garym;585651 Wrote: > magiccarpetride: Did you do an ABX of the two samples. How many trials > and what were the results. Foobar2000 has a nice utility for doing ABX > trials. One can easily avoid the expectation bias issue with a ABX test > and enough trials. If you tell me that you could pick one version over > the other, say 70-80% of the time (in a blind ABX test) over, say, 10 > trials, I certainly wouldn't argue that your results are expectation > bias. edit: of course, I'm assuming here that the null hypothesis is > no difference.
I don't subscribe to the phony ABX methodology. It introduces its own expectation bias into the equation, increases the stress level on the subject, and these factors in the end skew up the final results. Nothing solid and conclusive could ever come out of that witch doctor approach. -- magiccarpetride ------------------------------------------------------------------------ magiccarpetride's Profile: http://forums.slimdevices.com/member.php?userid=37863 View this thread: http://forums.slimdevices.com/showthread.php?t=82870 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
