How to publish your research data

In the last few years it has become standard practice for research funding bodies, institutions, and publishers to request or mandate that the research data underpinning your research outputs are also published and made available for validation and re-use by others.

The ‘reproducibility crisis’ widely reported in the media in 2015 – spurred by a paper published in Science which estimated that the findings of only one third of scientific papers are replicable – has accelerated this push towards open research data publication to increase the transparency and integrity of the research process and results.

Open research data publication and sharing has also been driven in the UK by the UKRI Common principles on research data (2016), which recommend that research data be made openly available with as few restrictions as possible in a timely and responsible manner.

But there are also significant personal benefits to publishing and sharing your research data: a recent study (2020) found that journal articles that included data access statements that linked to datasets had up to 25.36% (± 1.07%) higher citation impact on average.

So how do you publish your research data? And how can you publish your research data while protecting any legal or ethical obligations you may have towards your collaborators or participants?

 

Ten-step guide to publishing your research data

 

1. Decide which data you need to publish

You will not need to publish all your research data, but you are typically expected to publish the minimal dataset that supports the central findings of a published study. That is, the research data or materials that evidence your research process, substantiate your thesis, or which would be needed to replicate your findings.

2. Choose an appropriate repository

An easy way to preserve, publish, and share your research data is to deposit a copy of your research data or research materials in a research data repository.

You should check whether your publisher  or funder recommends a particular repository. For example, Scientific Data (Nature journals) maintains a list of recommended data repositories, PLOS hosts a list of recommended repositories by discipline, and a number of journals support the use of the cross-disciplinary repositories Dryad, Figshare and Zenodo (a repository that specialises in preserving software and code). Research data funded by the ESRC should be deposited in the UK Data Archive; and research data funded by the Wellcome Trust should be deposited in one of the Wellcome Open Research-approved repositories.

If your publisher does not recommend a particular repository, or if they demand a Data Publishing Charge (DPC), you can find a research data repository relevant to your discipline at the Registry of Research Data Repositories (re3data). National data centres or discipline-specific repositories are the best to use because they have the expertise and resources to deal with particular types and sizes of data.

3. Prepare your data for deposit

You will need to organise your files in a meaningful structure, convert your files to open or common formats wherever possible and whenever mandated by a publisher, and you will need to anonymise any sensitive data for which you have legal or ethical consent to share.

4. Document and describe your data

You will need to create metadata to accompany your deposited datasets in line with any repository or publisher metadata requirements.

5. Decide how open or closed your data will be

You will need to check that research data is published in compliance with any terms of participants’ consent and any copyright or intellectual property right-holders of the material. Ideally, consent for some form of data sharing will have been agreed before any research commenced.

While you must not publish data openly if you do not have the right to do so, there are several options for sharing data securely via a repository. For example: you can set access conditions; you can publish only a subset of the data; or you can publish a metadata-only record that includes a description of the data but not the data itself. Read our guidance on solutions for data sharing on our webpage restricting access to data.

6. Upload your data to your chosen repository and generate a Digital Object Identifier (DOI)

A DOI, or Digital Object Identifier, is a unique identifier assigned to an online object which acts as a persistent identifier or handle to permanently identify and link to it on the web. It can be used to track data citation metrics and to link related outputs such as journal articles, research data, and metadata records, thereby increasing your research’s visibility and impact.

7. Select an appropriate re-use licence

Creative Commons Licences are a type of open licence that you apply to your work to indicate you are happy to share the work under certain conditions. To help you decide which licence to choose, you should check any funder or publisher requirements, and read our guidance on copyright for researchers and licensing your work.

8. Add an embargo

This will set a date for the publication of your research data on the public interface of the repository. An embargo will protect your data as a ‘closed’ deposit until your article is published, or until any period of exclusive use of the data expires. Some repositories will also allow you to share the data anonymously with peer-reviewers during the embargo period.

9. Link the data to its associated publication

Include the DOI of the dataset or metadata-only record and a statement describing any legal, ethical, or commercial constraints on access to the data in your article’s data access statement. Once your article is accepted and a DOI for your article has been issued, add the article’s DOI to the metadata record of your dataset in the repository.

10. Register your dataset as an output in the VRE

Whenever you complete a research output you must add a record of it to your researcher profile in the Westminster Virtual Research Environment (VRE). After logging in with your University credentials, select ‘Add output’ > ‘Text-based output’ > ‘Dataset’, and complete the form as prompted.

 

More information

 

You can also publish your data as supplementary material to a journal article, in which case, you should consult the journal’s guide to submitting supplementary data. Alternatively, you can publish your dataset or a description of your data – a metadata record – in a data journal, a type of journal that specialises in publishing peer-reviewed data sets. To find an open data journal relevant to your discipline, search the list of peer-review open data journals hosted by the University of Edinburgh.

You can find detailed guidance on finding, assessing, choosing, and using a suitable research data repository to publish your research data on our Research data repositories webpage.

You can also contact the Research Data Management Officer for advice and support at research-data@westminster.ac.uk