FAIR principles

FAIR stands for principles describing good practice in the management of research data. Data should be as findable, accessible, interoperable and r(e)usable as possible.

The FAIR acronym encapsulates four principles that apply to all research data and should always be taken into account in data management. The Ģ��ֱ��'s Research Data Policy is built around these principles, as are the requirements of different funders for research data. In the data management plans, it is often necessary to take a position one way or another on how the FAIR principles are implemented in one's own case.

In essence, the FAIR principles aim at data that are as machine-readable as possible, and that can be processed by machines using the (meta)data available without human intervention. However, FAIR is a spectrum rather than a binary: for each data set, the aim is to achieve the best possible result for that particular data set. In many cases, it is enough to have a human element: a human being must be able to find, access, understand and use the data. This can be achieved in many different ways.

Research data should be discoverable. The world needs to know it exists.

Carefully describe your data in Converis, the research information system.
At a minimum, publish your the metadata in the university's JYX repository. This way the data exists in the eyes of outsiders. Metadata published on JYX will be given a DOI identifier that you can refer to in publications. Published descriptive data will be indexed in e.g. Google Dataset Search.
If you can publish the data yourself, please do so in a proper discipline-specific data repository or in JYX. Publishing the material or its descriptive data on your own website is rarely sufficient for genuine discoverability.

Tutkimusaineiston pitäisi olla saavutettavissa jollakin laadukkaalla tavalla.

By selecting a quality data archive, you can ensure that the data is easily downloadable and its metadata is accessible, even if the data itself is not.
Accessibility can vary: direct contact with the researcher, requesting access from the data repository, or open access, where files are directly downloadable.
Access may be restricted for good reasons, such as embargoes or confidentiality of the research data.
Where access is restricted, the service and technology must guarantee security and data protection.

Data should be in a format that is easy to use and understand.

Use file formats that can be opened with commonly available software.
Prefer file formats and ways of presenting information that make it easily accessible and editable.
If your field uses established vocabularies, standards or, for example, variables, use them.
Describe your data carefully. Save the description in a README file with the data.
The description should cover information related to the comprehensibility and usability of the data to the extent that another researcher can understand and use the data only with their help.
For interoperability of the descriptive information related to discoverability, use an appropriate data repository
Quantitative data can be presented in a database whose contents are in machine-readable format.

The conditions for reusability of the data should be made as clear as possible.

It is a good idea to give the data a licence, which explicitly states how the data can be used, whether it can be modified, whether it can be redistributed, etc.
Prefer established licences (e.g. Creative Commons for data, MIT or Apache 2.0 for software/code).
Make it clear who produced the data, how to refer to them or to the data, and who may have different rights to the data.

Ģ��ֱ��

FAIR principles

F - Findable

A - Accessible

I - Interoperable

R - Reusable

For more information on the FAIR principles, see for example: