Jan Netík

I am a PhD fellow at the Department of Statistical Modelling of the Institute of Computer Science of the Czech Academy of Sciences, a member of the Computational Psychometrics Group, and a doctoral student at the Faculty of Education, Charles University. At Schola Empirica, an applied-research non-profit, I also help design and analyze evaluations of educational programs.

My research sits at the intersection of psychometrics and natural language processing. I study how the difficulty of test items can be modeled directly from their wording using fine-tuned transformer encoders, and I develop item response theory methods and interactive tools that make psychometric analysis accessible to test developers.

Much of my work is open source: I co-develop the {ShinyItemAnalysis} ecosystem, including the {SIAtools} developer toolkit, and love contributing to projects such as 🤗 transformers, {mirt}, and the {tidyverse}, in both R and Python.