Multiple Sets of Initial Values Method for MLE‐EM and Its Variants in Cognitive Diagnosis Models

Cognitive diagnosis models (CDMs) are widely used to assess individuals’ latent characteristics, offering detailed diagnostic insights for tailored instructional development. Maximum likelihood estimation using the expectation‐maximization algorithm (MLE‐EM) or its variants, such as the EM algorithm with monotonic constraints and Bayes modal estimation, typically uses a single set of initial values (SIV). The MLE‐EM method is sensitive to initial values, especially when dealing with non‐convex likelihood functions. This sensitivity implies that different initial values may converge to different local maximum likelihood solutions, but SIV does not guarantee a satisfactory local optimum. Thus, we introduced the multiple sets of initial values (MIV) method to reduce sensitivity to the choice of initial values. We compared MIV and SIV in terms of convergence, log‐likelihood values of the converged solutions, parameter recovery, and time consumption under varying conditions of item quality, sample size, attribute correlation, number of initial sets, and convergence settings. The results showed that MIV outperformed SIV in terms of convergence. Applying the MIV method increased the probability of obtaining solutions with higher log‐likelihood values. We provide a detailed discussion of this outcome under small sample conditions in which MIV performed worse than SIV.

• Publication year

  2025

• Venue

  Journal of Educational Measurement

• Publication date

  2025-09-01

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1111/jedm.70005
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• A One‐Parameter Diagnostic Classification Model with Familiar Measurement Properties

  2023cited by this paper
• Item Parameter Recovery: Sensitivity to Prior Distribution

  2023cited by this paper
• DINA-BAG: A Bagging Algorithm for DINA Model Parameter Estimation in Small Samples

  2023cited by this paper
• A Testlet Diagnostic Classification Model with Attribute Hierarchies

  2023cited by this paper
• Improved Harris hawks optimization for non-convex function optimization and design optimization problems

  2022cited by this paper
• A Tensor-EM Method for Large-Scale Latent Class Analysis with Binary Responses

  2021cited by this paper
• Examining cognitive diagnostic modeling in classroom assessment conditions

  2021cited by this paper
• The Reliability of the Posterior Probability of Skill Attainment in Diagnostic Classification Models

  2020cited by this paper
• Estimating Cognitive Diagnosis Models in Small Samples: Bayes Modal Estimation and Monotonic Constraints

  2020cited by this paper
• GDINA: An R Package for Cognitive Diagnosis Modeling

  2020cited by this paper
• Improved Initialization of the EM Algorithm for Mixture Model Parameter Estimation

  2020cited by this paper
• A Comparison of Software Packages Available for DINA Model Estimation

  2019cited by this paper
• On the Estimation of Standard Errors in Cognitive Diagnosis Models

  2018cited by this paper
• Applications of Diagnostic Classification Models: A Literature Review and Critical Commentary

  2018cited by this paper
• Prior Sensitivity Analysis in Default Bayesian Structural Equation Modeling

  2017cited by this paper
• Local Optima in Mixture Modeling

  2016cited by this paper
• An Application of M2 Statistic to Evaluate the Fit of Cognitive Diagnostic Models

  2016cited by this paper
• The R Package CDM for Cognitive Diagnosis Models

  2016cited by this paper
• Obtaining Diagnostic Classification Model Estimates Using Mplus

  2013cited by this paper
• Minimum Discrepancy Estimation in Probabilistic Knowledge Structures

  2013cited by this paper
• Evaluating the Wald Test for Item‐Level Comparison of Saturated and Reduced Models in Cognitive Diagnosis

  2013cited by this paper
• Theory and Use of the EM Algorithm

  2011cited by this paper
• On the Analysis of Fraction Subtraction Data: The DINA Model, Classification, Latent Class Sizes, and the Q-Matrix

  2011cited by this paper
• DINA Model and Parameter Estimation: A Didactic

  2009cited by this paper
• Performance of Factor Mixture Models as a Function of Model Size, Covariate Effects, and Class-Specific Parameters

  2007cited by this paper
• Boosting Algorithms: Regularization, Prediction and Model Fitting. Rejoinder.

  2007cited by this paper
• Choosing initial values for the EM algorithm for finite mixtures

  2003cited by this paper
• Global Optimization

  1999cited by this paper
• Maximum likelihood estimation of a mixture of normal regressions: starting values and singularities

  1998cited by this paper
• A Probabilistic Model for Diagnosing Misconceptions By The Pattern Classification Approach

  1985cited by this paper

• No citing papers are available for this paper.