Tag Archives: Britain

Random journalising: another case of journalists misrepresenting stats

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.netAPGaylard and Gimpy have also blogged about it on their sites.

links for 2008-06-30

Guardian: James Murdoch calls daily/Sunday integrations ‘cost-cutting exercises’

James Murdoch has spoken out against integrating daily and Sunday newspapers into seven-day operations, dismissing the process as “cost-cutting exercises”.

He told staff at its UK newspaper division News International that the introduction of integrated seven-day operations in Britain had “diminished daily and Sunday rivals”.

Associated Newspapers claim monthly reach of 22 million

Associated Newspapers websites and print editions claim to reach a combined audience of 22 million adults a month – 45 per cent of all adults in Great Britain, according to figures released by the group’s digital division today.

The survey of 60,000 readers, which was conducted across Associated Northcliffe Digital’s (AND) network of 38 websites, used a questionnaire to profile the age, online behaviour and print reading habits of respondents.

Those polled were also questioned about specific subject areas of the larger sites to investigate how users differ between these sections.

The information gathered will be used to create advertising campaigns relevant to several sites or areas, a press release from the company said.

The final figures were produced by Survey Interactive, which is also involved in developing a new audience measurement tool for Trinity Mirror’s websites.

Speaking to Journalism.co.uk last month, Guy Lipscombe, managing director of Survey Interactive, said the polling technology can give figures on the ‘unduplicated reach’ of a title – the number of unique users to a site who do not also read a print edition.

The system, he said, is based on ‘measuring the people not computers’ who view a site, preventing duplicate records as a person accesses a site across a range of devices.

However, it is unclear as to how information would be gathered to disregard users that read both the print and online versions. Wouldn’t it be more interesting to see these ‘unduplicated’ stats?