At DANS, we make datasets available in the Data Stations with this purpose in mind: to enable them to be found, understood, and reused in a variety of contexts. In the ‘Reuse in Practice’ series, we share experiences of reusing research data. We speak to users who have reused data and to depositors about the potential for reusing their datasets. What works in practice? What challenges arise? What can others learn from these experiences?
A recent case study demonstrates that reuse not only occurs within research projects, but also plays a vital role in the training and professional development of researchers and data specialists. David Rayner, Education Coordinator at the Swedish National Data Service at the University of Gothenburg, Sweden, contributed this example.
A dataset as a starting point for learning reuse
In a training on open science delivered by Rayner, a dataset from the Data Station Social Sciences and Humanities (SSH) was used as a case study. The dataset “Data Discovery and Reuse Practices in Research” (Gregory, 2020) gave participants a practical starting point for understanding data reuse. They worked through the entire process, from finding the data to interpreting it in a new context.
Finding data in practice
An important part of the training focused on data discovery. Participants explored various strategies for locating datasets, such as searching related academic publications, conducting general web searches and using repository search engines like OpenAIRE.
This demonstrates that datasets can be found through multiple routes. The findability of datasets depends strongly on how they are linked to publications, included in aggregation services, and described through metadata.
From access to interpretation
After locating the dataset, participants assessed how suitable the data was for reuse: from reviewing the access conditions to interpreting the documentation and metadata to gain a better understanding of the dataset. The dataset functioned as a realistic example, enabling participants to apply their theoretical knowledge directly to a practical situation. This demonstrates the dual role of the DANS Data Stations as both digital archives and infrastructure that supports reuse.
Challenges in reuse
The case also made clear that reuse is not self-evident. Two key challenges emerged:
- Linking data and publications: the metadata did not clearly indicate the relationship with the associated publication, making it more difficult to assess context and provenance.
- Use of controlled vocabularies: the absence of standardised terms in the description led to differences in interpretation.
These issues are common across many datasets and emphasise the importance of consistent, high-quality metadata.
In this case, some of these challenges were overcome thanks to domain expertise. This demonstrates that successful reuse depends not only on the availability of data, but also on the user’s skills.
Training and guidance are therefore essential. Working with existing datasets enables participants to develop the ability to critically assess and interpret data, even when documentation is not fully standardised.
Looking ahead
Experiences such as this provide concrete input for improving datasets and services. They show where metadata can be strengthened, where links can be made more explicit, and how users interact with data in practice. DANS is actively working to improve thequality of the metadata and the functionalities of our data services in order to facilitate reuse more effectively.
This use case demonstrate that reuse extends beyond research alone. Using datasets for training strengthens the skills required for sustainable and open research practices.
For researchers and data professionals, reuse is not just an abstract principle, but a tangible practice that can be developed, supported by infrastructure, and realised through interaction with data.
Contribute to this series?
We invite researchers and data professionals to share their experiences with data reuse or share what the reuse potential of their data is.
