“I found the real world too messy”
Dr Rhian Daniel is a reader in medical statistics at Cardiff University, with a particular focus on developing methods that identify cause-effect relationships in complex observational health data. For example, ‘when a genetic marker is known to be associated with a human disease, Daniel works on methods to disentangle the numerous causal pathways giving rise to this association.’ Although her work is now strongly tied to applications in medicine, growing up Daniel preferred to stick solely to the maths.
“Science was never my favourite subject at school; I found the real world too ‘messy’ and preferred the beauty and exactitude of mathematics,” Daniel said. “However, as I got older (and as pure maths got harder), I discovered that medical statistics contains just the right balance for me between mathematics and the real world.”
Along the way, were several figures and early moments that helped Daniel come to this realisation. “Having two maths teachers as parents was a great start, then a fantastic first statistics teacher – Mr Jenkins – at school, and a wonderful lecture course on statistical theory from Alastair Young as an undergraduate at the University of Cambridge. But I think it was when I started the MSc in Medical Statistics at the London School of Hygiene and Tropical Medicine (LSHTM), that I really felt I had found what I wanted to spend the rest of my life doing.”
“I am forever indebted to him for teaching me so much”
Subsequently, Daniel embarked on a PhD, also at LSHTM, investigating missing data methods – methods which try to minimise the effects of missing data on statistical conclusions. Following her PhD, Daniel stayed at LSHTM for a further 8 years, first as a post-doc and then as a lecturer, where her research evolved to look at methods for learning about cause-effect relationships. “In particular, I have worked on the framework and methods for separating a causal effect of a particular exposure on an outcome, into its constituent pathways through a number of intermediate variables.”
At LSHTM, Daniel encountered some impactful mentors, whose guidance helped to shape her statistics journey. “My PhD supervisor, Mike Kenward, taught me (or at least tried to teach me) how to think rigorously as a methodological statistician. My mentor after that, Bianca De Stavola (another Maths and Computing 2020 Suffrage Science Awardee), taught me that when analysing real data there are even more important things to worry about than mathematical rigour. The tension between these two standpoints inspires me (or, maybe, worries me) every day. And then there is Stijn Vansteelandt. I am forever indebted to him for teaching me so much and for being so generous with his ideas.”
Opportunities to network with other academics at conferences, has also inspired Daniel and provided a welcome break from the constant challenge of the ‘never ending to-do-list’. “Every causal inference conference (UK, European and Atlantic) I’ve managed to attend is a rare ‘holiday’ from the usual day-to-day madness, giving time to learn and to talk to so many academics working on very similar problems to mine.”
“This is a personal highlight and an encouragement to keep at it”
A move ‘back home’ in 2017, saw Daniel join Cardiff University’s Division of Population Medicine, where she remains today. Daniel’s collaborations here include projects on diseases such as Type 1 diabetes and colorectal cancer screening, with a direct impact on society. “The Type 1 diabetes project I’m involved with, for example, led by Julia Townson, ultimately aims to develop an algorithm that would be installed on GPs’ computers in order to prompt earlier diagnosis of Type 1 diabetes in children using the information already routinely entered by the GP.”
In addition to her fundamental research, Daniel’s efforts to make her field more accessible is far-reaching. “I have written papers and taught courses that try to explain some of the trickier concepts in statistical causality to a wider audience, but without dumbing them down. When occasionally someone says to me “I was really confused about such-and-such-a-topic until I read your paper and now it makes sense”, this is a personal highlight and an encouragement to keep at it.”
Daniel’s nominator, Professor Ruth Keogh of London School of Hygiene & Tropical Medicine, also applauded her clear-thinking and, more recently, her work during the pandemic. “Rhian is a supportive and fun colleague who always comes up with a solution to a tricky problem, and is a generous supervisor of students and early-career researchers. Rhian is also a role model for mothers in academia, and during the pandemic has been combining her research, including scientific work relating to Covid-19, with co-caring for her two small children.”
“Every single task and interaction can accidentally exclude people if we’re not careful”
Drawing on her own experiences, Daniel believes the way to improve diversity and inclusivity in science is to focus on the smaller actions which collectively can bring about change. “Rather than there being a few big things to change, I think there are now a huge number of tiny things that all of us need to do all of the time in every aspect of what we do. From the way we ask questions at job interviews, to the people we invite to give seminars, the times at which we organise meetings, the students we encourage to think about PhDs: every single task and interaction can accidentally exclude people if we’re not careful, or can help to nudge STEM in a more diverse and inclusive direction if we get it right.”
“My advice to women starting an academic career would be to ignore the (perhaps now outdated?) advice that to succeed you need to be confident and ruthless, or worse still, to give a false impression of being so,” Daniel continued. “The best academics I know are kind and humble.”
The Suffrage Science Maths and Computing Awards 2020 were held on Friday 6th November, 2020. You can find out more, and watch a recording of the event, here.
Hear more about previous Suffrage Science Awardees on the Suffrage Science Podcast. You can subscribe on: Podbean; Spotify; Apple Podcasts, or wherever you get your podcasts.