FRIDAY, 26 JUNE 2015
The claims made are based on real clinical trial data, however the study was underpowered (with only 15 participants) and subjected to “p-hacking” (where adjustments made to study design or data to ensure significant results). Usually something unintentionally done by scientists, these studies are mostly caught by peer-review – however some journals are still willing to publish these studies, especially if they are more financially than scientifically motivated.The chocolate paper was published without peer review in the International Archives of Medicine in April. The accompanying press release, which left out the number of study participants and the effect size, was simply copied by many journalists; who Bohannon says are to blame for the sensational stories that resulted:
“You have to know how to read a scientific paper—and actually bother to do it. For far too long, the people who cover this beat have treated it like gossip, echoing whatever they find in press releases. Hopefully our little experiment will make reporters and readers alike more sceptical.”
There has been some criticism of the stunt, saying that while the media should be doing due diligence, it’s unrealistic for journalists to research every press release in depth. Additionally, there is concern the deception has damaged the public’s trust of scientists, as well as science journalists, and that this was an extreme method of pointing out problems in science journalism.
Mainstream media reporting of science has changed greatly with access to free online news sources, content is released very quickly and not always fact checked – maybe this best serves as a reminder to readers that sensationalist headlines aren’t always what they appear.
Read the original paper (now retracted from the International Archives of Medicine) here: http://www.scribd.com/doc/266969860/Chocolate-causes-weight-loss
Read John Bohannon’s full post here: http://io9.com/i-fooled-millions-into-thinking-chocolate-helps-weight-1707251800
Written by Caitlin Stewart.
