Themis wrote: > I don't think so. Your assumptions would be true if *actually* the > coffee was different 100% of the time... > But the test doesn't claim this. So, if the coffee prepared by the two > machines was different 50% of the time, the conclusion is that the > audience succeded at 100% of the samples.
What the test wanted to show has nothing to do with what it shows. It shows exactly a random selection between the two machines. There is nothing about the cause of the selection in the information provided. The result is exactly the same as you would expect of coin flips. -- Pat Farrell http://www.pfarrell.com/ _______________________________________________ audiophiles mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/audiophiles
