|
Activity Number:
|
125
- SPEED: Modernization of What, How, and Where We Teach Statistics Part 1
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 29, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics and Data Science Education
|
|
Abstract #306364
|
Presentation
|
|
Title:
|
Active-Learning for Bayesian Inference: An Introductory Exercise Using MandM's Candy
|
|
Author(s):
|
Gwendolyn Marie Eadie* and Daniela Huppenkothen and Aaron Springford and Tyler McCormick
|
|
Companies:
|
University of Washington and University of Washington and Weyerhaeuser and University of Washington
|
|
Keywords:
|
active learning;
Bayesian;
education;
eliciting priors
|
|
Abstract:
|
Using m&m's candy as a teaching tool for classes in freqentist statistics has been popular for many years, with the colour distribution of the chocolates often playing a dominant role. However, there are only a couple of exercises for Bayesian statistics that take advantage of these yummy candies. In this talk, I will describe a full-fledged active-learning exercise we developed that uses m&m's candies to introduce Bayesian inference to students in a fun (and tasty!) way. In this activity, students obtain a posterior distribution for the percentage of blue m&m's produced at the factory given a bag of m&m's they are given in class. The exercise gives them a chance to quantify their prior information about different m&m's colours before opening the data. I will describe how our exercise works in practice, in the context of the intended learning outcomes and lesson plan. This presentation may also include actual m&m's and audience participation.
|
Authors who are presenting talks have a * after their name.
