SORTEE Webinar: Assessing Computational Reproducibility in Ecology and Evolution

For this month's Workshop & Webinar series, SORTEE is pleased to announce that Dr. Steve Kambouris from the University of Melbourne will be presenting a webinar on assessing computational reproducibility in ecology and evolution.

Description:

Researchers are increasingly sharing both the data and code associated with research publications in open repositories and archives. This makes it easier for interested researchers to computationally reproduce the results reported in a publication, by re-running that code with the data. This presentation examines the activity of computationally reproducing published results, by presenting and discussing the outcomes of my attempts to reproduce results from the Ecology and Evolution literature published over 2015-17. I will cover the technical aspects of my efforts to get R code from different times and versions to run, highlighting both my successes and failures, but will also address more fundamental questions, such as how we might decide on the criteria to judge whether a reproduction attempt has been "successful" or not. I will also discuss the benefits (and difficulties) of working reproducibly when embarking on assessments of computational reproducibility of other work.

Presenter:

Dr. Kambouris is a Research Fellow at the Australian Council of Educational Research and a member of the MetaMelb research group at The University of Melbourne. His research focuses on computational reproducibility, including ways of working reproducibly, assessments of reproducibility, and where computational reproducibility fits in conceptually within the wider contexts of replicability and open science.