RSS
The Guardian
June 2, 2015
by Roger Pielke, Jr.
The past few weeks have seen some remarkable episodes in science.
Through a hoax, evocative of the Sokal Affair of the mid-1990s, John Bohannon showed how trivially easy it is to start a popular meme based on science. Bohannon ginned up a fake study showing that eating chocolate leads to weight loss, got it published and then was able to promote it onto the pages of several newspapers and television news outlets.
Far more significant than the hoax was the unraveling of a major study published in Science by Michael LaCour and Donald Green. LaCour and Green found that a single conversation with activists on the subject of same-sex marriage was “capable of producing a cascade of opinion change.”
The study was celebrated by major media across the United States, just talking to people who were until that point opposed to same-sex marriage was apparently enough to change their minds, leading to political change. The New York Times was quick to generalize the paper: “The findings could have implications for activists and issues across the political spectrum, experts said.”
Unfortunately, LaCour and Green was too good to be true. Last week Science retracted the paper, based on irregularities and false claims. The retraction led to a series of corrections among US media giants, including The New York Times, the Washington Post and National Public Radio, which had trumpeted the paper’s conclusions when it was released.
The Bohannon hoax and LaCour/Green retraction have a lot in common. Scientific research was manufactured, which resulted in claims that appealed to some popular views, and the media broadly and uncritically promoted the results, advocating popular actions in response.
These two episodes highlight a more general problem: a lot of nonsense is published in the name of science. Writing in The Lancet last month, editor Richard Horton argued that as much as half of all scientific papers may simply be “untrue.” He writes: “The apparent endemicity of bad research behaviour is alarming. In their quest for telling a compelling story, scientists too often sculpt data to fit their preferred theory of the world.” The media, journal editors and universities also share blame, he wrote.
The hoax and retraction should help us to understand that a big part of the problem that Horton laments has little to do with research misconduct or fraud, though that is a problem too, but rather science working exactly as it should. Read more …
