Reference
Roberts, J. S., & Thompson, V. M. (2011). Marginal Maximum A Posteriori Item Parameter Estimation for the Generalized Graded Unfolding Model. Applied Psychological Measurement, 35(4), 259-279. https://doi.org/10.1177/0146621610392565
Review
In their study, Roberts and Thompson (2011) implemented a marginal maximum a posteriori (MMAP) procedure to estimate item parameters in the generalized graded unfolding model (GGUM). The authors compared the MMAP method's performance with marginal maximum likelihood (MML) and Markov chain Monte Carlo (MCMC) procedures. They conducted a recovery simulation that manipulated sample size, questionnaire length, and the number of item response categories.
Roberts and Thompson (2011) found that MMAP item parameter estimates were generally the most accurate and had the smallest standard errors on average. MML estimates suffered considerably in accuracy and variability when the number of item response categories was small, and the true item locations were extreme. Additionally, the MMAP estimates were more computationally efficient than the corresponding MCMC estimates.
Based on their findings, Roberts and Thompson (2011) recommended the MMAP procedure for estimating GGUM item parameters. The study provides valuable insights into the advantages of using the MMAP method for parameter estimation in psychological measurement, highlighting its improved accuracy, reduced variability, and computational efficiency compared to alternative methods.
