Where possible, you should publish:
- the data that supports your published findings
- any code needed to recreate your results
- documentation that explains how the data were collected or processed.
Publish what you can
Publish your data in a manner that is 'as open as possible and and as closed as necessary'. If the data are sensitive and cannot be published, consider what aspects of your dataset can be published. For example, you may have raw data comprising responses to a questionnaire. In this case, the raw data from the participants may need to remain closed for ethical reasons. However, you can still publish the questionnaire and related metadata so that others can see what you did and build on your work.
Standalone datasets
You may have a large corpus of data relating to a specific project that you wish to share. You may also have a dataset that is not attached to a particular project but you wish to share for other researchers to explore, reuse and build on. Publishing such data into a repository will form a major research output in its own right. This would be especially so if you publish your data in repositories or data journals. This would allow others to cite your dataset.
Raw data, software and code
As a general principle, publish the raw data and/or code needed to recreate your findings and to confirm the integrity of your research. This may be quantitative data, qualitative data, images, movies, texts, and so on. You are encouraged to share data as widely as possible.
Cost implications
Depending on your data type and volume, it is important to consider that publishing your data may come with a cost. The best way to anticipate any costs associated with data management and sharing is to develop a data management plan. It is best to develop one early on in the research project’s lifecycle. This will allow you to budget for any data storage, management and sharing costs in grant applications.
University of Cambridge expectations
The University expects the sharing of any data that supports published findings. The data sharing should be as wide as possible. This is in line with the University's Research Data Management Policy Framework.
Funder and publisher requirements
Some funders also expect grant recipients to share the data from the funded work. Many publishers will now ask authors to share the data that supports the publication. At the very least, these are the data that you should publish. They often require that you publish your data in appropriate data repositories (discipline-based or institutional). Such data should be assigned persistent Uniform Resource Identifiers (URIs), such as Digital Object Identifiers (DOIs). This will increase the discoverability of your data. Funders and publishers often require Data Access Statements in journal articles that explain how the data can be accessed or why it cannot be published.
Benefits of publishing your data
There are many reasons why you should publish your data. Please see Why publish your data. We also have guidance on how to reuse data and make your data reproducible.
How to publish your data
Please see our guidance on How to publish your data.