Reproducibility

Definition

“A central tenet of science: To produce a reliable scientific body of knowledge, researchers must be able to trace the steps of each other’s work and verify that they yield the claimed results, or to explore why they don’t.”

Definition source: Reproducibility Network (n.d.). About.

Stakes

part of: opacity

related to: documentation; FAIR; accessibility; provenance

If research is not reproducible, it means the process has been untransparent. The research, as a result, is not trustworthy and lacks accountability.

Where does it occur in the lifecycle?

4 - Analyse

• Analysing and Interpreting Data

5 - Preserve & Share

• Create or choose an interface to publish data, analyses, visualisations and other publications

Questions to consider throughout your work

• Have you documented your analysis process sufficiently for others to reproduce it?
• How are you handling intersectionality in your analysis?
• Do your interpretations risk essentializing complex identities?

Examples

• Peels, Rik, Gijsbert van den Brink, Hans van Eyghen, and Rachel S. A. Pear. “Introduction: Replicating John Hedley Brooke’s Work on the History of Science and Religion.” Zygon: Journal of Religion and Science 59, no. 2 (2024): 2. https://doi.org/10.16995/zygon.11255.

Good-better-best practices

Good	Better	Best
Make your research data freely available.	Make data documentation available and downloadable	Keep detailed and openly accessible documentation of decisions made and actions taken.
		Adhere to the FAIR Principles.

Resources

• Harjes, Janno, Anton Link, Tanja Weibulat, Dagmar Triebel, and Gerhard Rambold. “FAIR Digital Objects in Environmental and Life Sciences Should Comprise Workflow Operation Design Data and Method Information for Repeatability of Study Setups and Reproducibility of Results.” Database 2020 (January 2020): baaa059. https://doi.org/10.1093/database/baaa059.