Aaron Rendahl takes deep dive into using R and Rstudio coding to standardize statistical analyses

Aaron Rendahl, associate professor of statistics and informatics in the Department of Veterinary and Biomedical Sciences (VBS) at the College of Veterinary Medicine (CVM), took a sabbatical leave from teaching during the Fall 2024 semester. Rendahl shared with us what he learned from his sabbatical, as well as how he will use what he learned, which will contribute to his teaching going forward. 

Can you tell us a little bit about your background and how you got into the field of statistics?

I have always loved the connections between mathematics and the world around me, and during my graduate work in statistics, I really came to appreciate the way that statistics could help answer scientific questions. After graduating, I ran the University Statistical Consulting Center for almost a decade, collaborating with researchers from all around campus. During that time, I made many connections within CVM, so when the opportunity for this position arose, I knew it would be a great fit.

Why did you decide to take a semester off at this point in your career?

In addition to my research collaborations, I teach two statistics courses in the graduate program, and, over the years, have made changes each year to fit my style and the needs of our students. However, although each individual change seemed to help, the overall arc had suffered, especially for the fall course, so I took the semester off from teaching to rework the course and consider more holistically how the elements of the course should fit together. I did continue many of my existing collaborations during the semester, though I did attempt not to take on new large projects.

What did you learn during your semester away from campus?

In my consideration of my fall course, I determined that one of the major issues I was struggling with was to balance student understanding of statistical concepts with the computing needed to actually do the analyses. I decided that the focus needed to be more on the statistical concepts, so I wrote an R package to standardize both the coding and the output of the analyses used in the class. While I still expect students to do all the analyses, the idea is to remove some of the distraction and frustration of the computing so students can better focus on the statistical concepts and ideas. I hope to present this work at the U.S. Conference for Teaching Statistics in July.

Is there anything you would like to bring from what you learned/experienced recently into your teaching or statistical consulting?

I used the package in my course this past fall, and it seemed to be a success; the feedback I got was mostly positive, and my TA and I noticed substantially fewer questions and difficulties with the computing. The package has also been useful in my collaborations; while when I wrote it, I primarily had in mind the pedagogical improvements, I also brought to the work my years of experience actually doing statistical analyses, and built in features that have been useful not only in the classroom but in actual research analyses.

Anything else you would like to share? 

Thanks to the department for funding this break; being away from the day-to-day of teaching was truly a gift both personally and professionally. I look forward to continuing collaborating with you! 

If you are interested in looking into the package I wrote, you can find it at https://aaronrendahl.github.io/umncvmstats/.