Cognitively Diagnostic Analysis Using the G-DINA Model in R

Abstract

Cognitive diagnosis models (CDMs) have increasingly been applied in education and other fields. This article provides an overview of a widely used CDM, namely, the G-DINA model, and demonstrates a hands-on example of using multiple R packages for a series of CDM analyses. This overview involves a step-by-step illustration and explanation of performing Q-matrix evaluation, CDM calibration, model fit evaluation, item diagnosticity investigation, classification reliability examination, and the result presentation and visualization. Some limitations of conducting CDM analysis in R are also discussed.

Knowledge Integration in Science Learning: Tracking Students' Knowledge Development and Skill Acquisition with Cognitive Diagnosis Models

“…As Table 4 shows, there exist 3 and 2 items with high RMSEAs in the pretest and posttest, respectively. For IDI , currently, there is no agreement regarding the value of good discrimination though a higher value corresponds to a higher discrimination power (Shi et al., 2021). In general, the IDI s in the posttest were relatively better than the IDI s in the pretest.…”

“…In this procedure, the evaluation of item quality was provided for both the pretest and posttest using the diagnosticity (de la Torre & Karelitz, 2009), the root mean square error of approximation (RMSEA), and the item discrimination index ( IDI ) for the CDM framework (Ma, 2020; Shi et al., 2021).…”

“…For item‐level diagnosticity in the CDM framework, the IDI checks the differences in success probabilities between respondents who have mastered all measured attributes and respondents who have not mastered any of the attributes (de la Torre, 2008; Lee et al., 2012; Shi et al., 2021). An item with a higher value of IDI indicates a higher discrimination power.…”

“…Additionally, in evaluating model fit, Chen, de la Torre, and Zhang (2013) reported that in the detection of the misspecified CDMs, the Bayesian information criterion (BIC; Schwarz, 1978) and to some extent, Akaike information criterion (AIC; Akaike, 1974) can be useful for the relative fit evaluation. Therefore in choosing the suitable model, we employed −2 log likelihood(−2LL), AIC, BIC, AIC3 (a variant of AIC that uses 3$$p$$ as the penalty term), AICc (the sample size adjusted AIC), and CAIC (consistent AIC) (Ravand & Robitzsch, 2018; Robitzsch et al., 2022; Shi et al., 2021).…”

Editorial of the Psych Special Issue “Computational Aspects, Statistical Algorithms and Software in Psychometrics”

“…Shi et al [19] show how to perform the analysis of a G-DINA model in the R packages GDINA, CDM, and cdmTools. The G-DINA model framework is central to the literature of cognitive diagnostic modeling.…”