Understanding When Statistics Are Badly Researched
News or Selling Papers?For me, this idea always brings to mind a TV interview during which a senior police officer was getting a grilling for a recent spate of stabbings in London. The officer – who had clearly done his homework – repeatedly told the reporter that, in spite of the recent events, knife crime had been trending downwards for years and continued to do so (which the officer attributed to a range of police initiatives). The reporter wasn't listening, and he repeatedly harangued the officer for a lack of police action. To me, it looked obvious that the news bites on knife crime which had been circulating in recent weeks had trumped all research. At that time, if you'd asked anyone whether knife crime in London was up or down, it would have been hard to find anyone who thought it was reducing. In the public's mind, the easy-to-recall news headlines were creating Availability Bias and trumping the real figures. The point is this: Be pretty suspicious of the statistics you see in the papers, and don't believe everything you read. There are some notable news media biases.
News or Advertising?On slow news days, papers will sometimes use articles that have been spoon fed by companies' or charities' public relations departments. So, a headline like "70% increase in ice-cream sales over the heat wave" will probably have originated with a phone call from the ice-cream sellers' corporation. A 2001 study in Australia found that over 40% of the "news" in their newspapers was fed by PR departments. It's not news. It's advertising. Such articles are lazy journalism.
And who's going to complain? Certainly not the paper's editor. He gets to fill column inches for next to nothing. And certainly not the vendor. He gets free advertising. And certainly not the reader. We're too trusting. It's a win-win-lose situation, and we're the losers.