A Bayesian inspired approach to reasoning about uncertainty: 'How confident are you?'

Abstract

This paper focuses on a task design that is aimed at eliciting young students' reasoning about uncertainty as it relates to their personal degree of confidence through a Bayesian inspired informal inferential reasoning about chance games. This Bayesian inspired approach is described and discussed based on preliminary analyses of data from a teaching experiment in a designed-based research study. With this approach, beginning by a hypothesis (or prediction) about the fairness of a game and revising it based on new information appear to come natural to the students. The change in strength of their personal level of confidence seems to vary by the conflicting results obtained by playing the game, the size of the data collected, and the multiple computer simulations conducted using TinkerPlots software.