![]() ![]() |
Propeller Head Plaza Technical and scientific discussion of amps, cables and other topics. |
For Sale Ads |
Use this form to submit comments directly to the Asylum moderators for this forum. We're particularly interested in truly outstanding posts that might be added to our FAQs.You may also use this form to provide feedback or to call attention to messages that may be in violation of our content rules.
Original Message
RE: No pain, no gain!
Posted by EBradMeyer on September 27, 2007 at 07:40:47:
Axon:
Well, you were following the evidence, like a good experimenter. Of course that way is often tiresome and annoying. If your blind test method gives you the answer after each trial, then there is a possibility that you could affect the overall result by deciding to stop at a certain point, but there's a relatively easy way around that: Decide beforehand how many trials you're going to do, and stick to it. There's a mathematical rescue of a sort if your results are positive too. A score of 10/10 is quite conclusive, and you can stop there.
Yes, you could get 8/10, be disappointed, and choose to go on to try for a better score, which is a bit slippery. But if you miss any at all, you know the effect is at least not obvious. If you do another ten, and again get 7 or 8, you're in that gray area, where effects probably make themselves heard, but not all the time. Remember too that in the case of polarity, asymmetrical distortion in your playback transducer aids audibility, so the device on which you can't hear it may be a better reproducer, not a less revealing one.
Finally, about that hypothetical 8/10 score: The usual confidence level that is asked for is 95%, and 8/10 is 94.5%, so that one is close enough unless you're writing it up for publication. After getting 8/10, I'd also call it fair to do another ten and total them -- but not to do another 10 and throw it out if you do less well. It's pretty easy to grasp what is cheating here and what isn't. -- E. Brad