In our efforts to modernize the statistics curriculum, we have experimented with R in lower-division courses for the non-major and in more advanced courses on statistical computing. We have found that incorporating R into our courses necessitates changing what we teach and how we teach. We spend more time on graphics where students critique, compose and create visualizations that tell stories and uncover structure in data. On-line data visualization tools for shared visualization, e.g. Many Eyes, offer an abundance of graphics with accompanying data for students to critique and reconstruct in R. Virtual earth browsers such as Google Earth offer an alternative medium for presenting spatial-temporal data where the interface makes it easy for the user to bring in auxiliary information, such as geographic features, political boundaries, and additional data. With Scalable Vector Graphics (SVG) students can create interactive and animated graphical displays for viewing in a browser. In our more advanced courses, students use R to present their findings in these new venues (e.g. Google Earth, SVG). At both the introductory and advanced levels, we aim to expose students to the vast amount of digital information available to them and show how graphics can be used to analyze and make sense of large complex data. We have found it rewarding to teach statistical thinking by having students construct rich visualizations with R.
Keywords: Introductory statistics; Graphics
Biography: Deborah Nolan is Professor of Statistics at the University of California, Berkeley. She has been recognized at Berkeley for excellence in teaching and undergraduate student advising and is noted for working with and encouraging all students in their understanding of statistics. Deborah is the co-author of Stat Labs: Mathematical Statistics through Applications with Terry Speed and Teaching Statistics: A Bag of Tricks with Andrew Gelman. Together with Duncan Temple Lang she has recently published the paper, Computing in the Statistics Curricula, which presents a broad set of computational topics our students should learn in their preparation to become statisticians and ideas on how to teach these topics.