TiredLegs;228181 Wrote: > Of course, you are correct in the case of someone scoring 10/10, but in > that case, nobody needs statistics. The minimum number of 30 is for a > general case, averaged out over all the possible outcomes.
I'm not sure what you're trying to say here, but I'm pretty sure it's wrong :-). Can you clarify? Where the heck did the number 30 come from? andy74 Wrote: > > Objectively, if a 100 persons does not hear the difference in the test > of 50 trials. And there is one who does all 50 tests with 100% right > answers. This means that there is difference doesn't it. And it also > means that those 100 persons do not simply hear it. Yes, you're right - it does mean that. The odds of anyone randomly getting 50/50 correct are about one in a thousand trillion. And that would be seen immediately in anything other than the most cursory statistical analysis of the results. The best thing to do then would be to isolate that person, and anyone else that did exceptionally well, and test them on their own. Most, if not all, blind audio test experiments are well aware of this "golden ear" possibility and look for it. -- opaqueice ------------------------------------------------------------------------ opaqueice's Profile: http://forums.slimdevices.com/member.php?userid=4234 View this thread: http://forums.slimdevices.com/showthread.php?t=38258 _______________________________________________ audiophiles mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/audiophiles
