“For every anecdote, there is an equal and opposite anecdote”
“The plural of anectode is not data”

Anecdotes are useful for generating hypotheses but useless for testing hypotheses.

Now that the good doctor has supplied us with the intellectual ammo, it’s time we start shooting down anecdotal ‘evidence’ wherever it rears its fallacious head.

Post duly linked to and promoted: http://the-attempts.blogspot.com/2010/12/plural-of-anecdote-is-anecdotes-not.html

And so are these two quotes:
“For every anecdote, there is an equal and opposite anecdote”
“The plural of anectode is not data”

Steve, do you mind if I translate it to portuguese and post it with linkback?

I think you don’t read older posts comments, so I’ll email you later with that request.

And Elmer,
Perhaps those weren’t the best examples, but they were the first things that came to mind. I was just using them to try and gain some tools for rational discourse in the future with people who advocate for whatever unconfirmed remedy/theory/what have you with a stockpile of anecdotes.

Seems to me that acupuncture and homeopathy are not the most relevant examples, since, beyond having mere anecdotal evidence to back them up, there is good evidence showing that they don’t work (I still think some kinds of dry needling that haven’t been studied adequately could possibly have non-placebo effects, but presumably we’ll see, eventually). Of course, many treatmnets commonly practiced by doctors have only anecdotal evidence to back them up.

The witnesses to Joseph Smith’s miraculous golden tablets were three people, because if it was just one guy nobody would’ve believed him.

I like that. It’s an easy, concise concept to remember. And I now feel better equipped to recognize confirmation bias.
Thank you both for the clarification.

In other words – millions and even billions of people can be systematically wrong.

RedQueen – I would say no, but the separation of discovery and confirmation is simply just another demarcation problem. Most such dichotomies are, in fact, continua – but that does not mean they are not real. There is a difference between science and pseudoscience, even though there is not bright line between the two but rather a continuum.

So while many scientific activities combine or blur the lines between discovery and validation, that does not mean these are not two distinct concepts that need to be understood.

..when the same observations start to occur in large numbers…does that lend any evidentiary strength to the claims that are being made?

The plural of anecdotes is not data.
However, a large number of independent subjective reports do strengthen the case for subjecting it to scientific study.