Reproducible Research Workflows in the Era of Big Data and Machine Learning

Workshop prepared for NOFE Young Investigator Day, 28.10.2025.

Source on codeberg.

Schedule

PART 1: “Why reproducibility?” (1 hr)

• DISCUSSION: what is reproducibility? what tools do you know and which you already use? how to do it better?
• NAMING THINGS: file organization, file names, file formats
• Short hands-on
• Check out the presentation

PART 2: “When to think about reproducibility?” (1hr)

• Walk-through of a typical scientific project: how to make it reproducible at each step
• Discussion and hands-on
• Check out the presentation

PART 3: “How to do reproducible reports?” (1hr)

• Quarto: create beautiful reproducible reports, programming language agnostic!
• Hands-on

Check out the exemplary reports:

• using R
• using STATA, from R

Requirements

To get the most of this workshop, you should:

• have tried analysis of data
• have tried writing about the results, creating a presentation or a poster
• have been frustrated with the need to re-run analyses and keeping track of all the versions
• be willing to change your habits to make your science more reproducible

Technical requirements

NB: We could not confirm the instructions for running STATA from Python, thus we do not recommend this way

• PC
• installed Quarto
• installed R, Python, or STATA
• installed VisualStudioCode or RStudio
• if using STATA:
  • using STATA in Quarto requires R to setup, so install that also
  • then, open terminal (e.g., powershell in Windows or terminal in VScode) and write “R” and press Enter
  • now, in R, write: “install.packages(‘Statamarkdown’)” and press Enter
  • if the installation was successful, you’re ready to go!