Talking to the media

DJS, IMA Bulletin September 2010

Compared to people like Marcus du Sautoy, Ian Stewart and Simon Singh, I am a newcomer to the media business. But since I got the job as Winton Professor of the Public Understanding of Risk in Cambridge I have, rather late in life, been learning something about talking maths to the media. Or, in my case, talking about probability and statistics in the guise of discussing risk. My CV now covers quite a lot of radio, both live and recorded, and writing articles in the mainstream media and blogs. And a bit of TV. Being old, I feel most comfortable with traditional media, and although I increasingly enjoy blogging, I cannot get used to wittering on Twitter.

Most of what I can say about talking maths to the media applies to talking anything to the media, and from my very limited experience I do have a few conclusions about working with radio and TV. Although media people may disparage media training, I am very pleased to have received some from the Medical Research Council. Apart from showing what can go wrong, and practising being grilled about difficult subjects, they emphasised a few general points that have proved really useful.

The first, and most crucial, is preparation. Working out the three main things that need to be said, and even the phrases that sound good, and making sure that they get said, while appearing to answer the questions beings asked. Even if a researcher has suggested in advance what the questions are likely to be, when it comes to the real interviewer they will ask just what they want and you have to be ready. Which is why it is acceptable, if asked for a quote or a recorded interview, to ask to call back a little later in order to have some time to prepare. But you do need to be available at all times and get back to people quickly.

Second, despite the popular image of the boffin with explosive hair, pebble spectacles and a top-pocket full of pens, it is good to try to be reasonably human, and possibly even amusing if that is your natural state. And definitely not defensive. I was recently challenged on this having made a set of possibly ill-advised predictions about the premier league which proved disastrously inaccurate. In 2009 I had done very well on More or Less, the excellent Radio 4 programme on statistics and numbers, making better predictions using a statistical model than the BBCs own pundit Mark Lawrenson. This year Today asked me to repeat the performance and the results were woeful. John Humphreys had a field day but I tried to take it lightly and even got in a mention of regression-to-the-mean, in that since I did better than expected last year a reduction in performance should only be expected. Humphreys considered this a pathetic excuse, but I was told afterwards he liked the item and have had encouraging feedback from colleagues.

Third, beware of being set up, when a producer has a script that calls for an ‘expert’ to express a desired opinion, and they have phoned around until they have found a suitable mug. I, dear reader, have been that mug. Fourth, expect to be cut. Twice I have been completely edited out of TV programmes after spending hours filming. Live is preferable.

All these are general media points you can get from any manual. What about maths in particular?

There appears to be a huge and growing appetite for accessible explanations of fairly technical issues. It is great if they can be made relevant to real world, but this does not seem to be essential: if they are communicated with enthusiasm, and maybe even passion, then there is an audience. The recent BBC Horizon programme about Infinity was very popular in spite of having no practical interest for anyone watching and just comprising ideas that, from personal feedback, seemed to enthral people with no mathematical background. “No equations, no jargon” is the standard mantra about science communication, but some technical language seems to be fine provided that it is clearly identified and the general tone is appropriate for, say, a non-scientific friend. The success of More or Less on Radio 4 backs this up.

Stories, analogies, and images are important, conveying the excitement and the fascination of the subject – in any case, nobody you talk to afterwards will remember any details whatsoever, except of course pedantic colleagues who may delight in finding fault. Spurn them.

If contacted by journalists for a newspaper article, don’t expect to be able to check what is written, so assume that everything said might be reported, and take care on the phone, since everything you say may be taken down and used as evidence against you. Writing your own articles provides the greatest opportunity: practice on your blog, submit to other blogs (eg the Times Science blog), submit to Comment is Free in the Guardian (although be prepared to then endure fatuous personal comments). The Science Media Centre is also a wonderful institution that puts journalist in touch with scientists.

I have particularly enjoyed the opportunity to write newspaper articles – you can Google ‘Journalisted Spiegelhalter’ for examples - and have been fortunate enough to build up a good relationship with the Times and, like Marcus du Sautoy, have received the obligatory letters from Miss Pick. The advantage of writing a column is being able to choose a topic, and so when Paul the Octopus started predicting World Cup results it seemed a wonderful basis for a story on chance and evidence. The Times at first thought this was not interesting and asked for another story, and then a week later desperately phoned me up asking for 400 words by that afternoon on the octopus as everyone was covering Paul. It was good to be vindicated.

But what is the point of all this activity? I have to admit that I don’t study the “public understanding of science” and don’t feel a particular need to academically analyse what I do (I also don’t mind splitting infinitives). I know that simply talking at people, without proper two-way engagement and participation, may be thought anachronistic, and in broader educational work I try and do better. But given the current media there is still a role for the ‘performance’ element in science communication, and long may it last.

Finally, I have to admit it helps being a professor from the University of Cambridge. But against that I am a grey-haired old male and haven’t been a rock star. But we can’t all be Brian Cox.