Quantitative critical thinking: Student activities using Bayesian updating

Aaron R. Warren

DOI: https://doi.org/10.1119/1.5012750

IF: 0.835

2018-05-01

American Journal of Physics

Abstract:One of the central roles of physics education is the development of students' ability to evaluate proposed hypotheses and models. This ability is important not just for students' understanding of physics but also to prepare students for future learning beyond physics. In particular, it is often hoped that students will better understand the manner in which physicists leverage the availability of prior knowledge to guide and constrain the construction of new knowledge. Here, we discuss how the use of Bayes' Theorem to update the estimated likelihood of hypotheses and models can help achieve these educational goals through its integration with evaluative activities that use hypothetico-deductive reasoning. Several types of classroom and laboratory activities are presented that engage students in the practice of Bayesian likelihood updating on the basis of either consistency with experimental data or consistency with pre-established principles and models. This approach is sufficiently simple for introductory physics students while offering a robust mechanism to guide relatively sophisticated student reflection concerning models, hypotheses, and problem-solutions. A quasi-experimental study utilizing algebra-based introductory courses is presented to assess the impact of these activities on student epistemological development. The results indicate gains on the Epistemological Beliefs Assessment for Physical Science (EBAPS) at a minimal cost of class-time.

physics, multidisciplinary,education, scientific disciplines