Quackery over the US Labor Day weekend
This is reposted in full from the Lancet blog post of Aug 24th, “‘Quackometer’ probes websites for pseudoscience“.
Andy Lewis of Oxford, UK wanted to see if it was possible to identify alternative medicine websites simply by analysing their language, and so began his study of “quackometrics”.
The result is his website http://www.quackometer.net where his “Quackometer” will analyse a website you’re curious about and rank it on a scale of 0 to 10 “Canards”, which Lewis maintains is the “internationally accepted SI unit for quackery.”
Practitioners of alternative medicine, Lewis says, tend to pepper their pitches with the same pseudoscientific language, terms like: “energy”, “vibrations”, “holistic”, and “quantum”.
So, “for a bit of fun” while he was between jobs, Lewis, who now works for a computer software company in Venice, “coded up an algorithm” that analyses the language used on a website, comparing the terms used with terms in six dictionaries, including one that identifies legitimate scientific terminology.
The programme then scores the site based on the ratio of pseudoscientific to scientific terms.
“Some terms, like aromatherapy, are dead obvious”, Lewis says, but others, like “quantum” are also used in legitimate scientific documents, making it harder to rate some sites.
The hardest sites to score, Lewis says, are newspaper stories on alternative medicine, which, although they might be skeptical, use many of the terms favoured by proponents of alternative medicine.
Lewis estimates the Quackometer currently gets it right only about 80% of the time.
So how does the Quackometer do?
Well, first I ran the website for the National Center for Complementary and Alternative Medicine at the US National Institutes of Health.
Score? “0” Canards
Although the site is laced with alternative medicine terms, the Quackometer gave the NCCAM’s site a “0″ score.
Why? It also detected the site’s skepticism:
“This web site has more quackery than my village pond.” the Quackometer says. “It is full of scientific jargon that is out of place and probably doesn’t know the meaning of any of the terms. However, the black duck can spot a fellow sceptic! The site is highly sceptical in language and is debunking.”
Next I ran The Lancet. Score: “0”
New England Journal of Medicine. Score: “0”
BMJ. Score: “2”
What, 2 Canards for the BMJ?
Asked about it, Lewis said the BMJ’s “2” score may have been due to an article on acupuncture they were running at the time.
And indeed, a retest today finds the BMJ back in the mainstream with a score of “0”.
To visit the Quackometer and run your own tests go to: www.quackometer.net