Bayesian Inference in the Introductory Physics Lab

Bayesian inference provides a consistent and scalable method for drawing conclusions given incomplete and uncertain data. Teaching Bayesian inference and data analysis in the introductory physics lab has several benefits. Bayesian inference provides a unique framework for updating prior knowledge with experimental data to obtain probability distributions for model parameters. Learning Bayesian methods requires students to follow logical thought processes. Introductory physics students will hopefully carry their understanding of Bayesian methods to other disciplines where the misuse of traditional frequentist methods has become problematic. Bayesian methods should not, of course, supplant traditional frequentist methods. When presented alongside traditional methods, they can be compared and contrasted, leading to a deeper understanding of both methods. We will show that Bayesian methods can be taught effectively at the introductory physics level, present examples of Bayesian inference in the introductory lab, and demonstrate appropriate computational tools.