Fitting Latent Space Item Response Models using Joint Maximum Likelihood Estimation
Fitting Latent Space Item Response Models using Joint Maximum Likeliho...
Rotate the person and item latent space parameter matrices to an echel...
Selecting the Latent Space Dimensionality using K-fold Cross-Validatio...
Simulating Data according to the Latent Space Item Response Model
In Latent Space Item Response Models, subjects and items are embedded in a multidimensional Euclidean latent space. As such, interactions among persons, items, and person-item combinations can be revealed that are unmodelled in more conventional item response theory models. This package implements the methods from Molenaar & Jeon (in press) and can be used to fit Latent Space Item Response Models to data using joint maximum likelihood estimation. The package can handle binary data, ordinal data, and data with mixed scales. The package incorporates facilities for data simulation, rotation of the latent space, and K-fold cross-validation to select the number of dimensions of the latent space.