The University of Southampton Research Data Management Policy has a requirement that all significant Research Data should be held for a minimum of 10 years and may be longer where the data is actively used. Funders also have retention requirements and some research data will also be subject to legal requirements.
The Digital Curation Centre (DCC) has a summary of the requirements by the major funders. If your funder is not on this list please use the Sherpa Juliet service to check if your funder has any requirements for research data.
It may be tempting to keep everything but that has drawbacks as it can be more difficult to find the truly important material. It is also worth remembering that research data can be subject to Freedom of Information (FOI) requests.
You may have your own view on how long you need or want to retain data. This will be influenced by the discipline you are working in, the type of data created and whether further work or publications will be based on it. Factors that may influence retention include:
Data which you decide not to keep should be destroyed securely.
Further guidance is available:
It is not a requirement that all research data must be held within the University. Discipline specific repositories and funder requirements may mean that research data will be held elsewhere. You should consider what services you may require to meet the retention requirements applicable to your data
You can find discipline-specific data repositories for your subject via Re3data.org.
If you are funded, check your funder policy for recommended data repositories e.g. ESRC and NERC support dedicated data centres. Also consider whether any agreements with your collaborators include requirements for data deposit.
Regardless of where you deposit your data, you need to add a dataset record to our data catalogue (via Pure)
This is a growing and fast moving area. Some publishers are now requiring the deposit of supporting data with the article, while others require that a link to the data is provided. You will need to take this into account when considering how long you will need to retain the data and may influence your choice of storage location.
You can deposit small datasets (gigabytes in size) in Pure for long-term storage. If your data is a terrabyte or over in size, contact email@example.com to discuss how to deposit your data. For further guidance see Depositing data.
You can request a DOI for your dataset to include in the funder acknowledgement and data access statement in your publications and we can organise this. Ideally you should request the DOI prior to submitting your manuscript. See DOI for Data.
You may be able to publish your data in a data journal with an accompanying data paper (a separate entity for any research papers based on the data).
Discipline specific repositories and funder requirements may mean that research data will be held elsewhere. You should consider what services you may require to meet the retention requirements applicable to your data.
The largest share of costs for data are incurred in preparation and ingest to the selected storage service, as shown by the costing tool provided by the UK Data Archive. Extended data retention periods may have some additional costs that will impact your project directly or they may be covered by the full economic costing included in your proposal. Invariably data retention periods will outlive projects, so you may want to consider how this will be funded as part of your data management plan and/or in your proposal – check with your funder’s guidelines.
Over time costs will be incurred for storage, typically based on the volume of data stored for a given retention period, and for additional services, for example active data management such as reformatting to counter possible format obsolescence. The latter is now regarded as less of a problem for popular formats, but may need to be considered for specialised data formats.
In some cases the costs and benefits of data storage and retention decisions may need to be assessed and justified for funding purposes. The KRDS (Keeping Research Data Safe) Benefits Analysis Toolkit may be used for this.
Research data represents an investment not just from the funder and the University but also by the individual researcher. However, as part of the deposit process you will be asked to consider what should happen at the end of the retention period and who is responsible for carrying this out. Under the University policy the review process will be the responsibility of the lead PI’s Faculty. (see also Secure Destruction of data)