psiz

psiz is a Python package designed for building and analyzing psychometric models. It provides tools for implementing various item response theory (IRT) models, such as the Rasch model, the two-parameter logistic (2PL) model, and the three-parameter logistic (3PL) model. The package aims to facilitate the process of psychometric analysis by offering a flexible and user-friendly interface.

Key features of psiz include the ability to estimate item and person parameters, perform model diagnostics, and generate visualizations of model results. It supports both standard IRT models and more complex extensions, allowing researchers to tailor analyses to specific research questions. The package is built on top of scientific Python libraries like NumPy, SciPy, and TensorFlow, leveraging their computational power for efficient model fitting.

psiz is particularly useful for researchers in fields such as education, psychology, and healthcare who conduct assessments and need to understand the properties of tests and the abilities of individuals. It enables the creation of adaptive testing systems, the identification of differential item functioning (DIF), and the evaluation of measurement invariance across different groups. The package encourages reproducible research by providing a structured framework for psychometric modeling.