Today’s focus is on a subject near and dear to my heart, research data management. I’ve been teaching on this subject both within my own library and to groups in various other departments around Cambridge as part of last year’s Research Ambassador project. I was even invited to teach bioinformatics PIs from all around the world how to create good data management plans as part of an international training conference they were attending. I believe good data management is fundamental to the research process.
I am fortunate enough never to have lost data myself, but I have heard horror stories of laptops stolen, USBs left in public computers, and the resulting loss of entire PhD theses/underlying data. I even met a researcher once who had backed up all her data onto an external hard drive, but left the hard drive connected to her home computer, which was subsequently stolen in a burglary, in an extraordinary streak of bad luck. When teaching research data management, among other things, I recommend to my students that they back up their data using at least two different methods, and in at least two different physical locations.
Data loss is not the only aspect of research data management: researchers also need to think about file naming conventions, sharing methods (particularly if the research is being undertaken collaboratively with a team based at multiple institutions), and making data associated with publications available openly. These elements may have costs associated (for example storage of data on an open access repository), and this needs to be considered at an early stage of the research process so that costs can be factored in to grant applications. I strongly recommend using a tool such as DMP Online, which has templates for data management plans required by most major funding bodies. These can be completed by multiple users, so a group applying for a grant can share in the creation of such plans. If a particular funder’s template is not represented on DMP Online, there is a generic template that can be used. This resource makes the process of creating a data management plan a lot simpler, as the templates make it very clear what sort of information needs to be considered and included. It is free to use.