Support and training
Contact the library:
Henrik Holtvedt Andersen
Research data throughout the project
Advice for research data management in different project phases.
Planning
At the start of the project, review and assess the following:
What data types are you collecting?
- Observational data are recordings or descriptions of phenomena. The need for preservation is typically high because such data are difficult to reproduce.
- Computational data are data from models, simulations, and other calculations. Whether input and/or output data, as well as the algorithm, should be preserved will vary.
- Experimental data are results from controlled experiments. If the conditions of the experiment cannot be recreated, there may be a need to preserve the data.
- Source data include documents, recordings, registers, and other sources that become research data when used as foundational material in research. Often, these are already sufficiently well preserved elsewhere, and it is sufficient to document where they are located.
How open or closed the data should be stored, both under and afterwards
- Open or Freely Accessible (Green):
Used if disclosing the information to unauthorized persons or unintended modifications do not cause harm to public interests, the university, individuals, or partners. - Restricted (Yellow):
Used if disclosing the information to unauthorized persons or unintended modifications could cause some harm to public interests, the university, individuals, or partners. - Confidential (Red):
Used if disclosing the information to unauthorized persons or unintended modifications could cause harm to public interests, the university, individuals, or partners. - Strictly Confidential (Black):
Used if disclosing the information to unauthorized persons or unintended modifications could cause significant harm to public interests, the university, individuals, or partners.
More about managing personal data
Classification of data from the University of Oslo
Data management plan (DMP)
A data management plan describes how research data will be handled throughout a project and how openly they can be shared after publication and upon project completion.
It includes the following information:
- About the DMP
- About the research project
- Legal and ethical considerations
- Storage and protection of data during the project
- Reuse or collection of data
- Data documentation during the project
- Processing, analysis, and interpretation of data
- Archiving and publishing of data
- Responsibilities and resource requirements
It is up to you how you design a data management plan (DMP), but there are templates available to help you, such as Sikt’s template for creating a data management plan.
The Research Council of Norway, the EU, the ERC, and other funding agencies require the submission of a data management plan.
Organizing research data
In a research project it is important to have descriptive file names, standardized and non-propietary file formats, and a good file structure.
UiT has made a detailed guideline of how to organize your data.
Assessment of Storage Solutions and Access Control
The size of the dataset—both in terms of file size and the number of files—affects which storage solution you should choose and how the data should be organized and documented throughout the project. It can also impact the project workflow in terms of quality assurance, file transfer, access control, and analysis.
Based on the analysis plan, data size, project complexity, and other factors (e.g., planned fieldwork), assess whether there is a need for equipment, additional storage space, software, or other resources that you do not currently have access to. Contact IT for support with this and refer to our storage guide.
If others besides yourself need access to the data during or after the project, carefully consider who should have access to what and how they can obtain it. This is especially important when the data contain sensitive or classified information.
Implementation
During the project, it is important to have good control over storage to ensure data quality and avoid data loss.
Organisation and documentation
It is important to document how you handle your research data throughout collection, organization, analysis, and archiving as part of good research practice.
An example of tools that you can use for documentation are
- a clean text document (non-formated, non-proprietary)
- notebook like OneNote
When archiving data, documentation should be included in a ReadMe file. A ReadMe file is a detailed guide to the dataset, enabling other researchers to understand and reuse the archived data.
Here is an example of a template for a ReadMe file.
Closing
What should be made available following the principle "as open as possible, as closed as necessary."
Description of your research data
Documenting Provenance via a ReadMe File: Specify where and by whom the data were collected, described, and archived.
A ReadMe file is a detailed guide to the dataset, enabling other researchers to understand and reuse the archived data.
Here is an example of a template for a ReadMe file.
File formats
Data has to be exported to archive-friendly open formats. About preferred file formats: uit.no/dataverseno/deposit/prepare/
Lisences
Lisences should be internationally approved, machine and human readable, and put as few restrictions as posible on open access, reuse and distribution of the data.
We recomend the license CC-BY 4.0 if possible. That means that the data kan be reused, modified and also used commercially as long as they are properly cited.
Archiving
Publishing research data and associated metadata in a research archive ensures long-term preservation and reusability. Research data should be made available no later than the publication of the scientific article.
Choice of archive
A discipline-specific archive will make your data more visible to other researchers in the field. If no suitable discipline-specific archive is available, we recommend using a general archive.
The archive should facilitate effective retrieval by adhering to the FAIR principles.
If you want to learn more about choosing an archive, we recommend OpenAIRE's guide.
Discipline-specific archives
You can look for a fitting discipline-specific archive at www.re3data.org
General-purpose archives
- DataverseNO – A national archive for general research data from all disciplines. MF has its own collection, curated by MF. It is CoreTrustSeal-certified as a reliable and sustainable archive ensuring long-term preservation of archived data. Each dataset receives a DOI and an automatically generated citation for publications. Data from the archive are searchable in services like Google Dataset Search. The archive also includes version control, registering and displaying all changes made to the dataset after publication. More information and user guides can be found here: DataverseNO Info Page.
- Sikt – A national archive managed by Sikt (Knowledge Sector Service Provider). It focuses on preserving social sciences and humanities research data, as well as some data from medical and health research. NSD curates archived data. For data containing personally identifiable information, Sikt’s archive is recommended.
- Zenodo – The EU’s research data archive, allowing the archiving of data and other documents from all disciplines. Does not curate archived data.
Reuse
Much research is based on data from existing sources.
Data use from existing sources
Source data include documents, recordings, registers, and other sources that become research data when used as foundational material in research. Often, these are already sufficiently well preserved elsewhere, and it is enough to document their location.
When reusing data from other sources, it is important to provide information about:
- Where the data come from
- Which version was used
- How the data were collected
- Who is responsible for the data
- Whether the data need to be cleaned or processed before use in the project
- Other necessary information to ensure result reproducibility
Lisence agreement
It is important to know what type of lisence or other agreement applies for the data and how they can be used.
Published text is often subject to copyright, which may restrict further sharing of the material. If the material has a CC license, reuse and sharing are easier. For other materials, permission from the publisher or the author is required for republication. Some publishers allow reuse with restrictions, such as for teaching and research purposes within the institution. In such cases, the material cannot be archived openly.
For material from Norwegian online sources, it may be useful to contact Nettarkivet at the National Library of Norway.