We all know that people tend to learn things through experience, for better or worse. Many times, that process achieves good results; other times, selective memories or unaccounted factors lead to shaky interpretations and bad "knowledge". However, our views and thinking are also shaped heavily by our peers. Especially in Internet communities like this one, people flock to those of similar mind and viewpoints. Groups are self selecting and maintaining. There, they get a distorted viewpoint of what percentage of people it is that really agree with them. I don't mean to bring up politics, but that's a good example: a lot of times, people think that all their friends and most everybody they know voted _____, so how could _____ win instead? Or some such.
Of course nobody knows the real answer, but I'm wondering what percentage of people stopping by something like head-fi sees some of the discussions, decides that they're dealing with a bunch of loons (rightly or wrongly), and never bothers posting. Personally, in real life and other forums, I know of people who don't bother posting but just check the headphones / IEMs sections for some opinions whenever they want to buy something and ignore the rest... but those are just anecdotes, not statistics. I would assume that those liable to believe in large effects of *insert audio product or situation here, e.g. demagnetizing strip X* are more likely to sign up and stay. For example, if you wanted to figure out how many sets of headphones the average person has (even among those who own at least one), sampling head-fi is very obviously a biased slice of the population—even more so if you look at posts, as those with more posts are liable to be repeat purchasers or otherwise have been around for a while.
This kind of goes with audio manufacturers too. The designers at boutique and audiophile-oriented companies are much more likely to believe certain things themselves, have doubts about tests and thinking that diminish the value of what they work on, and so on. Especially if you ignore those working on audio in the chip designers, pro audio, and low-margin home theater receivers and the like (though many of these would be included), they probably think that what they're doing is important and makes an audible difference—assuming that, without rigorously testing otherwise, because what would that help? Otherwise, it'd be harder to find the motivation and interest to continue.
In the thread that Currawong locked:
What kind of engineers were we talking about? Certainly not the ones posting in Sound Science. Probably not university profs and so on. How about the scientists? You'll find differing viewpoints if you look at different groups. It's just hard to get a handle on the percentages.
Edited by mikeaj - 6/9/13 at 7:38am