Our researchers create, collect and measure large amounts of research data every day, and some of this data is used to support our internationally-recognised research. Researchers expect the University to provide secure and resilient storage during a project, on a cost-recovery basis, in order to prevent data breaches and data loss. Funders expect researchers to publish the data (and the code) which supports their findings at the end of a project unless there are ethical, commercial or public safety reasons for not doing so. In the spirit of open scholarship, researchers are expected to demonstrate they are working transparently and reproducibly. Researchers are expected to use discipline-specific file formats and metadata tags (where these exist) to improve interoperability and re-usability. Researchers are expected to document their data to improve comprehensibility for themselves and for others. Researchers studying a human population are expected to write a plan which details how they will protect and manage the personal data pertaining to their study participants. Research Data Management is concerned with all of these topics and more.
An excellent paper on Current practices in research data management and sharing in the UK has been written by Ian Carter. This is recommended reading for anyone interested in RDM today. Another outstanding paper on Good enough practices in scientific computing has been published in PLOS Computational Biology. This highly practical paper covers data management, writing code, collaborating with others, project organisation, tracking changes and writing a manuscript.
Research data flows through a life-cycle. At the beginning of a new project, you write a project plan. Funders now expect you to write a data management plan also. A data management plan conveys to your funder and to others that you have a strategy for responsibly mananging your research data during and after a project. Once you receive your funding, you collect, create or measure your research data depending on your methodology. You might have a small amount of data or Petabytes. Either way, you need to store your data safely. Then you analyse or process your data, and you need appropriate collaboration and data analysis tools to do so. You might need to write your own data processing scripts. Your analysis leads to the discovery of something new and you write a paper. Your paper gets accepted. Funders now expect you to publish your research data and code simultaneously with your paper. You may need to anonymise your data if it contains personal information. When you publish your data in an open data repository, your data becomes easier to find because a DOI and metadata are assigned to your dataset. You cite your own data in your paper. When others find your data, they might re-use it in their research project, provided they cite you. Thus the research data life-cycle continues. (Diagram by Jisc licenced under CC BY 4.0)
Please e-mail the University's Research Data Manager, Nicholas Syrotiuk, if you require assistance with any aspect of data management.