This month, the University appointed its first Director of Data Science and eResearch following the development of its first Data Science and eResearch Platform Strategy last year.
We sat down with Associate Professor Shawn Ross to learn more about the future of the field, what researchers can learn, and how he’s taking his experience in the Department of Ancient History to this new role.
The fields of Ancient History and Data Science and eResearch seem like they are worlds apart. Are there any similarities?
Ancient History and Classics were fairly early adopters of approaches like computer assisted text analysis and agent-based modelling (amongst others). Archaeologists use geographic information systems (GIS) and statistical analyses day in and day out, but struggle to manage the diverse data that fieldwork generates.
What made you interested in this directorship and what experiences will you draw on for the role?
My experience building eResearch infrastructure, and the cooperative work with a range of field researchers in a number of disciplines that arose from it, drove my interest in the directorship. Working with researchers to sort out their data reveals weaknesses in research design pretty quickly, and that led me to confront the problems with transparency and reproducibility that many disciplines are now facing. Every solution to those problems I’ve seen involves adopting digital research methods.
The University has launched a new Data Science and eResearch strategy. What do you see as the pillars of this strategy?
I’d say the ‘pillars’ of the project are:
- Ensuring staff have the resources they need to accomplish their research to the highest possible standard.
- Building staff awareness, confidence, and capacity regarding digital scholarship. Reaching beyond ‘innovators’ and ‘early adopters’ to the majority of potential beneficiaries is something I will be focussing on as part of my role.
Where do you see the field of eResearch going in the future?
In the future, I think that eResearch will really focus on making research designs and outcomes more robust, especially with regard to transparency and reproducibility. In order to evaluate, reproduce, or replicate results, all steps of the research project and all ‘research objects’ (not just data but, for example, the code underlying all statistical analyses) have to be available.
What is your advice to researchers who are curious about exploring data science and eResearch but have yet to do so?
I’d say the best way to start is to bring a research problem, or an early-stage research project, to someone who has done data science or eResearch work in the past, and see what improvements to your usual research approaches digital techniques can offer. It’s always easier to work with someone else, at least at first, if you want to explore digital scholarship. We’ve had students in our Digital Methods MRES unit marvel at how much time they can save, or how much more they can do, through the judicious application of technology to their research. The other thing I would say is that introducing digital methods to a project usually involves an up-front time investment for a later pay-off in efficiency or quality.
To learn more about the strategy and priorities, visit the Data Science and eResearch page.