When you are ready to share your data, either externally or within the research community, you will need to think about depositing it in a repository. This is often referred to as archiving. Data that cannot be shared should also be archived in a repository as this will ensure it is preserved for future reference, made available for research validation, held securely, backed up appropriately and is discoverable.
Where should I deposit my data?
Your journal may have data sharing requirements, either to deposit your data as supplementary to an article or paper, or to make your underpinning data available in a subject specific or institutional repository. Some funders provide a data repository, others require sharing but do not stipulate where. For more information on Funder's policies click on the following link,
Deposit in Discovery (Pure) or externally?
The University repository Discovery is available to all research staff and postgraduate research students. A key aim of data sharing is to make your data discoverable and reusable. There may be a repository that is better suited to your data than Discovery. You can search for a repository at the Registry of Research Data Repositories.
If you choose an external repository you should create a record of your dataset in Discovery (Pure) that can link to the hosting repository.
How to deposit your data with Discovery
You can upload your data directly to Discovery (Pure) - however, if individual files exceed 20GB you should get in touch with email@example.com in advance.
To deposit directly in Discovery
On your Pure personal overview page, you will find a "Datasets" option is within the left hand menu. When you click on the plus sign "+" button a new template will open allowing you to add details of the dataset.
All mandatory fields are marked with a red asterisk; these are Title, People, Dataset managed by, Publisher and Data made available. The record will not save unless these are filled in. Add further information in any other fields that are relevant.
In the Data availability section you can add your data files to Electronic data. All file formats are accepted and multiple files can be uploaded at the same time.
You will be asked to select an appropriate Data Licence from a drop down menu and the Type of data you are depositing (e.g. Image, text) to all files. Press OK.
In Relations to other content please link your dataset to relevant Publications, Projects or other Datasets in Discovery.
Finally, at the bottom of the page under Status, save the record as "For Validation".
You will receive a Digital Object Identifier (a DOI is a persistent URL) for your dataset in due course. If you need a DOI to reference your dataset in a Data Access Statement within your publication in advance of depositing your data contact us at Discovery@dundee.ac.uk
For more information on Data Access Statements follow this link,
There is a more detailed guide on How to submit your dataset to Discovery (pdf)
If your data include personal data, special categories of personal data, or security sensitive information then do not send them via email. You can coordinate deposit with the research support team using Box. Once we have your data we will create a record and deposit on your behalf.
You may not be able to safely share data with content that is sensitive - however there is a value in archiving such data. When you prepare your data management plan and ethics application you will be asked to outline your plans for collecting and safeguarding the data you collect.
The safe destruction of information should also be considered. There is University guidance on the proper disposal of information on the University web pages.
When designing your research you should seek permission from research participants to grant access to other researchers where possible. This might be anonymised data, and it might be under controlled access - where researchers are required to make a data access request. E.g. the UK Data Service make controlled data available to accredited researchers following approval of their data usage by a Data Access Committee. Users are given training prior to access.
If personal data are identifiable or pose a risk to individuals even after anonymisation, it may not be possible to provide access to that data. The data are however retained by the repository or institution as they may be used in support of a research integrity investigation.
For further guidance on data protection visit the University Data Protection pages where you will find contact details for the University Data Protection Officer.
Commercially sensitive data may be released only under restrictions set out by a commercial partner.
What data do I deposit?
There is a value in all your data - even negative results. However, there will most likely be some preparation work required to make data reusable. This could be the organisation of data with an accompanying ReadMe file to help users understand how to utilise the data. It could also include information about or access software used to process the data.
Depending on the length of the project and the volume of data collected this could take some considerable time. At a minimum researchers should share the data that underpins the research findings they have published in an article, paper or report.
You may want to publish more than the minimum requirement for data sharing in which case you can publish a large data set and assign that a DOI for citation. Subsets of that data that relate to specific papers or elements within papers can be assigned their own DOIs and published as a separate but linked record.
What if my datset is very big?
There is no limit on the size of data the University will publish, but datasets that run into multiple Terabytes will require special arrangements. Get in touch with the research support team at firstname.lastname@example.org for further guidance and support.
Data published in external repositories
The University Policy to Govern the Management of Research Data requires that a central record of known datasets deposited in public access repositories will be maintained by the LLC. As such any research data deposited with external services, such as Research Council or international repositories, should be made known to the LLC for recording in the University's research information system Discovery (contact email@example.com). This is to allow the University to ensure data continues to be discoverable and compliant with funder mandates.
The University policy to govern the management of research data is available on the University web pages.
The dataset record created in Discovery will link to the data in the external repository.