Ben Goldacre highlighted another case of statistics used badly in the Guardian on Saturday: the claim that Britain’s happiest places have been mapped by scientists, according to the BBC and many newspapers.
Erm, says Ben Goldacre, there’s a slight problem with that. He shows how sampling has yet again been misused by journalists. “This entire news story was based on nothing more than random variation,” he reports.
“This is called sampling error, and it quietly undermines almost every piece of survey data ever covered in any newspaper.”
When Goldacre talked to the scientist behind the research, Dr Dimitris Ballas, he said: “I tried to explain issues of significance to the journalists who interviewed me. Most did not want to know.”
‘Ithika’, originally flagged this up, posting in the Bad Science forum and has written about it at dougalstanton.net. APGaylard and Gimpy have also blogged about it on their sites.
Not to foget Gimpy of Gimpy’s Blog, who beat me to the punch!
http://gimpyblog.wordpress.com/2008/09/03/the-bbc/
Apologies to Gimpy! I’ll add in the link now.
No worries, Ithika does deserve the credit for finding the story and making the point.
Just to let you know that ‘Ithika’ on the Bad Science forums is the same person as the author of the article on dougalstanton.net, ie me! 😉
Thanks Ithika aka Dougal. All duly noted and now updated!