What data to share?

Research data refers to all data created in the course of research and includes source code, algorithms, simulations and models, experimental notes (paper or digital), physical samples, questionnaires, video and audio recordings, protocols, and other forms of information supporting traditional publication (Applied Sciences Research Data Management Policy).

In order to determine what data should be openly shared ask yourself the following questions:

What information would another colleague need to understand

• what you did,

• how you did it and

• how the data can be reused?

Which data can be shared?

• Data restrictions can apply depending on the type of data (personal data) or by contractual agreements with third partners (confidential data).
• Sometimes data is too large to be shared (more than TBs)
• See valid reasons to not share data.

When can it be shared?

Generally it is easiest to share data at the same time as the publication/article. This way you can still correct errors if there are any, and you’re still familiar with the data organisation.

If the data cannot be shared:

• Consider sharing the metadata about the data in a repository Increases findability/discoverability.

• Explain to others what they have to do to request access. For example, 4TU.ResearchData offers restricted access and a way to request access to these restricted datasets.

TU Delft Research Data Framework

Ensuring that research data, code and any other materials needed to reproduce research findings are appropriately documented, stored and shared in a research data repository … unless there are valid reasons not to do so.

Faculty of Applied Sciences Research Data Management Policy

The minimal requirement is the deposition of the numerical data (processed data) underlying the figures in academic papers and PhD theses. Deposition of datasets not directly linked with publications or raw data/software and processing scripts used to derive the processed data is encouraged.