03 Jun 2015 Why a journalist scammed the media into spreading bad chocolate science

Earlier this year, headlines around the world trumpeted an exciting bit of news that seemed too good to be true: “Eating chocolate … can even help you LOSE weight!” as Britain’s Daily Mail put it. From India to Australia and Texas to Germany, news organisations shared findings published in the International Archives of Medicine in late March. The problem? The study was real but based on pure junk science.

And the person behind it, John Bohannon, would be the first to tell you that.

In fact, that’s exactly what he did recently, in a story for io9 that’s gone viral. Bohannon, a science journalist who also has a PhD, lays out how he carried out an elaborate hoax to expose just how easily bad nutrition science gets disseminated in the mainstream media.

Diet science, Bohannon stresses, is still science – and reporters need to know how to cover it. “You have to know how to read a scientific paper — and actually bother to do it,” he writes. “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 skeptical.”

To be clear, the study involved was real — a randomised controlled trial. Bohannon and his partners, a German television reporter and his collaborator, really did recruit 16 people for a study on dieting. And they found that the ones who followed a low-carb diet and also ate a 1.5-ounce bar of dark chocolate daily lost weight faster than the control group that was dieting alone.

So what’s the problem, you ask? There are plenty.

For starters, as Bohannon explains in great detail, the study design itself was flawed — it had too few subjects, and the research measured too many factors, making it likelier that some random factor would appear to have statistical significance. (Bohannon does an excellent job of explaining the specifics himself.)

Then there’s the journal that published it — a so-called pay-for-play publication, which failed to carry out peer review of the findings. As Bohannon himself exposed in another sting for the journal Science a couple of years ago, there are lots of these publications that will publish bad research for a fee…..

NPR.org: Read the full article

Related reading:

I Fooled Millions Into Thinking Chocolate Helps Weight Loss. Here’s How.