Why share data?
At the heart of the scientific process is a raw material without which research would be impossible: research data. Oftentimes a researcher or research group works with data and then stores it on a USB stick in a drawer after the project is done. Sharing one's data, however, enables round-the-globe academic synergies. By sharing data, other researchers can test, replicate and re-analyse them.
Enabling meta analyses is another reason for more openness with one's data. In some disciplines, that encourage an open data culture, errors in well-established measuring methods or tools have been discovered by meta-analyses that showed inconsistencies across research data. In this way meta analyses based on shared data can help avoid false and distorted results.
What are the advantages of sharing data?
- New world-wide cooperations can emerge.
- Research results become more visible through citation of published data. 
- Openly accessible data can inspire new or complementary research ideas.
- Data sharing encourages an open scientific culture.
Who supports data sharing?
Reserach data and publications can be archived and published in online repositories. In recent years a large number of repositories have come into existence. Depending on the discipline, the choice can be quite varied.
An overview of existing repositories world-wide as well as a search function with many useful filters is provided by re3data. It is possible, for example, to filter for repositories from specific disciplines or for repositories who provide different access options (open access, restricted access).
The CoreTrustSeal is a certification organisation that has originated from the ICSU World Data System (WDS) and the Data Seal of Approval (DSA). The CoreTrustSeal Data Repository certification replaces the DSA certification and WDS Regular Members certification. The CoreTrustSeal is an international, community based, non-governmental, and non-profit organization and offers every interested repository a core level certification based on the DSA–WDS Core Trustworthy Data Repositories Requirements catalogue and procedures. The overall 16 guidelines seal guidelines are mainly founded on 5 criteria:
1.) The data can be found online.
2.) The data are accessible without limitation and ownership of the data is clear.
3.) The data are available in a common format.
4.) The data are reliable.
5.) The data are citable via a persistent identifier.Mehr erfahren
nestor seal for trust-worthy long-term archives
Nestor is a German competency network for long-term archiving of digital artefacts, which focuses on the harmonisation efforts of all partner institutions (archives, libraries, museums). The nestor seal comprises a list of 34 criteria including legal, budget and staffing aspects of the certified institution. The stored data are evaluated according to their quality, citability, re-usability and long-term durability (data formats and storage mediums).Learn more
Another possibility is is the publishing of research data in data journals. These journals are dedicated to the description, documentation and evaluation of research data. Data journals usually do not require a scientific publication, but a description of the data in the form of a paper or a Data Curation Profile. The data set itself is archived in a repository with the journal article referencing the data set via a persistent identifier.
What to consider when sharing data
- Is the repository or data journal suitable for your discipline?
- Can you discern whether the repository will be operational in the long term?
- Do the offered acces options fit your needs (closed/restricted/open access)?
- Which data formats and meta data standards are being accepted/used?
- What information does the provider give about data security?
- Is copyright taken into account and is data citation mandatory?
- Have their data licensing and data reuse contracts been legally vetted?
- Are the research data that are archived on the repository findable by discipline-specific search engines?
- Sensible/personal data: are data anonymised/pseudonimised before uploading?