If you are required to share your research data by your funder, or you are interested in making your data openly available, there are types of data which can be shared. Generally, this falls into four categories
1. Data which can be used to validate your results. This is sufficient data such that a researcher at a different institution is able to download, understand and re-analyse your data using the same techniques that you used. It is recommended that you share as a minimum, the data which underlies any figures, tables or charts in your publications.
2. Raw data, which could potentially be aggregated with other similar datasets and stored in a database for re-analysis.
3. Data which you do not intend to use for a publication, but which may still have some value if shared e.g negative research results.
4. Physical data - This data may or may not underpin your research publications. As this data is physical, it will not require sharing on a online repository. However, a safe storage location would be required and instructions on how to access it should be made available online.
All RCUK funded research papers require a Data Access Statement, whilst it is also recommended for good practice in the Aston RDM Policy. A dataset DOI (Digital Object Identifier) should also be included in your research publications. This Data Access Statement is required to be in your publication prior to submission. For this reason, the metadata describing your data needs to be uploaded onto Aston Data Explorer prior to submission. The data itself can be uploaded onto Aston Data Explorer anytime prior to online publication.
Some funders also require data which you do not intend to use as a part of a publication to be shared. For the EPSRC, the metadata should as a minimum be available at a maximum of 12 months following data creation or 12 months following the end of the research grant.
Data can be deposited in Aston Data Explorer and stored for a period of 10 years from last access. You can log into Aston Data Explorer using your university login credentials. See the below attachment for a tutorial. Alternatively, data can be stored in another open access repository. See Storage and Archiving for examples of repositories.
There are certain types of data which should either never be shared, or only shared under strictly controlled circumstances. Below is a list of types of research data which in the majority of cases should not be made openly available for sharing. Please note, even if your data is not shared due to one or a combination of the reasons below, some funders (e.g RCUK) have additional requirements which must still be met such as the Data Access Statement
1. Commericalisation - If data is potentially to be commercialised and/or have a patent application is made, the researcher must contact the RKE (Research & Knowledge Exchange) in the first instance. If following a successful patent application, a researcher would like to share their data and/or publish their research, they would need to contact the RKE in the first instance to see whether this would in any way be detrimental.
2. 3rd Party Data - For 3rd party data which was used in your research, a metadata link to the dataset can be uploaded onto Aston Data Explorer. Likewise, a metadata only link can be made to your research data stored in an external repository, whether or not it was created whilst working at Aston (for example, see http://researchdata.aston.ac.uk/363/). If you wish to upload the actual data, permission must be sought from the owner of the data.
3. Ethics - Must meet GDPR requirements. Data which contains personal information can be shared if informed consent has been obtained for it to be collected and shared. The consent must be freely given, specific and unambiguous (for example, see https://doi.org/10.17036/researchdata.aston.ac.uk.00000257). Under certain circumstances, data can be shared without consent being obtained such as when it is fully anonymised, where it would fall outside of the scope of the GDPR. See the UK Data Archive advice
4. Commercial Funders - Data which has been produced in partnership with a commercial funder should not be shared
5. Security Risk - Data which if shared can pose a security risk (e.g terrorism, the location of a particular rare animal specie) should not be shared
6. Cost considerations - If it would be unreasonable to store a dataset which is very large e.g over 1TB, it should not be shared. Alternative means of sharing however could be pursued, such as sharing a partial dataset or data/metadata which will allow the re-creation of the dataset.
A license type must be selected which determines how your data can be reused following sharing. This is undertaken on Aston Data Explorer. To maximise the impact of your research, it is advised that one of the more liberal licenses is selected. If needs be, more restrictive licenses can be applied to more sensitive data.
Image courtesy of fotor.com