The term data sharing can have different interpretations. It may refer to informal, restricted sharing within a research team—including external collaborators—or to open, public sharing of data beyond the original project’s duration.

Both of these types of sharing are supported by research data management practices such as Data management planning and Organising and describing data, but require additional tools as well.

1. Informal data sharing

One common type of data sharing is the collaboration that happens within research groups and projects as part of the day-to-day research process. Best practices for informal data sharing include:

  • Ensuring that data are stored securely and that only authorised individuals have access.
  • Establishing clear guidelines for data sharing and ensure appropriate agreements are in place when sharing with external partners.
  • Using recommended data-sharing tools wherever possible.

Information about data sharing and collaboration services provided by ICT – including tools for sharing data with third parties and for transferring large volumes of data between Imperial College London and external sites - can be found on these web pages: Sharing and collaboration tools and  Sharing files.

2. Depositing data in a public repository

It is also increasingly common to share data more widely as a form of publication. The best way to make data widely available and reusable is to deposit it in a public repository (also known as an archive or data centre). This also meets requirements for long-term data preservation. Most repositories assign a Digital Object Identifier (DOI) or other unique identifier to your dataset, enabling it to be cited and tracked like a regular publication.

Deposit your data in a disciplinary or subject-specific repository where one exists. Some funders provide their own repositories or recommend preferred options. If no specialist repository is available, consider using a generalist repository such as Zenodo, Figshare, or Dryad.

For help choosing a suitable repository, see our webpage Finding a data repository

See our Data deposit checklist for guidance on how to prepare data for deposit with a repository.

3. Sharing sensitive/confidential data

Not all data can be made publicly available. There may be valid ethical or legal reasons for restricting access (e.g., protecting sensitive or confidential information).

Some repositories (e.g., the UK Data Service) provide tiered access levels that allow datasets to be shared under controlled conditions. Users typically need to register and sign a data-sharing agreement. For highly sensitive data, access may require approval from a committee or designated contact person.

The Centre for Open Science maintain a list of approved protected access repositories

Guidance on how to responsibly share sensitive data can be found on our web page Sharing sensitive data.

4. Supplementary information

If the dataset is very small, some journals may accept it as supplementary information and host it themselves. However, sharing data as supplementary files is generally not recommended because:

  • Discoverability may be limited, making files harder to find
  • There may be restrictions on file size and format
  • There is no guarantee of long-term preservation
  • Supplementary files rarely receive their own DOI, making citation and long-term tracking difficult

A better option is to deposit the data in a suitable repository to ensure long-term preservation and accessibility.

5. Available on request 

A common method of data sharing has been to make data available upon request, but this is not recommended because:

  • Requests may go unanswered
  • Responses may be delayed
  • The data may no longer be available
  • The contact email may become inactive

Again, the better option is to deposit the data with a suitable repository. This method should only be used if no other option exists—e.g., if you have sensitive data that cannot be made publicly available and you do not have the option of publishing with a repository that offers restricted access.

Wherever possible, use a shared email address for your research group or department rather than an individual address.

6. Writing a Data Access Statement

All articles published by researchers affiliated with Imperial should include a Data Access Statement. This is now a requirement of many funders and publishers. The UKRI open access policy requires all research articles to include a Data Access Statement even where there are no data associated with the article.

Your Data Access Statement should contain the following information about the data underlying the article:

  • How the data can be accessed: Where it can be downloaded or who must be contacted to request access. This should always include a web link (preferably a DOI or other persistent identifier), or—only in rare cases—an email address
  • The terms under which the data can be used: Any embargo in place, whether a general licence applies, or whether a data-sharing agreement must be signed before access can be granted

If you cannot share your data, you must still include a statement briefly explaining why.

Examples of Data Access Statements can be found on our web page How to write a data management statement.