A Project-Based Approach to Statistics and Data Science

Abstract

In an increasingly data-driven world, facility with statistics is more important than ever for our students. At institutions without a statistician, it often falls to the mathematics faculty to teach statistics courses. This paper presents a model that a mathematician asked to teach statistics can follow. This model entails connecting with faculty from numerous departments on campus to develop a list of topics, building a repository of real-world datasets from these faculty, and creating projects where students interface with these datasets to write lab reports aimed at consumers of statistics in other disciplines. The end result is students who are well prepared for interdisciplinary research, who are accustomed to coping with the idiosyncrasies of real data, and who have sharpened their technical writing and speaking skills.

Keywords:

• Statistics education, interdisciplinary
• project-based learning

Acknowledgements

I am grateful to Amy Wagamon, Robin Lock, Jeff Witmer, Shonda Kuiper, Talithia Williams, Rick Cleary, Matt Neal, and the Isostat listserv for sharing teaching materials with me and answering my questions as I learned the content of the courses I developed. I also thank my colleagues at Denison for guidance about how statistics is used in their fields, for teaching me advanced topics, for datasets, and for serving as problem supporters to my students. Special thanks go to Olga Nicoara, Nestor Matthews, and Andy McCall. This work was supported by Denison University’s Center for Learning and Teaching, Pedagogical Practice Projects Grant

Notes

1 This idea is due to Rick Cleary, of Babson University.

2 Adapted from a conversation with Shonda Kuiper in 2016.

3 This idea is the result of a conversation with Lisa Dierker, of Wesleyan University.

Additional information

Notes on contributors

David White

David White is an assistant professor in the Department of Mathematics and Computer Science at Denison University. He received his B.A. at Bowdoin College in 2008, M.A. (in computer science) at Wesleyan University in 2012, and Ph.D. (in mathematics) at Wesleyan University in 2014. He has developed courses at Denison in statistics and computer science. His research is primarily focused on abstract homotopy theory, a subfield of algebraic topology that touches most other fields of math. He has conducted research with undergraduates in statistics, discrete mathematics, and computer science. Course material for the two courses described in this paper can be found at the following two links: http://personal.denison.edu/*whiteda/math401spring2016.html; http://personal.denison.edu/*whiteda/math242fall2015.html