DINA Model and Parameter Estimation: A Didactic

By Jimmy de la Torre. 2009

631 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Experimental and Cognitive Psychology, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/6db6f9a212028c5b3e5b74863852897c1830c5e2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Chia-Yi Chiu
2. Young-Sun Lee
3. L. DeCarlo
4. Y. Park
5. Wei Tian
6. Yanlou Liu
7. Tao Xin
8. Carolin Strobl

In Experimental and Cognitive Psychology

1. Chia-Yi Chiu
2. Wei Tian
3. Yanlou Liu
4. Tao Xin
5. Young-Sun Lee
6. Y. Park
7. L. DeCarlo
8. Achim Zeileis

In Management Science and Operations Research

1. Chia-Yi Chiu
2. Young-Sun Lee
3. Y. Park
4. L. DeCarlo
5. Wei Tian
6. Yanlou Liu
7. Tao Xin
8. Achim Zeileis

Citations in the corpus

• The Generalized DINA Model Framework. (2011, methodology): “Examples of specific CDMs include the deterministic inputs, noisy “and” gate (DINA; de la Torre, 2009b; Junker & Sijtsma, 2001), and the reduced reparametrized unified model (R-RUM; Hartz, 2002; Roussos, DiBello, Stout, Hartz, Henson, & Templin, 2007); examples of general CDMs include the log-line”
• A Cognitive Diagnostic Modeling of Attribute Mastery in Massachusetts, Minnesota, and the U.S. National Sample Using the TIMSS 2007 (2011, methodology): “Cognitive diagnostic models were developed to provide more targeted information in the form of score profiles that resolve the limitation of IRT models ( de la Torre, 2009 ).”
• On the Analysis of Fraction Subtraction Data: The DINA Model, Classification, Latent Class Sizes, and the Q-Matrix (2011, methodology): “, de la Torre & Douglas, 2004, 2008; Henson et al., 2009; Junker & Sijtsma, 2001). Templin and Hoffman (2010) have shown how to incorporate the model into software for second-generation structural equation modeling, namely Mplus (Muthén & Muthén, 1998-2007).”
• What CDM Can Tell About What Students Have Learned: An Analysis of TIMSS Eighth Grade Mathematics (2015, methodology): “Further discussion of assumptions and statistical requirements of the DINA model are described in de la Torre (2009) and Lee et al. (2011).”
• Statistical Refinement of the Q-Matrix in Cognitive Diagnosis (2013, methodology): “…use maximum likelihood estimation (MLE) procedures to estimate model parameters that are then used to assign examinees to proficiency classes (de la Torre, 2008, 2009, 2011; de la Torre & Douglas, 2008; Haertel, 1989; Hartz, Roussos, Henson, & Templin, 2005; Henson, Templin, & Willse, 2009; Junker…”
• An Application of M2 Statistic to Evaluate the Fit of Cognitive Diagnostic Models (2016, methodology): “…purposes, we used M2 and RMSEA2 statistics to investigate the performance of CDMs among the deterministic inputs, noisy ‘‘and’’ gate model (DINA; de la Torre, 2009; Junker & Sijtsma, 2001), the compensatory reparameterized unified model (C-RUM; Hartz, 2002), and LCDM regarding fraction…”
• A Note on Comparing Examinee Classification Methods for Cognitive Diagnosis Models (2011, methodology): “For example, de la Torre (2009) fit the DINA model to a subset of Tatsuoka’s (1990) fraction subtraction data and reported guess and slip parameters with means of .105 and .138, respectively, whereas Cheng (2009) reported means of .58 and .07 for guess and slip parameters, respectively, when…”
• On the Estimation of Standard Errors in Cognitive Diagnosis Models (2018, methodology): “The marginal log likelihood that is maximized to estimate q, given the data sample X ¼ fxig for individuals i ¼ 1; . . . ;N is given by lðq;XÞ ¼ log Lðq;XÞ½ ¼ log YN i¼1 XL l¼1 pl Pr Xi ¼ xij lð Þ; and can be maximized using the EM algorithm as described in de la Torre (2009).”
• Neural Cognitive Diagnosis for Intelligent Education Systems (2019, methodology): “DINA (De La Torre 2009; von Davier 2014) and IRT (Embretson and Reise 2013) were two of the most typical works, which model the result of a student answering an exercise as the interaction between the trait features of the student ( θ ) and the exercise ( β ).”