Data management plan
A data management plan (DMP) describes the research data you intend to collect or generate, and how you plan to describe, store, and preserve the research data during and after the project.
Main content
The purpose of a data management plan is to ensure that research data is handled safely and responsibly during a research project and to prepare the data for long-term preservation. Data management plans are also useful for researchers because they:
- make it easier to identify and address key issues related to research data early in the project (such as consent for data sharing or copyright)
- help outline any additional costs or resources needed for managing research data (e.g., extra storage space)
- allow for early planning of data management needs and monitoring of data-related activities throughout the project
- help others make use of your data if you need to share it
WHY write a data management plan?
The Research Council of Norway and the EU's framework program, Horizon Europe, require projects to have a data management plan (DMP). Additionally, The University of Bergen Policy for Open Science states that all research projects led by UiB must include a data management plan.
The purpose of the data management plan is to outline how research data will be secured, both during the project and for future reuse. A data management plan is also useful for identifying the costs associated with data management and storage, as well as planning how these costs can be covered.
The researcher should establish a data management plan early in the project, ideally within six months of its start. The plan is a living document and should be updated as needed throughout the project.
Requirements from funders include:
See also Nasjonal strategi for tilgjengeliggjøring og deling av forskningsdata .
DMP 50 seconds
WHAT is a data management plan?
The Research Council of Norway describes the data management plan as a living document that should accompany the research project. It should detail which data will be generated, how the data will be described, where the data will be stored, and if and how it may be shared.
The content of a data management plan will vary across disciplines, but Science Europe's practical guide to data management outlines the following key points:
Description of Data, Data Collection, and Reuse of Existing Data
- How will new data be generated/collected and/or how will existing data be reused?
- What data (type, format, and volume) will be generated or collected?
Documentation and Data Quality
- What metadata and documentation (e.g., method descriptions or data organization) will accompany the data?
- What data quality control measures will be implemented?
Storage and Backup During the Research Process
- How will data and metadata be stored and backed up during the research process?
- How will data security and the protection of any sensitive data be ensured?
Legal and Ethical Requirements and Guidelines
- If personal data is processed, how will compliance with data protection and security legislation be ensured?
- How will other legal issues, such as intellectual property rights and ownership, be managed? What legislation applies?
- How will potential ethical issues be addressed, and what ethical guidelines are followed?
Data Sharing and Long-Term Preservation
- How and when will the data be shared? Are there potential restrictions on data sharing or reasons to place an embargo on the data?
- How will data for preservation be selected, and where will it be stored for the long term (e.g., in a data archive)?
- What tools or software will be necessary to access and use the data?
- How will the application of a unique and persistent identifier (such as a Digital Object Identifier (DOI)) for each dataset be ensured?
Responsibility for Data Management and Resources
- Who (e.g., role, position, and institution) will be responsible for data management?
- What resources (e.g., budget and time) will be allocated to data management to ensure the data will be FAIR (Findable, Accessible, Interoperable, Re-usable)?
For more information see DMP_checklist and Cessda Data Management expert guide.
HOW to write a data management plan
There are several tools available for creating data management plans. By using these, one can develop data management plans that comply with requirements from various research funders. These plans can also be shared and edited by others.
Recommended DMP tools:
- DMPonline from Digital Curation Centre (NB! Use the "Science Europe" template under Reference > Funder Requirements)
- Data Steward Wizard from Elixir (Particularly for life sciences, there are templates adapted for Norway that support exporting a machine-readable data management plan)
Other tools:
- EasyDMP from Sigma2/Sikt (adapted for Norway and support exporting a machine-readable data management plan)
- Data management plan from Sikt
- argos from OpenAire
For tips and assistance with completion, we recommend reading Cessda Data Management expert guide, ELIXIR's RDMkit, Science Europe guidelines or DCC's checklist.
The Research Council encourages open publishing of DMPs. This can be achieved by downloading the developed DMP from the tool used and publishing it on platforms such as Zenodo.
Examples of data management plans
Example DMPs and guidance from Digital Curation Centre
Curated collection of Horizon 2020 DMPs from University of Vienna
DMP Catalogue from LIBER Europe
In addition, there are many examples of data management plans available on platforms such as Zenodo or on the websites of DMPonline and DMPtool. Note: These plans are not curated or quality-assured in any way and should not be used as templates.
Frequently asked questions (FAQ)
What is research data?
Research data can be defined as all data produced or collected by the researcher in a research project. This includes information, particularly facts or numbers, gathered or generated to support claims in the literature, such as statistics, results of experiments, measurements, fieldwork observations, surveys, interview recordings, images, etc.
What are the FAIR principles?
The FAIR principles are a set of guidelines for open research data, standing for:
- Findable - data should be discoverable.
- Accessible - data should be easily accessible.
- Interoperable - data should be able to be opened, understood, combined, reused, and processed without restrictions, both now and in the future.
- Reusable - data should be reusable.
For research data to be reusable, its quality must be high. Therefore, both data and metadata must be findable, accessible, and interoperable.
Where can I archive and share my research data?
To make your research data visible and accessible, you should choose a discipline-specific archive. Such archives can be found on re3data.org, which is the largest and most comprehensive registry of available data repositories. An alternative is fairsharing.org.
If you cannot find an appropriate discipline-specific archive, as a researcher at the University of Bergen (UiB), you can archive your data in DataverseNO.
Please also see our page on Open access to research data.
What is metadata?
Metadata is structured information that describes, explains, locates, and makes it easier to retrieve and use a source of information. In short: data about the data. To help make your data reusable and accessible for you and others in the future, you must create and archive accurate metadata alongside your data.
The Digital Curation Centre (DCC) provides an overview of discipline-specific metadata.
Do I have to make all my research data openly available?
Not necessarily. You must decide which data to store, which data you want to keep, which data should not be openly available, and which requirements apply from funders, the university, and/or any legal or regulatory requirements.
If there are no specific requirements, we recommend researchers consider the following: Which data are necessary to reproduce or validate the results? Note that this may include code. Which data have potential for reuse by others?
The Digital Curation Centre (DCC) offers useful guidance: ‘Five steps to decide what data to keep’
Are there "best practices" for handling research data?
The following components are necessary for good management of research data:
- Use short and descriptive filenames
- Choose archival-worthy file formats
- Track different versions of your documents
- Create metadata for each experiment or analysis
See Deposit Guide for DataverseNO or DataONE Best Practices Primer for more information.
What are archival-worthy formats?
Archival-worthy file formats ensure that your data can be read by anyone, including in the future. Certain file formats are more likely to remain readable over time than others. Such formats are typically:
- Non-proprietary
- Open, with documented international standards
- Widely used in the research community
- Use standard character encoding, preferably Unicode (e.g., UTF-8)
- Uncompressed
When archiving your data, ensure you upload your files in an archival-worthy format in addition to the original file format. Also, make sure all your files contain a valid file extension, such as .txt, .pdf. More information and examples of archival-worthy formats can be found at the Dutch national expert center for archiving research data: Data Archiving and Networked Services (DANS)
