Professor Simon Baron-Cohen of Cambridge University’s autism research team wrote a piece for New Scientist recently about media distortion:
WHEN media reports state that scientist X of Y university has discovered that A is linked to B, we ought to be able to trust them. Sadly, as many researchers know, we can’t.
He talks about the Observer debacle of 2007 and how that paper made a total hash of leaked data. At the time Baron-Cohen said:
The draft report, he says, “is as accurate as jottings in a notebook”
The big furore was that the paper was working on prevalence rates and mentioned a few rates of autism from 1 in 58 to 1 in 200. Guess which one the Observer decided was the more newsworthy? Baron-Cohen again:
Baron-Cohen says their study of Cambridgeshire children, which has been running for five years, comes out with a range of figures from one in 58, to one in 200, depending on various factors. The draft report, he says, “is as accurate as jottings in a notebook”. He adds that the data is with public health officials, who are crunching the numbers.
So it was really a total non-story. Some scientists were working on a paper about prevalence and had some unadjusted figures. At that point, someone in the team decided to leak the unadjusted report to the press who swallowed it hook, line and sinker.
Researchers now believe as many as one in 60 children has some form of the condition.
Up and down the blogosphere at the more credulous blogs and news pages the 1 in 60 figure is touted about. Indeed, only yesterday on the Age of Autism John Stone was claiming Baron-Cohen’s silence about the Daily mail article in his New Scientist piece as proof definite that the 1 in 60 figure was accurate:
The most significant thing about Simon Baron-Cohen’s recent New Scientist grouse about media irresponsibility and science (HERE) was that he did not mention the publication just a few days earlier in the Daily Mail of his latest – if long delayed – figure for the prevalence of autism in the UK school population of 1 in 60….
Given the importance of this figure – a true rate 66% higher than formerly acknowledged – the long term reticence of Baron-Cohen and the study’s sponsor Autism Speaks UK is dismaying – indeed Science Media Centre and Autism Speaks UK were still apparently trying to deny it to the Mail ahead of publication of the article.* But the silence of all these parties, and most particularly of Baron-Cohen after the Daily Mail article came out suggests that they did not have a leg to stand on.
When John Stone speaks with certainty in his voice, that is a certain clue that something is not as it seems. So I decided to email Professor Baron-Cohen to get his take on the Daily Mail. After a chat about what it meant he obliged me with an exclusive quote:
The Daily Mail was irresponsible in reporting on the results of our study before it was published in a scientific, peer-reviewed journal and where the details of the study are publicly available for scrutiny. The study will be published in the British Journal of Psychiatry on June 1st 2009.
My own opinion based on the discussion we had and that quote is that the paper might not be quite what it appears from the abstract posted at IMFAR. At any rate Professor Baron-Cohen is right that the Daily Mail have – just like the Observer – acted very irresponsibly. So are those that are reporting a UK prevalence of 1 in 60. Irresponsible and very previous.