Wow. Ouch.
All I'm saying is, if you're just looking for a group of folks to state their opinions, and you clearly state how the test was conducted, how is that "incorrect"? I'd agree if there was a specific technical point you were trying to prove, and the test was done poorly to that end, but I think that lies simply in the goals of the test.
I think I'm with Ken on this one, FWIW. Questions directed solely at the opinions of relatively small groups are very unlikely to be best served by a double-blind method, which is appropriate for asking focused, often "technical", questions, and powered by large numbers. Asking a couple dozen people whether they can hear a difference between cord A and cord B, even under double-blinded conditions, is very unlikely to provide a result with a usefully narrow confidence interval. It would be better to do something like what Bongomania did some time ago, in actually measuring the signal input and output of various cords and comparing them.
A well-designed qualitative study is much more likely to provide useful answers to questions involving multiple and complexly interactive variables like tonal preference, for example.
