Michael Leon Chrzan

Data Scientist bridging practice, policy, and advanced analytics through machine learning, causal inference, and NLP to drive equity and evidence-based improvement in education and beyond.

Package: pairadigm - Python Package for Text Measurement

Author of a Python library designed to streamline the creation of high-quality, continuous measurement scales from text using Large Language Models (LLMs). The package enables researchers and practitioners to develop robust measurement instruments for analyzing text data at scale, with applications in education research, social science, and beyond.

pairadigm workflow

New resources developed using pairadigm:

  • EmoPair (Dataset): An open-source textual affect dataset that updates the EmoBank corpus. Available on GitHub

  • Reward-EmoPair (Hugging Face Model): An open-source 355M-parameter RoBERTa-large reward model optimized with a multi-task pairwise decision framework to evaluate text-based emotional intensity. Available on Hugging Face

Package: edfinr on GitHub - Education Finance Data Analysis in R

Contributor to an open-source R package supporting education finance data analysis. The package facilitates accessing, processing, and visualizing longitudinal K-12 finance data to inform policy and research. I collaborated on development, documentation, and testing to make education finance data more accessible to researchers and policymakers.