This seminar is meant as a small contribution to improve statistical literacy in the scientific context. Indeed, empirical research relies on statistical methods. We start with a brief introduction to Bayesian inference, making sure it is accessible to scientists and engineers of all backgrounds (using plain language and explanatory diagrams). The benefits of the Bayesian approach include uncertainty quantification (for all model parameters and predictions) and the ability to interpret results. We give a practical demonstration of how to build a statistical model and compute predictions, going through a cognitive science example (more specifically in learning assessment). To do this, we use PyStan, the Python interface to Stan. Stan is the state-of-the-art, free and open-source engine for Bayesian inference.

Marianne Corvellec holds a PhD in statistical physics and works as an industry data scientist. She is also an independent researcher affiliated with the Institute for Globally Distributed Open Research and Education (IGDORE). Her research interests include data science, education, and assessment. Since 2013, she has been a regular speaker and contributor in the Python, Carpentries, and FLOSS communities.