Data archiveren in stappen
The infrastructure that enables sharing research data is layered:
- First a researcher saves his or her data locally or in a data lab. In most cases, the data will not be available for others to view yet.
- The institute will save the data for the short term, for example in an institutional repository.
- After selection, the data will be saved in a data archive, for example 4TU.Centre for Research Data or EASY.
- The data can be made available under various conditions: under embargo, closed access or open access. All cases bring data publications. The most optimal form of data publication availibility is open access, because then the data can be used without use restrictions.
To visualise the layered structure the Edinburgh Research data blog discusses 'The four quadrants of research data curation systems'.(2)
Whereas research projects usually have a short lead-time, data archives often focus on continuity. For instance:
- Long-term storage.
This is for reusing data.
- Facilitating interoperability.
In order to be able to compare and combine data sets with each other, it is important that metadata are assigned consistently and that metadata standards and data formats are used.
- Facilitating a data citation network.
Being able to cite data and link data and literature leads to more transparency in academic science and furthers scientific integrity.
The White Paper 'Sustaining Domain Repositories for Digital Data'(3) describes the roles of data archives in detail.
The below accordion lists a number of categories for data archives and provides one or more examples per category. In the searching for data section we will discuss how to find the appropriate data archives in more detail.
Case Wageningen University
The university library of Wageningen partners (in Dutch)(20) with DANS and 4TU.Centre for Research Data to transfer research data from the institution to the data archive.
A double interview(21) held with researcher Frits van Evert (Agrosystems Research) and library employee Annemarie Patist explains how that process generally works: