For what it's worth: Telling the subjects to raise their hand when they noticed a change wasn't necessarily "unblinding" the test, as long as there was a possibility that no change would happen during the course of the test. Having multiple subjects in the same room, even with their eyes closed, pretty much ruined the test, though.
If you ever want to re-do this test, isolate your subjects (test one at a time), and perform multiple tests per subject (on the order of 12 or more), including tests where no change occurs. Come up with the criteria for a "positive" BEFORE the trial starts, such as raising a hand within 30 seconds of the tweak. Then consider any other hand-raising to be a false positive, even if it's at X+1 seconds. Otherwise you'll end up adjusting your criteria to fit your data, which makes the whole process suspect. Then at the end you'll have a nice big pile of data to work with. Then it will likely be a pretty good single-blind test. It will not be double-blind because you are in the room with the subjects, both to make the tweak and to observe the hand-raising (and ninja-like as you may believe yourself to be, your subjects may be able to sense you making the tweak). Invest in a one-way mirror if you want to try for double-blind. -- CatBus ------------------------------------------------------------------------ CatBus's Profile: http://forums.slimdevices.com/member.php?userid=7461 View this thread: http://forums.slimdevices.com/showthread.php?t=32466 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/lists/listinfo/audiophiles
