Fitting Latent Space Item Response Models using Joint Maximum
Likelihood Estimation
Description
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.