Dr Jennifer Rogers, research fellow at The University of Oxford, discusses the interplay between statistics and biological research for World Statistics Day.
“The best thing about being a statistician is that you get to play in everyone’s backyard” is the now infamous quote from John Tukey and it is one that I whole heartedly get on board with. I have been a medical statistician for about eight years and in that time I have had the opportunity to work on research projects that have been focussed on epilepsy, cardiovascular disease, malaria and diet and nutrition. But why does being a statistician offer us these exciting opportunities? Well I believe that statistics has to be undoubtedly one of the most important tools in scientific research and biology in particular. Biological research generates huge amounts of data, but without statistics, that data remains just data and the exercise of collecting it merely academic. It is statistics that unlocks the information within that data and allows us to examine relationships between outcomes and associated variables and make predictions about future events.
I work in clinical trials where the interplay between statistics and clinical biostatistics could not be more meaningful and relevant. Statisticians are integral to every stage of the clinical trial process, from design, the collection and monitoring of data, through to the analysis and presentation of results. But it isn’t just in clinical trials where the need for statistics is clearly evident, the analysis of observational data has resulted in many impressive public health advances. John Snow famously used data to study the pattern of a cholera epidemic in Soho in 1854, thus identifying the Broad Street Pump as the source of the outbreak. This study is thought of as the beginning of modern epidemiology. And then in 1950, Richard Doll and Austin Bradford Hill established the link between smoking and lung cancer and initiated the British Doctors Study to examine this link further.[2,3] This result is now conclusively proved and has been of huge benefit to public health.
But statistics doesn’t only have a place in medical research, it is used throughout biological research in general, be it genetics, botany, ecology; any time where an experiment is conducted and numerical data collected, statistics will be essential in turning that data into knowledge. And of course, no summary of the relationship between statistics and biology would be complete without mentioning R.A. Fisher. R.A. Fisher was a statistician and biologist who is often thought of as the founding father of modern day statistics, evolutionary biology and genetics. His book, The Genetical Theory of Natural Selection helped define population genetics and The Design of Experiments is considered a foundational work in experimental design.[4,5]
So what does the future hold for statistics and biology? As computational capacity has increased alongside rapid developments in molecular research technologies, mathematical and statistical techniques have advanced, transforming the field of bioinformatics. The vast amount of biological data that is now available for analysis brings with it new challenges for the statistician with respect to data mining, machine learning, pattern recognition and visualisation. These challenges, driven by biological applications, help develop our statistical profession and, in turn, it is these statistical developments that unlock the scope for biological advancement. The subsequent opportunities for ground-breaking and powerful research never cease to amaze me.
Statistics and biology have enjoyed a long and rewarding relationship to date, and it is one that will most definitely continue to be successful going forward.
 Snow, J. On the Mode of Communication of Cholera. London: John Churchill, 1855.
 Doll R, Hill AB. Smoking and Carcinoma of the Lung. British Medical Journal. 1950;2(4682):739-748.
 Doll R, Hill AB. The Mortality of Doctors in Relation to Their Smoking Habits. British Medical Journal. 1954;1(4877):1451-1455.
 Fisher, RA. The Genetical Theory of Natural Selection. Oxford: Oxford University Press, 1930.
 Fisher, RA. The Design of Experiments. Edinburgh: Oliver and Boyd, 1935.