practitioners when they use CDMs to analyze the data with a restricted Q-matrix design and make them be aware of the potentially contaminated classification results if ignoring attribute hierarchies.

KW - attribute hierarchies

KW - classification accuracy

KW - cognitive diagnostic models

KW - restricted Q matrix

UR - http://www.scopus.com/inward/record.url?scp=85045683653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045683653&partnerID=8YFLogxK

U2 - 10.1177/0146621618765721

DO - 10.1177/0146621618765721

M3 - Article

VL - 43

SP - 255

EP - 271

JO - Applied Psychological Measurement

JF - Applied Psychological Measurement

SN - 0146-6216

IS - 4

ER -