Originally Posted by mrlittlejeans
I saw that. It implies that 24 people were able to accurately identify the two systems though in order to prefer one over the other. This would tend to argue against the fact that amps/cd players sound indistinguishable from each other when listening is done blind.
Statistically, it's called guessing. If one were to mark A in every box of a multiple choice test with three choices, if a hundred and twenty test questions evenly distributed, they'd be right 1/3 of the time which is what you have here; guessing. If they're right, they can say, see, I'm good. But if the test were done twenty times, those who correctly guessed the first time, would be eliminated in subsequent testing till maybe you had one or three out of a group of twenty people guessing 26%. The more tests done, the lower the guess rate; statistical average, IIRC is 17%. The more DBT done, the fewer individuals will have consistently made the desirable choice; process of elimination; meta studies. Would you go to a dentist who has an admitted one in four hit rate?
In the case of the aforementioned test, you had what seem to be, five individuals. And out of these five individuals you have 38 tests and "ONLY" 26% accurate identification of the better equipment, yet you have a much larger statistical number of fails, either not being able to make a choice or the "inferior" equipment being chosen. When a test shows 74% not being able to tell good gear from obviously "inferior" equipment with the poorest quality of interconnects and mountings, this point and this point alone should cause one's eyes to open. The test results are consistent with similar tests I've read over the last twenty years.
It never hurts to know what each participant's hit rate was.
Everybody should spend their money anyway of their choosing but friendly audio-heads encourage fellow audio-heads to do so with their eyes open. After all, it's the civil thing to do.
-Edited by BeeMan458 - 8/13/12 at 2:01pm