Research data management
What is research data management?
There are many different facets to Research Data Management (RDM). It involves the effective planning, collection, organisation, storage, preservation and, where possible, the sharing of data both during and beyond the lifespan of a research project.
Funder policies
The Research Council Common Principles on Data Policy states that research data should be “made openly available with as few restrictions as possible in a timely and responsible manner.” The data policies of each individual research council are broadly aligned with these common principles and should be interpreted within the framework offered by the RCUK, now UKRI, in their Guidance on best practice in the management of research data. Typically, the policies outline the funder’s expectations with regards: data management planning; data sharing and access; time limits for deposit and retention; how funding can be used to offset RDM costs and compliance monitoring.
For concise summaries of the policies and links to detailed policy documentation of the major funders in the UK and EU, please visit the DCC website. The SHERPA / JULIET database also provides current and comprehensive information about the data archiving and data sharing requirements of all research funders.
Data management planning
Successful RDM is a process that starts at the initial research planning stage with the formulation of a data management plan (DMP). This document describes how you will handle the data generated in the course of your research project and what will happen to it afterwards. Most major funding bodies require the submission of a DMP as part of the grant application: this is mandatory for six out of the seven research councils, although the other, EPSRC, expects a DMP to be in place for all projects awarded funding.
The content of your DMP will vary depending on your research context and the information required by funders. Your DMP is not a static document and should be reviewed and updated throughout the research project. Therefore, when writing, then implementing, a DMP you should:
- Pay close attention to the data policies of relevant funders and/ or publishers, as well as the NTU RDM Policy and the DCC guidance for best RDM practice.
- Use the web based tool DMP Online to help you write your DMP. This offers templates for specific funders, step-by step guidance, as well as options to share and review your plan.
- Refer to the Checklist for Data Management Plan produced by the DCC. This offers a comprehensive overview of the different elements of RDM that you should consider and plan for.
Contact the Library Research Team for further advice on compliance with university, funder and journal data policies, and the development of your DMP.
Creating and organising research data
Legal and ethical guidelines
NTU works continuously to embed research integrity within its institutional research culture. It is essential that you maintain the standards for research data set out in our Code of Practice for Research.
Personal, sensitive or confidential data warrant additional safeguards. Any research involving human participants requires the correct handling of all associated data in line with NTU’s Research Ethics Policy and procedures, as well as any ethical guidelines issued by funders. In addition, the General Data Protection Regulations (GDPR), that comes into effect 25 May 2018, govern the processing of personal data and the circumstances under which that data can be shared. The ICO offers exhaustive and up to date guidance on GDPR.
The UK Data Service provides clear strategies for the gathering and processing of research data, including:
Adopting these will help you to maintain the integrity of your data, within the legal and ethical frameworks, without precluding opportunities for data sharing.
Active research data storage
You need to identify the most appropriate and robust storage solutions for your research data during your project. Use NTU Research Data Storage: a guide to help inform your decisions about the storage that is most suitable for the type of data your research is generating. You should back up your data on at least two forms of storage. You may also find the MANTRA Data Storage and Security Tutorial a useful tool.
NTU OneDrive is available for data storage during the active phase of your research. This offers:
- 5TB of data storage
- Access to your data anytime, anywhere using multiple devices
- Automatic file syncing across all of your devices
- Sharing and collaborative functionality, so you can share files and set permissions for multiple people to edit documents
- GDPR compliant data privacy and security measures
- Version history and file recovery up to 90 days, so you can restore previous versions of documents or retrieve deleted files
However, this cloud-based service is not appropriate if your research involves sensitive or confidential data. In this instance, you should apply for access to NTU network data storage. You may also require alternative data storage arrangements if:
- You are involved in a collaborative research project
- The amount of data generated is expected to exceed 5TB.
When your research meets one, or more, of these criteria, please complete the Active Research Data Storage Request Form and email it to the LIBResearchTeam@ntu.ac.uk
Organising and securing your data
You can take several practical steps to not only mitigate the risks of data loss and corruption, but also maximise research efficiency. You should implement:
- Agreed file structures
- Consistent file naming conventions
- Version control
- Data encryption for sensitive or confidential data.
Additionally, your data should be accompanied by documentation that details things such as the filing and naming protocols you have employed. This is particularly important if you are working as part of a collaborative research project. It will also help other researchers understand your data more fully if you plan to make data available for future reuse. For further advice about these techniques go to the ‘Format your data’ and ‘Document your data’ sections of the UK Data Service website.
Can, and should, you share your data?
You should archive your data in an appropriate repository so that data that underpin your outputs are assigned a persistent identifier and preserved for at least ten years. You have a range of options for archiving and sharing; you do not simply have to decide to make your data open, or to keep it closed. Ultimately, your funder’s position on data sharing will influence your decision about where and how to share your finalised data. However, planning carefully for data sharing from the very outset, will give you flexibility in controlling access to the data and its future reuse, whilst remaining compliant with funder policies. Your options include any, or a combination of:
- Imposing an embargo period, if, for example, you need to continue to exploit your data for either publication or commercial purposes
- Giving different levels of access to different parts of your data: some data could be restricted, whilst other data could be shared with others
- Creating different versions of your dataset. Therefore, even if your research involves sensitive information, you could share anonymised data and keep any data with personal identifiers private
- Limit the sharing of your data to certain users, for instance bona fide researchers
- Make signing a data use agreement that dictates how the data can be used a condition of data sharing
- Applying a licence that specifies how the data can be used. See, ‘How to License Research Data’ by the DCC
Data access statements
As a minimum, you should include a data access statement in your publication that explains how, and under what circumstances, if any, the underlying data can be accessed. If the data is closed, you will need to offer a reason why and note if, and when, any embargoes will be lifted. Additionally, a record describing your data (a metadata record) must exist in NTU’s IRep.
Contact LibResearchTeam@ntu.ac.uk to talk through the options for archiving, registering, sharing and licensing your data in more detail.