ESSB Repro Checks
Have your paper checked for reproducibility
Do you want to make sure your next research paper is reproducible? Then have it checked!
How does it work?
If you have results from a quantitative analysis that you are about to publish, you can send the data and code to me for an internal review. I will perform a code review to see if your code can run on another machine and if it will produce the same output as on your machine.
Let me know you are interested and provide some background about your paper or project (via email or via the form).
If possible, share your data and code on a public repository, so that I can access your materials. In case this is not possible (e.g., for privacy sensitive data), we will figure out an alternative arrangement.
I will try to (computationally)1 reproduce the results. This means I will try to regenerate the outputs including figures and tables of your paper.
Based on the previous step, I will provide you with a short feedback report including suggestions on how to increase the reproducibility of your findings.
Please send me a message via Teams or email (e.klapwijk [at] essb.eur.nl) if you are interested in piloting reproducibility checks at ESSB. You can also fill in the form on the submit your paper page and I will contact you.
The why
Why is this important? The repro checks:
Ensure a basic quality level of your results (being computationally reproducible).
Lower the barrier to share your analysis code with others.
Provide you with feedback to improve your data documentation and code.
Additional resources
If you want to learn more about (computational) reproducibility, please join one of our workshops! Here are some interesting papers, projects and other resources that might help to get you started:
The ESSB Repro Checks are heavily inspired by the following projects:
Reprohack: Building Communities of Practice in Reproducibility
The “reproducibility statistician” project at the the School of Psychology at the University of Sussex
Project background
The ESSB Repro Checks are part of the eScience Fellowship project Implementing institutional reproducibility checks to promote good computational practices awarded to Eduard Klapwijk, Research Data Steward at ESSB.
The project is supported by Sven van den Burg, Research Software Engineer at the eScience Center.
Footnotes
Computational reproducibility means that rerunning the same analysis on the same data should reproduce the same results↩︎