Teaching statistics to Real People: adventures in social stochastics

Presenter

• Rachel Fewster (University of Auckland)

Abstract

Students arrive in our first undergraduate course in statistical theory with many different backgrounds and motivations: from mathematically able students in their first year at university, to math-phobes who have put off the dreaded theory course to the last possible moment in their degree programme. One thing unites all these students: they are real people, immersed in their own worlds and biologically programmed to be social learners. Over the last few years I have tried to put Homo socialis to pedagogical advantage by experimenting with team-based activities in class tutorials. I will outline some of the successes and failures, from illustrating p-values through animated cartoons to a never-to-be-repeated investigation into the carpet-buttering tendencies of dropped toast. Importantly, I will discuss the ramifications of social learning for developing capacity for solo thinking and study. While there are obvious advantages of teamwork for fostering versatile thinkers who can alternate between language, diagrams, and mathematical notation, I will suggest that there are also more subtle effects at play. An abstract problem that has been discussed in a group automatically becomes a real-world experience. I will show how a classroom game to “catch the spies” translated on a final exam to a mature understanding of maximum likelihood estimators, suggesting that the classroom activity provided the tools for students to reach a deep understanding once they were “home alone” in private study.