Abstract:Multivariate Item Response Theory (MIRT) is sought-after widely by applied researchers looking for interpretable (sparse) explanations underlying response patterns in questionnaire data. There is, however, an unmet demand for such sparsity discovery tools in practice. Our paper develops a Bayesian platform for binary and ordinal item MIRT which requires minimal tuning and scales well on relatively large datasets due to its parallelizable features. Bayesian methodology for MIRT models has traditionally relied on MCMC simulation, which cannot only be slow in practice, but also often renders exact sparsity recovery impossible without additional thresholding. In this work, we develop a scalable Bayesian EM algorithm to estimate sparse factor loadings from binary and ordinal item responses. We address the seemingly insurmountable problem of unknown latent factor dimensionality with tools from Bayesian nonparametrics which enable estimating the number of factors. Rotations to sparsity through parameter expansion further enhance convergence and interpretability without identifiability constraints. In our simulation study, we show that our method reliably recovers both the factor dimensionality as well as the latent structure on high-dimensional synthetic data even for small samples. We demonstrate the practical usefulness of our approach on two datasets: an educational item response dataset and a quality-of-life measurement dataset. Both demonstrations show that our tool yields interpretable estimates, facilitating interesting discoveries that might otherwise go unnoticed under a pure confirmatory factor analysis setting. We provide an easy-to-use software which is a useful new addition to the MIRT toolkit and which will hopefully serve as the go-to method for practitioners.

Latent Variable Selection for Multidimensional Item Response Theory Models Via L-1 Regularization

Latent Variable Selection for Multidimensional Item Response Theory Models via $$L_{1}$$ Regularization

Accelerating L1-penalized expectation maximization algorithm for latent variable selection in multidimensional two-parameter logistic models

Joint Maximum Likelihood Estimation for High-dimensional Exploratory Item Response Analysis

Joint Maximum Likelihood Estimation for High-Dimensional Exploratory Item Factor Analysis

Applying Unidimensional and Multidimensional Item Response Theory Models in Testlet-Based Reading Assessment

A Restricted Latent Class Model with Polytomous Attributes and Respondent-Level Covariates

A Fused Latent and Graphical Model for Multivariate Binary Data

Latent variable selection in structural equation models

A Latent Trait Model for Multivariate Longitudinal Data With Two Sources of Measurement Error

Robust Measurement Via A Fused Latent and Graphical Item Response Theory Model

Multidimensional item response theory models for testlet-based doubly bounded data

Bayesian Nonparametric Monotone Regression of Dynamic Latent Traits in Item Response Theory Models

Item-Weighted Likelihood Method for Measuring Growth in Longitudinal Study With Tests Composed of Both Dichotomous and Polytomous Items

A Doubly Latent Space Joint Model for Local Item and Person Dependence in the Analysis of Item Response Data

Sparse Bayesian Multidimensional Item Response Theory

The Improved EMS Algorithm for Latent Variable Selection in M3PL Model

Regularized Variational Estimation for Exploratory Item Factor Analysis

Multidimensional Item Response Theory in the Style of Collaborative Filtering

Mapping unobserved item-respondent interactions: A latent space item response model with interaction map

Generalized Linear Latent Variable Modeling for Multi-Group Studies