When it comes to the visual representation of scientific information, in a scientific context, does aesthetic matter? In my day job at the British Library, I’ve spent the past year curating the upcoming Beautiful Science: Picturing Data, Inspiring Ideas exhibition. This experience has given me a phenomenal opportunity to think about the way we communicate and discover things in science. And, I think there’s a strong case to be made for beautiful science.
The visual representation of data is a fundamental part of what it means to be a scientist today. Whether a single data point plotted on a graph or a whole genome sequence, data visualisation helps us to examine, interpret, and contextualise information in a way that numbers and statistics often do not. Moreover, at a time when we are expected to process ever-increasing volumes of information, visualisations are often more readily digestible than some of the more ‘traditional’ alternatives; as the increased prominence of colourful ‘data viz’ work in the pages of our newspapers, websites, and in-flight magazines would attest.
So you could be forgiven for thinking that data visualisation is a new fad that has emerged, hand-in-hand, with the current era of ‘Big Data’ and ‘Open Data’. However in Beautiful Science, we explore the rich historical legacy that has inspired and enabled the work of scientists today. The exhibition shows how scientific discovery has advanced alongside advances in technology for capturing and representing data, and how the ways in which we think about information in the fields of public health, meteorology and evolutionary biology are changing as a result.
Like today, the Victorian era witnessed a great proliferation in the number and variety of data graphics and visualisations. In part, this was spurred by a new availability of statistical information and widespread collection of data on everything from vital statistics to the weather. Psychologist Michael Friendly argues that advances in statistics, the availability of data, and new technology created an environment in which information graphics could flourish.
Florence Nightingale was one person to take advantage of this ‘perfect storm’, producing her famous ‘rose’ – Diagram of Mortality of the Army in the East – in which she masterfully demonstrates that in the Crimean War, vastly more soldiers were perishing from poor hospital conditions than from wounds inflicted on the battlefield. In an effort to push the government to implement vital reforms, her simple, clear diagram persuasively communicated her point and helped develop support for the Red Cross.
Other notable figures from this period who used data graphics to tremendous effect include the epidemiologists William Farr and John Snow – creator of the cholera ‘ghost map’ – and the amateur meteorologist Luke Howard and French engineer Charles Joseph Minard. Many of the techniques that they invented are still in use today; their graphics have a strikingly modern sensibility and they easily stand head-to-head with some of the best graphic design work being done today.
Our brave new world of online communication offers enormous opportunities to scientists interested in increasing the impact of their research. These days you can tweet, blog, and even dance your PhD! But I’m going to suggest something a little provocative: why not also invest some time in your figures? In curating Beautiful Science, I was surprised – and also dismayed – to discover that much of the really good data visualisation work was being done by designers, for lay communications. There were certainly exceptions—scientists are absolutely capable of producing exceptionally engaging figures as our exhibition attests– but many figures that I encountered in research publications were, if not downright ugly and difficult to read, simply rather tedious.
Why is this? Is it a lack of training in the visual arts? Or a focus on function over form in peer review? Or perhaps researchers are just not that bothered, so long as the message gets across to the specialists in their field? Many of us were, at some point, introduced to Microsoft Excel – or MATLAB, or R, or SPSS – and we seem to have assumed that because we have learned how to use these tools, we are competent enough to make figures for publication. And yet, in spite of the power of these tools, it seemingly remains very easy to create uninspiring graphics. It need not be so.
Good figures take time, and they require choices to be made regarding the format of the data, the symbology, the layout of the figure… not to mention colours, fonts and even word- and letter-spacing. Both form and function are possible in the presentation of scientific data, and this need not take away from the integrity of the information presented. Rather, by increasing the visual impact of their data, researchers may also improve the communication of their research and results—and dare I say it, raise the impact of their own research. A figure that is both easy to read, as well as eye-catching will have greater reach, not just within the scientific community, but also beyond it.
Beautiful Science demonstrates how the visual display of scientific data is integral to the process of scientific discovery and communication. It is through this lens that researchers can tease out and identify trends, patterns and associations to draw conclusions and communicate the meaning of their data and findings. Without attempting to impose a singular aesthetic upon the visualisation of scientific data, it is nonetheless clear that the careful, considered, and beautiful presentation of data can – as was shown by Florence Nightingale – change the world.
Beautiful Science: Picturing Data, Inspiring Insight opens at the British Library, London on February 20th and runs through 26 May 2014 and is sponsored by Winton Capital Management. Elements of this blog post were published previously in Research Fortnight.
The Case for Beautiful Science by PLOS Blogs Network, unless otherwise expressly stated, is licensed under a Creative Commons Attribution 4.0 International License.