• Home
  • The Statistical Society of Canada's
Excellence in Data Journalism Award, 2016

The Statistical Society of Canada's
Excellence in Data Journalism Award, 2016

24 May 2016 10:06 AM | Anonymous

We are pleased to announce that the inaugural 2016 Excellence in Data Journalism Award winner is THE GLOBE AND MAIL'S ELECTION FORECAST, a web page created by journalist Matt Frehner in collaboration with political scientist Paul Fairie. Designed and developed by Jeremy Agius and Julia Wolfe, produced by Chris Hannay. This web site aggregated all of the election polls in the lead-up to the October 2015 Canadian federal election, applied a "uniform swing" model to make election predictions, and ran a thousand Monte Carlo simulations to estimate the probabilities of various electoral outcomes.  It also allowed the reader to run their own Monte Carlo simulations to view their own potential election outcomes.  It provided colorful graphical illustrations of the riding-by-riding results of each simulation outcome.  An accompanying article described the methodology in greater detail.

The award committee was impressed with the forecaster's use of sensitivity testing, including providing probabilities of various outcomes as opposed to simply the most likely outcome.  We also liked the ability for readers to run repeated iterations of the simulations, and see different results each time.  The accompanying methodological article was a bit terse at the beginning, but it did make a good effort to explain to general readers how the numbers were analyzed and presented.  And, the topic (federal election predictions) was extremely important and widely discussed during much of 2015, so the web page had significant impact on improving the Canadian public's understanding of the data analysis aspects.  (Indeed, it was apparently the second-most highly viewed Globe and Mail web page for all of 2015.) For more information, visit The Globe Election Forecast at:  http://www.theglobeandmail.com/news/politics/globe-election-forecast-2015/article25377958/and see the background methodology at:  http://www.theglobeandmail.com/news/politics/how-our-election-forecasting-model-works/article25371747/

The Statistical Society of Canada is an organization whose mission includes the development of a public awareness of the value of statistical thinking and the importance of statistics and statisticians in Canadian society. Science writers play a central role in educating Canadians about statistical results and concepts. To recognize and encourage excellence in data journalism and its positive impact on Canadian society, the SSC is proud to sponsor the Excellence in Data Journalism Award.

The CSWA has been delighted to partner with the SSC in creating and judging this award. That collaboration flowed from our belief that great data journalism is a goal all science communicators and all non-science communicators should have in a world in which understanding and depicting statistics has become the intrinsic structural underpinning of much of modern reportage.