Step 4: Sharing and Publishing Data

Introduction

Research data form the basis of insights and scientific progress. Data collected or generated during the research process are not only relevant to one’s own work but also to others. On one hand, research often takes place in collaboration with other researchers and partner institutions, requiring data to be shared and exchanged across institutional and national boundaries so that they can be further processed from different perspectives. On the other hand, third parties such as other researchers and the public may have an interest in existing data, independently of the original research project and purpose. This enables them to verify or reuse the data instead of having to collect the same data themselves. To provide this access, research results and data must be published and thereby made available.

In principle, according to good scientific practice and the FAIR principles, research results should be brought into the scientific discourse for reasons of transparency, reusability, and connectivity of research [1]. The basis for this is “making research results publicly accessible” [1], for example, by publishing research data in a recognized data archive. Additionally, requirements from funders, partners, and publishers must be considered. At the latest by this stage, researchers should decide whether and how they can provide third parties access to which data. Not all data produced are relevant or suitable for publication, and shared data do not always equate to open data. When selecting and preparing data for publication, attention should be paid to the quality of the data.

Why should research data be shared and published? [2]

• Increases one’s own visibility, citability, verifiability, and reproducibility
• Facilitates the exchange and preservation of knowledge
• Enables reuse instead of new data collection—saving costs
• Prevents data loss—data remain available for further research
• Facilitates collaboration and the development of new partnerships
• Promotes an open science culture—the societal benefits of open data
• Fulfills the requirements of third parties (university, funders, publishers)

For scientific datasets to be easily found and reused by others, standardized formats and metadata should be used. The accompanying metadata can differ significantly depending on the data type and academic discipline.

Clear identification, as well as permanent referencing and discoverability, is achieved by assigning Persistent Identifiers (PIDs)—durable, digital identifiers. Frequently used identifiers for data and digital objects include DOIs (Digital Object Identifiers) and URNs (Uniform Resource Names), as well as ORCID (Open Researcher and Contributor ID) and ROR (Research Organization Registry) for the unique identification of individuals or organizations.

In all cases, storage media are needed where data can be securely stored and made accessible to others with different permissions and visibility levels. The following are recommended:

• Disciplinary, institutional, or generic repositories (storage location for digital objects, archive)
• Data journals
• For sharing within collaborative projects: cloud services (of your own university or through the Lower Saxony Academic Cloud)

When publishing data, information regarding their reuse must be provided, typically by assigning a license, such as a Creative Commons license.

Data publication can occur in various ways. Provided there are no legal barriers, open access publication is especially recommended to make data available to a broad audience of potential users.

Legal Aspects & Ethics

When sharing and publishing data, legal and ethical frameworks must be observed (see excursus: Legal Aspects in RDM).

• First, research consequences and ethical considerations should be evaluated.
• The ownership and copyright of the data must be clarified before publication.
• Contractual agreements and confidentiality obligations with third parties must be honored.
• Research results pertaining to patent applications should not be published.
• For personal data, an ethics approval or consent form is necessary, and data security must be ensured through anonymization, pseudonymization, or security standards such as ISO27001.
• Depending on the field, further legal or ethical considerations may play a role when deciding whether to publish data. For example, in biodiversity research, species protection concerns may prohibit the publication of exact coordinates of species locations.

[1] Deutsche Forschungsgemeinschaft (2022). Leitlinien zur Sicherung guter wissenschaftlicher Praxis. Kodex. https://doi.org/10.5281/zenodo.6472827, Leitlinie 13: Herstellung von öffentlichem Zugang zu Forschungsergebnissen
[2] https://www.tu-braunschweig.de/forschung/forschungsdaten-transparenz/forschungsdaten/grundkurs-forschungsdatenmanagement/archivieren-publizieren-und-teilen-von-forschungsdaten