2024/09/13 by James Sharpnack, Phoebe Mulcaire, Sharpnack, James +7
Computer Science · Decision Sciences · #62P15 #Advanced Statistical Modeling Techniques #Applications (stat.AP) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Psychometric Methodologies and Testing #Technology and Data Analysis
paper · pdf · doi:10.48550/arxiv.2409.08823
openalex publication_date 2024/09/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Item response theory (IRT) is a class of interpretable factor models that are widely used in computerized adaptive tests (CATs), such as language proficiency tests. Traditionally, these are fit using parametric mixed effects models on the probability of a test taker getting the correct answer to a test item (i.e., question). Neural net extensions of these models, such as BertIRT, require specialized architectures and parameter tuning. We propose a multistage fitting procedure that is compatible with out-of-the-box Automated Machine Learning (AutoML) tools. It is based on a Monte Carlo EM (MCEM) outer loop with a two stage inner loop, which trains a non-parametric AutoML grade model using item features followed by an item specific parametric model. This greatly accelerates the modeling workflow for scoring tests. We demonstrate its effectiveness by applying it to the Duolingo English Test, a high stakes, online English proficiency test. We show that the resulting model is typically more well calibrated, gets better predictive performance, and more accurate scores than existing methods (non-explanatory IRT models and explanatory IRT models like BERT-IRT). Along the way, we provide a brief survey of machine learning methods for calibration of item parameters for CATs.