Following on from the award-winning documentary ‘Joy of Stats’, on 18th October 2012, BBC Four broadcast a documentary called ‘Tails You Win: The Science of Chance’ presented by David Spiegelhalter, Winton Professor for the Public Understanding of Risk at Cambridge University. In the beginning of the documentary, we see Professor Spiegelhalter peacefully punting along the river by Kings College (Probability of falling in whilst punting 1/200 or 0.005/0.5%), to eventually jumping out of an aeroplane for his first ever skydive. Post Olympics, Spiegelhalter brilliantly explains his data visualization on a running track. For every day that we are 5kg overweight or drink 3 beers or smoke 2 cigarettes a day, we can on average expect to lose half an hour of life. A micromort is a 1 in 1,000,000 chance of death. Given a 7,000 micromorts risk of a 58-year old man before his next birthday, the extra risk of a skydive – 7 micromorts – does not make much difference, and is much the same extra risk as running a marathon or driving 40 miles on a motorbike.
Here, Statistics Views Editor Alison Oliver talks to Professor Spiegelhalter about the aims behind the progamme and how statistics has played an influential part in his research and career.
1. With the exceptional career that you are continuing to enjoy, now as Winton Professor for the Public Understanding of Risk at Cambridge University and not to mention your innumerable contributions to media and teaching, what was it in the first place that inspired you to pursue a career in statistics and mathematics? When and how did you first become aware of statistics as a discipline?
I did maths at Oxford, and at first enjoyed the pure maths and couldn’t stand the statistics – we learnt it with absolutely no connection with problems or the real world, and no data whatsoever. But then the pure maths got too tricky, and we got taught Bayesian statistics which was beautiful and sensible, and then I went to UCL where I found that statistical inference was both important and exciting.
2. Congratulations on the documentary ‘Tails You Win – The Science of Chance’ after the award-winning ‘The Joy of Stats’ with Dr Hans Rosling. What was your main objective that you wished to get across to the viewers at home?
The joy and excitement of trying to understand how chance works in the world, but also the basic difficulty of trying to pin down what exactly it is.
3. From your participation in events like the Millennium Mathematics Project and other school projects associated with Cambridge University, it appears you’re very passionate about getting children interested in mathematics. How have these projects been progressing and what do you hope to be the outcomes?
I would love to have just a small impact on how probability and stats are taught in schools. I am quite good at the ‘inspirational’ bit, but it also requires skilled educators to design materials and questions that are engaging and lead to better understanding, both for further education and for daily living. There’s a huge interest in improving maths education – much of the attention is focussing on A level, but we are aiming at younger students.
4. You obviously have a very successful research career. How do you feel now about the pressure to publish? Has your view changed over the years?
I’ve spent years going through the grind of writing papers, sending them in and getting them criticised by referees. It’s been a good discipline, and I still need to contribute to the department’s REF submission, but it’s a relief not to be under the pressure that young researchers are under. I still get things rejected, and roundly abuse the editors as poltroons. Although they often have a point.
5. Do you have any advice for students considering a university degree in statistics?
Do as much maths as you can. Take notice of what is going on in the outside world of statistics. Take the Bayesian ideas seriously.
6. What do you think the most important recent developments in the field have been? What do you think will be the most exciting and productive areas of research in statistics during the next few years?
In my career I have seen statistical inference extend from problems with a few parameters to Bayesian Markov chain Monte Carlo methods that could handle hundred of parameters. But now is the era of really big data, whether it’s internet, finance, genomics, imaging, astronomy, bioinformatics and so on. I am now a dinosaur, and no longer attempt to keep up. These are some of the most exciting new areas, and I wish the people who are studying them luck. But there also smaller problems that are of great interest and importance, say in forensics, adaptive clinical trials, and in handling messy routine data to help in deciding policy.
7. What do you see as the greatest challenges facing the profession of statistics in the coming years?
To establish an appropriate identity among the army of people who will be handling data – to continue to argue that formal techniques of inference are vital and that answers do not appear simply from visualisation and presentation, important though they are.
8. As one of the architects of BUGS software, do you think user-friendly statistical software such as BUGS and R may reduce the demand for the expertise of statisticians?
It depends what a ‘statistician’ is. Presumably it is someone who defines themselves as a statistician. There will always be a demand for people who can design studies, know what inferential techniques are appropriate, and communicate the answers. Just running the software is the minor part of the job (although fun).
9. Are there people or events that have been influential in your career?
Adrian Smith introduced me to Bayesian ideas 40 years ago and his passion for the subject was immensely important. I was blessed to know the previous generation of statistical ‘giants’, such as Dennis Lindley, George Barnard, Peter Armitage, David Cox, John Nelder and so on, who were always extremely encouraging and generous. And many of my contemporaries have been a constant source of inspiration – I have been incredibly fortunate.