Politics is broken: The Guardian’s Alberto Nardelli on the fragmentation of party support

Written by Brian Tarran on . Posted in Politics

It’s almost the end of party conference season in the UK – a time in which political leaders take stock of the past, propose grand visions for the future and throw policy ideas through the ringer of debate and discussion. Party bigwigs talk about opportunities, of course, and challenges too. And one of the biggest challenges they face is the fragmented nature of modern politics. 

Running against the odds – The probabilities of the London Marathon ballot

Written by Anthony Hatswell on . Posted in Sports

The late Chris Brasher, founder of the London Marathon, described the event as the ‘Suburban Everest’ - something everyone can aspire to do. Over the years, an increasing number of people have done just that. At the first London Marathon in 1981 there were 7,000 runners, in 2016 there will be 37,000 (and that ignores the growth in the sport due to new events around the country).

The stats in sport panel at RSS 2015: the bigger picture

Written by Oz Flanagan on . Posted in Sports

In the absence of huge amounts of money, gaining a competitive edge in sport requires creative innovation. As Moneyball most famously dramatised, statistics is one avenue that can deliver that edge. But different sports can have vastly different levels of noise to wade through before a signal is found. This question was the basis for a panel discussion at this year’s RSS Conference in Exeter. The panel consisted of a mix of sports coaches, academics and a statistician, who are all trying to discover an insight into sport through data, that would otherwise remain hidden.

How to communicate the uncertainty in economic statistics

Written by Charles Manski on . Posted in Economics & Business

Government statistical agencies often report official economic statistics as point estimates. Buried within the documents describing their data and methods, there may be an acknowledgment that these are in fact estimates subject to error, but they typically don’t quantify the size of these errors. News releases in turn then present the official estimates with little if any mention of potential error. But I think their failure to communicate this uncertainty gives an incomplete view of the statistics. In my opinion, agencies need to measure and report this more prominently in their news releases and technical publications.


Significance magazine is available by subscription or as a member benefit in joining the American Statistical Association, or the Royal Statistical Society.

Please visit the website of the American Statistical Association or the Royal Statistical Society for information on becoming a member or to renew your membership.