Presented by:
Chester Ismay & Andrew Bray, Reed CollegeAbstract
A common feature in undergraduate programs is the inclusion of a year-long senior thesis, and many of these utilize data analysis and computing. One goal is for all such theses to feature fully transparent and reproducible analyses. This is a significant challenge, as these theses are occurring in departments across the sciences, with students (and faculty advisors) that have varying degrees of familiarity with the tools that enable a reproducible analysis. To lower this barrier to reproducible analysis, we have put together an R Markdown thesis template which allows students to work in a reproducible, literate-programming environment without needing to know LaTeX. The other initiative that we've worked on is to utilize the new "shared project feature" in RStudio Server. Many intro stats courses use group projects where students apply models to an area and data set of their interest. This mode of collaborative data analysis is a learning outcome in and of itself, but in the past it has been fairly error prone due to the workflow of emailing files back and forth between group members. Shared projects allow all of the collaboration to take place in a shared environment on an RStudio Server. Instructors can also be shared on the projects, allowing them to periodically check-in and provide feedback. This can also be used for grading individual student labs that use R. These two technologies (thesis template and shared projects) can also be used in tandem - having the thesis student and advisor(s) shared on the same project.
Resources
- A set of accounts have been created on a series of RStudio servers for the duration of eCOTS 2016. See http://tinyurl.com/ecots-rstudio for an editable google drive file with login information.
- http://ismayc.github.io/ecots2k16
Recording
(Tip: click the fullscreen control)
Having trouble viewing? Try: Download (.mp4)
(Tip: right-click and choose "Save As...")