Modernizing the Introductory Statistics Course: Challenges and Rewards

• Activity Number: 109
• Type: Invited
• Date/Time: Monday, August 1, 2011 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Mark Hansen and Deborah Nolan*+
• Companies: University of California and University of California
• Keywords: education ; introductory statistics ; graphics ; simulation

Abstract

In our efforts to modernize the statistics curriculum, we have experimented with using R in lower-division courses for the non-major. Our goals are to expose students to the vast amount of digital information available to them and how statistics can be used to analyze and make sense of these data. We have found that incorporating R into our courses necessitates change in both what we teach and how we teach. We spend more time on graphics, where students create visualizations that tell stories and uncover structure in data. Also, students layer small computational steps to construct simulations to understand a statistical concept or assess whether data matches expectations (whether those expectations are for a plot or summary statistic). One challenge we have encountered is how this approach does not integrate well with traditional curriculum and textbooks that emphasize normal approximations over exact and simulated distributions, univariate over multivariate approaches, simple graphics such as stem-and-leaf plots over, e.g., density, mosaic, violin plots, etc. It is time to abandon the old curriculum and design a modern statistics course from scratch.