Frequently asked questions (FAQ) on research data management
What are research data?
Research data are data generated in the course of scientific projects, e.g. through digitisation, research, experiments, measurements, surveys or interviews. See Forschungsdaten, Beschreibung des Handlungsfeldes der Allianzinitiative Digitale Information.
What is research data management?
Research data management (RDM) combines a variety of measures and procedures, methodological, conceptual, technical and organisational, to handle research data within its digital lifecycle and beyond. RDM therefore encompasses the creation, processing, enrichment, publication and archiving of data.
RDM is part of good scientific practice. Research funders, as well as publishers and journals, increasingly require that data associated with a publication be made available for re-use. RDM measures and procedures can be laid down in a data management plan.
Why is research data management important?
- fulfils the requirements of research funders,
- saves time and resources,
- reduces the risk of data loss through measures such as data documentation, backup and archiving,
- makes research output verifiable and reusable,
- increases the visibility of own research.
Data management plan (DMP)
What is a data management plan?
A data management plan (DMP) should be drawn up before the start of a research project or during the application phase. It describes the entire life cycle of the data, from collection to archiving, as well as measures to ensure that the data remain usable and traceable after the project ends. Many research funders expect information on the handling of research data as part of a funding application.
What contents should a data management plan contain?
A data management plan should adress the following points:
- General information on the research project
- Existing Data
- Data generated in the project
- Data organisation
- Administrative and legal aspects
- Archiving, data exchange and data publication
- Responsibilities and duties
- Costs and resources
What requirements must be met when submitting research proposals to the DFG?
In its guidelines on handling research data, the DFG recommends that the data generated be made available to third parties as soon as possible for at least 10 years. The project-specific costs incurred for the re-use of the research data can be applied for within the project.
All higher education institutions and non-university research institutions must implement the DFG's Guidelines for Safeguarding Good Scientific Practice in a legally binding manner in order to receive funding from the DFG. Research data management is particularly addressed here.
The data handling should be explained in the proposal. This should include standards for documentation, metadata, data formats and data organisation. Existing data repositories and archives should also be considered for making the data accessible. Furthermore, there are numerous discipline-specific guides.
What requirements must be met when submitting research proposals to the BMBF?
Depending on the announcement, research data are of varying relevance.
In general, the BMBF expects a "utilisation plan", in which, among other things, scientific and/or technical possibilities for re-use are to be presented.
In some cases, the implementation of research data management must already be presented in the proposal and is part of the review criteria.
What requirements must be met when submitting research proposals to Horizon Europe?
A data management plan is mandatory. Initially, only a document with RDM considerations is required for the application, which must be updated in case of significant changes as well as at the end of the project.
Where can I find more information on creating a DMP?
A compilation of useful tools and templates for creating data management plans can be found on the website Forschungsdatenmanagement Bayern and on the information portal forschungsdaten.info.
Finding, Citing, Re-use
Where can I find research data?
Research data for your own research area can be found in institutional or subject-specific repositories. An overview of repositories is provided by the Registry of Research Data Repositories (re3data).
In addition, you can find already registered datasets via data portals such as DataCite or the European Union Open Data Portal.
How do I cite research data?
The correct citation of research data is part of good scientific practice and indispensable for the use and re-use of own and third-party data.
A citation should include information on the author, publication date, title, publication agent and an identifier. Optionally, further information, such as the resource type, can be added. The following quotation style is recommended: Creator(s) (publication date): Title. Data repository or archive. Version. Persistent identifier (preferably as a link).
Further information on citing research data can be found here.
Storage, Publishing, Archiving
How can research data be stored?
The easiest possibility to store research data is to store them on centrally provided network drives from the university. Data sets smaller than a couple of terabytes can be stored there unbureaucratically and without consultation. In case you expect larger data sets, you should contact the University IT-Service before your research proposal to clarify how your data will be stored. These forms of storing data require an individual system to manage the data and metadata by the researcher.
Research data repositories are better suited for storing research data. If repositories are already established for a discipline, it is advisable to store the data there. If no repository is available, the data can be stored in the research data repository of the University of Bamberg (currently under construction).
