Friday, October 30, 2020 12:50 p.m. - 1:40 p.m.
NR 8107 Graduate Student Seminar Series
Jane Nolan
Masters Student, Natural Resources Science and Management Program
Computational reproducibility is a standard which asks: if an independent researcher were given the same data and followed the same analytical methods presented in a study, would they reach the same conclusions?
A few years ago. I was part of a team working to assess the state of computational reproducibility in wildlife ecology. Working on that project has inspired me to think about how to make my own thesis work reproducible. This is not in anticipation that anyone will ever want to reproduce my thesis - rather, it is an exercise in best practices, Reproducible science is robust science.
My research (a look into variation in carbon storage capacity across a forest-prairie gradient) is well-suited to being made reproducible - I am using publicly available, pre-existing data, my analysis is code-based, and most of it will be completed using open-source software. But even in this ideal set of circumstances, ensuring
reproducibility requires forethought and planning. I will walk through what that process looks like and encourage others to take steps toward reproducibility in their own work as well.