Computational Reproducibility and the Methods Hub

Materials for the 2025 ICWSM workshop session on the use of Jupyter4NFDI, MyBinder and the Methods Hub by Taimoor Khan, Lorraine Saju, and Arnim Bleier. The materials have been presented in the following workshops:

• Social Science Meets Web Data: Reproducible and Reusable Computational Approaches (R2CASS)
• 1st Workshop on Validating Social Text-as-Data Methods (VaLiSTAD)

Target Audience

The sessions are intended for participants in the social sciences who have at least some experience with Python or R and are interested in learning more about tools and workflows for making their research results reproducible.

Learning Objectives

By the end of the session, participants will:

• Understand the purpose and capabilities of the Methods Hub, and how to publish, discover, and execute tutorials and methods on the platform.
• Gain initial insights into key concepts and best practices in computationally reproducible research.
• Be able to use frameworks and tools such as Git, Python and R packages, and Binder to enhance reproducibility.
• Be equipped to publish reproducible computational analysis pipelines using Jupyter4NFDI and MyBinder in either Python or R.

Content

• Methods Hub Portal – Taimoor Khan
• Tools and Workflows for Reproducible Research – Arnim Bleier
• Reproducible Research in Action: A Hands-On Binder Tutorial – Lorraine Saju

Acknowledgments

These materials are based on earlier versions of this session, including:

• Workflows for Reproducible Research with R & Git
• The GESIS KODAQS Academy
• Presentations at the CESSDA Conference 2024
• GESIS Notebooks.

This work received funding from the German Research Foundation (DFG) through the project NFDI4DS (no. 460234259), in cooperation with Jupyter4NFDI (no. 521453681).

Please link to the workshop GitHub repository