How can research data be published?
Within the open-access declaration, the university management asks researchers from the university to publish their research. Within the Open Science movement the publication of research data is highly regarded, too.
For publishing research data, both institutional, cross-disciplinary (e.g. Zenodo, Dryad or Figshare) and subject-specific repositories are possible. The latter should be preferred, as they offer the advantages that subject-specific standards and metadata schemes can be better taken into account and more specific indexing and search options can be provided. If there is no suitable repository for your discipline yet, you can publish your research data in the research data repository of the University of Bamberg.
In addition, there are special "data journals" that specialise in publishing research data. A (not complete) overview can be found at www.forschungsdaten.org/index.php/Data_Journals.
How do I find a suitable repository?
To search for a suitable and trustworthy repository, you can use the Registry of Research Data Repositories (re3data) nutzen. re3data offers search entries via the subject area, data type or operator of a repository.
What should be considered when publishing research data?
- Suitability of the repository for the discipline
- Long-term availability of the repository
- Data formats and metadata standards used
- Legal aspects (e.g. data protection, copyright, personal rights, licensing)
How can research data be archived?
The German National Library takes over the long-term archiving of publically available research data that are the result of a published dissertation, like mentioned above.
The central systems of the University IT-Service allow, for the time being, an indefinite storage option. But this does not constitute long-term archiving in the strict sense. When storing data in a subject specific repository, it has to be clarified in how far a long-term archiving is part of the offer.
Digital reproductions can be stored in the long-time archive “Rosetta” of the library network of Bavaria. In that case, please, contact the university library.
Data that are classified as archive worthy under the Bavarian Archive Law (BayArchivG) and are supposed to be long-term archived at the University of Bamberg will be managed by the university archive. The university archive decides in consultation with the researcher or superior institutions, e.g. institutes or research associations, about the archive-worthiness of the research data.
Documentation and metadata
What are metadata and metadata schemas?
Metadata is data that contains structured information about other data. They describe the data in order to increase their discoverability and ensure traceability for subsequent users. Furthermore, they enable similar research data to be linked, provided that standardised metadata schemas or standards are used. This contributes decisively to discovery and re-use.
Different types of metadata are distinguished:
- bibliographic metadata: Information such as title, authors, description or keywords enable the citation of data and make it easier to find.
- administrative metadata: Provides information about file types, locations, access rights and licences and is used to manage the data.
- descriptive metadata: These are structured very differently depending on the discipline and provide information on the content and context of origin of the data. For research data, this type of metadata is particularly important for its discoverability and traceability. For this reason, there are different metadata standards that specify information relevant to a subject discipline.
Metadata schemas or standards ensure that metadata can be compared and linked with the metadata of other research data. For this purpose, the terms used must be standardised, i.e. uniform terms are provided with fixed definitions.
Specific metadata standards already exist for some disciplines. The Research Data Alliance and the Digital Curation Centre provide an overview. If no subject-specific standard exist yet, generic ones such as the DataCite Metadata Standard can be used.
Law and licences
Which licences can I choose for the publication?
Publishing data under a specific licence allows you to define a permissible form of use and presents it in an easily understandable and comprehensible way. To ensure broad reusability and transparency of your research data, the use of free licences is recommended.
The most widespread free licence model is Creative Commons (CC), which, in addition to copyright and ancillary copyright, also covers database producer rights in the most recent version. CC licences are suitable for texts, images and data.
Open Data Commons was designed specifically for data publishing.
What legal aspects do I have to consider?
- Copyright: Research data may be subject to copyright protection. This must be examined in each individual case. As a rule, research data do not reach the necessary level of creation, but may be subject to ancillary copyright (e.g. photographs). When research data are re-used, any copyright protection must be checked.
- Data protection: The collection, use and disclosure of personal data is subject to strict regulations under the Data Protection Act. Information contained in research data by means of which a person is identified or identifiable must be removed for publication and archiving. If necessary, anonymisation of the data is also suitable.
A detailed overview of the topic of law and research data is provided by forschungsdaten.info.