Why FAIR data management?

The FAIR principles were introduced in 2016 to provide guidelines that improve the findability, accessibility, interoperability, and reuse of digital assets. Maastricht University (UM) recognizes the importance of FAIR within the context of Open Science.

  • These principles emphasize machine-actionability, which means that computational systems can find, access, and interoperate data. For example, search engines can easily find data sources that are FAIR as opposed to data sources published in personal websites.
  • Making data more easily available for discovery and reuse for the wider research community.
  • For individual researchers, it increases the visibility of their research, opens up new collaboration opportunities, and ensures compliance with the requirements of funding agencies and journals.
  • For society at large, it fosters public trust in research, facilitates data reuse in the private sector, and thereby boosts innovation.
How to be FAIR?

It’s easier than you think, you can follow these resources:

  • The FAIR principles explained

Video credits: Martínez-Lavanchy, P.M., Hüser, F.J., Buss, M.C.H., Andersen, J.J., Begtrup, J.W. (2019). ‘FAIR Principles’. In: Holmstrand, K.F., den Boer, S.P.A., Vlachos, E., Martínez-Lavanchy, P.M., Hansen, K.K. (Eds.), Research Data Management (eLearning course). doi: 10.11581/dtu:00000049

  • The adoption of the FAIR principles in terms of 6 RDM processes

 

Take a look at the FAIR courses

Prepare data for publication

When getting your data ready for publication, think about licenses, documentation, repositories, and whether your research benefits from databases and data APIs

Choosing a License or Terms of Use

Why it’s important:

  • Licences inform users about permissible uses. For example, a license once prevented a commercial entity from monetizing a public research dataset.
  • Licences act as a legal fingerprint. For instance, a researcher could prevent the unauthorized adaptation of their dataset in secondary usage.

Consider:

Provide Documentation

Why it’s important:

  • Makes data discoverable and understandable: for example, a researcher from a different field can easily grasp the insights of a dataset if it’s well-documented and has good metadata.
  • Prevents misuse or misunderstanding: for example, a well-documented dataset on plant extracts prevented a toxicology error in a subsequent research project.

Consider:

  • Opting for standardised metadata formats, like Dublin Core or DDI. 
  • Describing every dataset component: variables, units, sources.
  • Use of non-proprietary formats such as .csv, .txt, .json. 
  • UM RDM Code of Conduct.
  • MUMC+ guides 

 

 

Make use of Institutional Repositories

Why it’s important:

  • Ensures longevity and discoverability.
  • It boosts academic recognition: datasets, are discoverable in scholarly aggregators (e.g. SCOPUS) when they are deposited in recognized repositories.
  • It is compliant with security recommendations.
  • The dataset gets a unique identifier (DOI).

Consider:

Using Databases and Data APIs

Why it’s important:

  • Databases and APIs can significantly enhance research accessibility and efficiency. For instance, complex research might benefit from a database, streamlining data access and analyses.
  • However, while they offer numerous advantages, they also come with some risks in terms of security and data privacy.
  • Nonetheless, they promote collaboration. A well-documented database or API can enable global researchers to swiftly engage with your data, boosting collaborative efforts, while keeping the overview of who access and when.
  • Still, having a database assumes that it is hosted on a server. Thus, it requires maintenance.

Consider:

  • Having a database or an API for research data is the most efficient way to ensure compliance and have an overview of data access. Yet, it might not always be the solution. Evaluate if your research project genuinely demands it.
  • Local ICT faculties might offer hosting but be aware of the inherent responsibilities. Self-hosting can increase risks if not properly managed.
  • A database, especially with a public IP, demands security measures. Even minor oversights can lead to significant breaches.
  • FAIR Principles: while platforms like https://smart-api.info/ can help make your API more FAIR, ensure that the data you’re making findable and accessible doesn’t infringe upon privacy regulations.
Data Availability Statements

A data availability statement (or data access statement) is a brief explanation included in publications describing where the data (or software) associated with the publication is available, and under which conditions it can be accessed. Data availability statements can also be provided if the manuscript is based on secondary data, does not rely on data, or if the data cannot be made available, for example, for privacy or ethical reasons. The statement should then include the reason for this.

Why it’s important:

  • A data availability statement promotes transparency, allowing others to validate and reproduce results.
  • It demonstrates your commitment to responsible research and increases the visibility of your data by linking the manuscript with it.
  • Publishers and funders may require a data availability statement.

Consider:

  • Data availability statements should be provided at the submission stage of a manuscript. Either under a specific “data availability” or “data accessibility” section, or within the “acknowledgements” section.
  • The data availability statement should answer three main questions:
    • Is the data available? (if not, why?)
    • Where is it available?
    • How can it be accessed?
  • It is best practice to provide the DOI associated with one’s data entry in a data repository such as DataverseNL or the DANS data stations. Providing a personal email address is not appropriate.
  • Guidelines and examples on writing data availability statements: Taylor & Francis, Springer Nature, University of St Andrews, Yale University
Contact the data stewards for support