Good research data management from the very beginning will help your project run as efficiently as possible.
At the beginning of your project, there are a number of factors you may want to take into account when planning how you (and your colleagues) will manage your data. The RDM Service can help you with one-to-one clinics or group workshops.
Most external funding bodies expect supporting data to be preserved for the long term and shared where possible. Check this list to find out the specific requirements of your funder.
If you’re not externally funded, you are still obliged to follow the SGUL Research Data Management policy [PDF], which requires that you keep your research data for at least ten years.
You will also need to be aware of the relevant Information Governance issues and will need to complete an SGUL IG Checklist for Research Projects Self Assessment.
It’s likely that you’ll need to write a Data Management Plan for an external funding application. Templates and SGUL-specific guidance on how to complete these are available at DMPOnline. Instructions on how to access DMPOnline for the first time is available. You are strongly advised to use DMPOnline for access to the SGUL guidance and sample answers and training is available from the Library on this. Check the Library’s training pages to see when the next session is running.
DMPOnline is a free and easy to use tool that will help you write a Data Management Plan, as requested by most external funders. It has guidance and exemplar answers specifically tailored to SGUL researchers, so should be your first port of call when writing a DMP.
- Navigate to DMPOnline and from the DMPOnline home page, click the ‘sign up’ button
- Enter your SGUL email address, select St George’s, University of London from the dropdown list of organisations and choose a password
- DMPOnline will then send an automated email for you to confirm your registration
- Once you’ve done this, you’ll be able to sign into your DMPOnline account and link it with your SGUL account so that in future you’ll be able to log in via the ‘sign in with your institutional credentials’ link, underneath the orange ‘sign in’ button
Create a plan by clicking the ‘create plan’ button on the main page after logging in. Select the funder, e.g. MRC, from the dropdown menu. The second dropdown menu should automatically suggest institutional guidance tailored to SGUL researchers, but if not, then search for ‘St George’s University of London’ in this dropdown menu. If you tick the ‘DCC guidance’ box then you will also be given advice from the Digital Curation Centre (DCC) to help fill in the DMP. Click ‘create plan’ at the bottom, confirm your choice and enter the plan details on the next page. Then use the white and orange tabs to move through the DMP and enter information.
You will only be presented with the questions that are required by that funder’s DMP template. The SGUL guidance is intended to help you write a robust plan that identifies potential problems with solutions. For example, your data may contain confidential data but could be shared if it underwent redaction, anonymisation or de-identification processes and thus including staff time for performing this in the DMP. For some questions, exemplar answers have been provided for SGUL researchers, such as for storage and backup procedures and other relevant policies. You may use these and add or amend these as you choose.
You’re able to save and share a plan at any point and choose what permissions those people have.
- ‘Read-only’ access – these people will only be able to see the plan but not make changes
- ‘Editors’ can add to the plan
- ‘Co-owners’ can add to the plan, edit the plan details (e.g. title, grant number, authors) and set any access restrictions to other users
As part of the DMP, you will need to think about documenting and archiving your data – being organised at the beginning will really help when you’re gathering data and when you wish to share it at the end of the project.
It’s rare to simply be able to deposit your data or share it with others without explaining how you organised it, e.g. ‘Column A = blood pressure’. It is strongly recommended that you provide documentation that describes your data for other researchers, so that they can understand how the data were processed. Its best practice to provide a ‘readme’ file alongside your dataset that captures all this information, usually in a text file format (.txt).
Useful contextual documentation for a ‘readme’ file includes:
- A brief summary of the research project – who produced the data, how it was collected and prepared for analysis
- Description of details of any alterations/coding that has been carried out on the data
- A list of errors/missing values, how these occurred and how they are shown in the data
- Definitions of descriptions and annotations used at file/within-file levels (e.g. column headers, units of measurement)
- List of equipment/models/software used to create data or conduct analysis
- Questionnaire/interview templates, consent forms and other documentation that you used during your project
- If there are metadata standards for your discipline, try to use these as it will make it much easier to catalogue and share your data with others at the end of the project if you are all using the same taxonomy
Last Updated: Wednesday, 31 August 2016 16:06