Online Program Home
My Program

Abstract Details

Activity Number: 128 - Curricular Considerations for Statistics and Data Science Education
Type: Contributed
Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Education
Abstract #329246 Presentation
Title: Revisit: a Statistical Teaching Tool in R
Author(s): Tiffany Eunice Chen and Emily Watkins* and Norman Matloff
Companies: UC Davis and UC Davis and University of California at Davis
Keywords: Revisit; education; teaching; reproducibility; student
Abstract:

The R package 'revisit', developed largely in response to the growing concern over lack of reproducibility in research, serves as an excellent teaching tool for students. Its primary interface is a GUI written as an RStudio add-in. Pedagogically, the package aims to (a) develop good statistical habits in the students, and (b) aid the exploration of alternative analyses with the growing list of built-in case studies. For example, consider the potential effects of outliers. In terms of (a), the package includes built-in wrapper functions, such as one wrapping lm(), named lm.rv(). The latter not only performs lm()'s operations but also adds commentary, such as "Note that a robust regression model gives substantially different results from lm() in this instance. Consider deleting some outliers." Concerning (b), the package would then allow the student to remove the outliers, and "replay" all the analysis she had done earlier. This allows the student to conveniently explore a variety of analyses on the same data, say using/not using a certain predictor variable.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program