Your Research data is your craft

All PGRs at the University of Birmingham are required to have a data management plan in place as part of their first Annual Progress Review. In this post, Angeliki Andrikopoulou, Research Data Librarian in Scholarly Communication Services in Library Services, explains why they are important and where you can get support.

Research data management (RDM) is a crucial element of the research lifecycle as it consists of activities related to creating, collecting and reusing research data during a research project. Data governance skills and knowledge are required from all researchers, either creating new data or reusing data.

Angeliki set up for her RDM drop-in, in the Researcher Suite on the first floor of the Main Library.

In this digital era, research is data-centric. Research data is a significant output that has become increasingly important and equally valuable over the last years equal to journal articles, books, and other research outputs. Consequently, data management planning is the first and most appropriate step to undertake at the beginning of the project to ensure data quality.

A helpful way to visualise the data management plan is as a path you must step on throughout your research project. A data management plan will allow you to plan and make decisions about all the critical research data lifecycle stages beforehand and save you valuable time and effort. For this reason, writing and updating this important document should be a meticulously performed act.

Furthermore, data management planning will guide you in making your data open, if appropriate, and FAIR. To make your data FAIR, you must follow four principles: Findability, Accessibility, Interoperability and Reusability. FAIRness ensures that your research outputs are valid, reusable and reproducible, increasing the impact of your research, the quality and the resilience of your data. Data management planning and FAIR are unique tools that will guide you in organising, publishing and possibly sharing data in appropriate formats and manner that others can understand. Your data is your craft, the result of your hard work and should be considered a valuable asset and looked after accordingly. Adopting those tools will ensure that you and others and society in general can benefit from your research outputs.

Library Services provide various support dedicated to your research data journey. Recently, a new facility was added to this support. Scholarly Communications Services have launched a weekly RDM Drop-in session. It takes place in person every Friday 1-2 pm in the Researcher Suite on the Main Library’s First Floor, and no booking is required. We anticipate that this relatively new drop-in session will help to increase researchers’ awareness and improve their skills related to RDM. Thus, if you have queries or wish to discuss your research’s particularities concerning research data, grab your coffee or tea and join a research data librarian every Friday after lunch -no pre-booking is required. Dates can be found on the library website, or you can contact us if you require further information.

The BEAR Necessities of Research Computing

In this post, Debbie Carter, Research Training and Engagement Officer in IT Services, summarises how the Birmingham Environment for Academic Research (BEAR) can support PGRs at the University of Birmingham.

When I started writing this blog post, I wanted to address the most common questions we are asked by PGRs about BEAR (Birmingham Environment for Academic Research).

Debbie (left) and Aslam at the BEAR drop-in session, Main Library, April 2022

Let’s start with data, as this is the fundamental starting point for any PGR. You’re going to be managing data of some sort, whether you have produced it yourself, or are re-using existing data. Depending on the type of data, you might also need to use software to process it. BEAR services are designed to help researchers with managing these processes, supported by a team of software engineers and data experts.

We often meet PGRs who store data on their laptop, USB stick, or Dropbox-style storage, but this is not recommended by the University, and we definitely don’t want to risk anyone losing their precious resource! See “Where should I store my research data?” or our information on storing sensitive data. Whether you have a large or small amount of working research data to store, the BEAR Research Data Store (RDS) offers secure storage, backed up each night to two data centres on campus, so you can be confident that your data is safe. Ask your supervisor to fill in our project request form to get access. If your research is in Life Sciences, we have dedicated CaStLeS resources available.

When you leave UoB, data will remain in the RDS under the authority of the project PI. If after 5 years the project has ended, the data will be archived for a short period of time before being removed. Make sure you copy any data off that you will need for writing up papers after your thesis is complete, as you will not be able to access data on the RDS once your PhD is marked as complete – this depends on your School/Institute and could be as soon as you hand in your thesis.

Our Linux-based High Performance Computing (HPC)/High-Throughput Computing (HTC) supercomputer, BlueBEAR, can be used to process data faster than on your own computer. There are over 1000 software applications installed, from ABAQUS to zstd (see BEAR Applications), and you can ask us to install other software too. Our website explains how to activate your BEAR Linux account – this is needed to log in to BlueBEAR. We also provide web-based access to selected applications via the BlueBEAR Portal.

Often researchers need to use BEAR systems but are not experts in software and programming. We offer Software Carpentry workshops in Git, MATLAB, Python, R, and Linux courses, starting from the basics to prepare you for using BlueBEAR. Information and dates for courses are on our training webpages.

One of our main methods to assist researchers is via the IT ServiceDesk. Regular drop-in sessions are held both online and in-person to give you access to the BEAR team and ask questions. From tricky software problems to sensitive data, come along and we can help! BEAR Software offers a range of support and advice on the use of research software including how to optimise BlueBEAR jobs to get the most from our facilities, provided by a rapidly expanding team of Research Software Engineers. Or you can email us and we are happy to help!

Is research a risky business?

Topics around project management come up fairly regularly on this blog, because I think that getting to grips with these kinds of techniques is really helpful in terms of managing your time.  But good time management isn’t the only thing that’s going to keep your research on track.  There’s another strand of project management which is equally useful – risk management.  This is about acting now to increase the chances of things going well.

We all engage in risk management on a daily basis: when we look both ways before crossing the road, for example.  We assess the risks and make changes accordingly: if we judge a road to be particularly fast/busy, we might walk a bit further to the pedestrian crossing.  But risk management in research goes a lot further than health and safety. 

Planning to manage your data

We’ve talked a bit about Research Data Management (RDM) on this blog before, with a post from our Research Repository Advisor in Library Services and another from a current PGR.   However, now that there is a requirement for all PGRs who started their research programmes in or after September 2017 to produce a Data Management Plan (DMP) in advance of their first annual review, it seems a good time to revisit this topic, with a focus on DMPs.

A DMP is a living document that outlines how data are to be handled during and after a research project.  A good DMP will protect you against data loss and ensure you have well-documented data to assist with writing up and possible future data sharing.

Big data, small data, no data

There are many aspects to a successful PhD project and challenges to master on your way to graduation. You most certainly are aware that you should acquire e.g. writing and referencing skills, but how much time have you spent thinking about the research data management activities you might need to undertake as part of your research?  None yet? Time to get started with our introduction to research data management!

Spotlight on the RDF: “Appropriate practice”

In this series of “Spotlight on…” posts, we’ll be delving into the detail of the descriptors in Vitae‘s Researcher Development Framework (RDF).  Each one of the sixty-three descriptors is a characteristic of an excellent researcher, and we’ll be looking at how UoB PGRs can develop these characteristics.

The RDF  descriptor “appropriate practice” is one which is easier to define through its opposite:  academic malpractice is any activity – intentional or not – that is likely to undermine the integrity essential to scholarship and research.  Examples of academic malpractice include plagiarism and falsification/fabrication of results.