​Open data (sharing)

Sharing research results is an established academic practice, whether through publication or through more informal means with colleagues and collaborators. The increasing digitisation of research means that it has never been easier to share data on a more detailed level.

If you're planning a research project, it's worth checking whether there is already data available that you might be able to use. This may show up as part of a literature review, but there are a number of dedicated data archives and repositories which you can use e.g. re3data, Biosharing, UK Data Service, NERC data centres, Wellcome Trust guidance on data repositories, Dryad, Nature list of recommended data archives

If you're setting out on a research project, it's worth checking whether there are already data available that you might be able to use. This may show up as part of a literature review, but there are a number of dedicated data archives and repositories which you should take a look at too.

Benefits of Sharing

There are a number of reasons why you might consider sharing your own research data:

• Sharing supports research integrity by allowing the analysis to be easily verified
• Sharing can be a source of new collaborations, as your work is more discoverable
• Published articles whose underlying data are also published often receive more citations
• Published data can often be used in novel ways not expected by the original data creators, such as large-scale meta-analyses
• Where shared data are reused this can be used by the originator as evidence of impact, helping career progression
• Many funding bodies require data from funded projects to be publicly available where possible (e.g. RCUK, and EC policies on access to research outputs).

Restricting access to data

There are a number of reasons why you can justify withholding your research data. Withholding data means taking a decision not to openly publish them, even if there are obligations to funders or publishers to openly share the outputs of research.

Research funders' data sharing policies typically require that research data are made as openly available as possible, recognising that there may be legal, ethical or commercial reasons why access to some data needs to be restricted. These restrictions typically apply at all stages of a project so that the research process is not damaged by inappropriate release of data.

Access restrictions can change over time. For example, it would be unusual to share any data during the active phase of a project before findings have been published and continued access restrictions may be required to allow time for commercialisation. However, after the end the project and once patent applications have been filed, it may be possible to release the data.

Partial access: Access to data doesn't mean that everything has to either be openly available to the public or completely restricted. There are many types of access control that can be applied to all or part of your data, some of which will still allow you to share your data, but only with specific users under regulated conditions. This should always be considered in preference to a complete restriction on the whole dataset.

How to share data openly

When making your data freely open in in a repository (eg Worktribe outputs, type dataset) you should also provide some additional information to get the most impact from your work.

1. Assign a DOI. This is a unique ID for your dataset and can be included in you associated publications. The library can provide a DOI for you on deposit in Worktribe. You can also assign DOIs to dataset metadata records where the data is not openly available, but is finable by other reserachers.
2. If you have any other unique IDs for you or your work include this in the description of the dataset wherever possible eg ORCIDs and grant IDs, and apply appropriate creative commons licenses to clarify ownership and use of data.
3. Provide a Readme file with the dataset. The purpose of this is to make sure people who use your data understand it and use it appropriately. It also backs up the robustness of your methodology by allowing space for you to fully explain how the data was generated. Download Readme Files Explained

You can also do all of the above for dataset metadata records where the data is not openly available, which makes your work findable by other researchers.