Well, its an experiment, albeit written by a 2nd or 3rd year undergrad, however it actually "proves" nothing. Its good, because its attempting to do something about the issue, and it has good face validity, but completely lacks construct validity. This isn't really a criticism, because the guy probably knows full well that it lacks construct validity and that's not his point. I'm pretty sure his point is to run the exercise of experimental design and do a little something in the process.
According to his own conclusions as well though, he finds in favour of the null hypothesis, that wood type in fact "does not contribute significantly to the sound of the amplified instrument".
I have no intention to turn this thread/post into a master class on experimental design, so I'm not going to break down the flaws point by point. Any social science undergrad who's done a basic stats class can do that easily.
Salman, I'm starting to think we're arguing for different aspects of the same point. We on the 'dark side' as you've put it, all concede that there is in fact a difference that can be assumed to be due to wood (if everything else is controlled for), however we also assert that the difference is negligible and likely not easily discerned by the human ear in a properly designed experiment.
So far, no one has been able to put together a coherent argument that proves (factually or logically) that our 'dark side' viewpoint is in fact wrong or incorrect. Most seem still to come over to this side after thinking about this quandary at length, and I think more are still on their way over.
On its face, it sure seems like wood would make a significant difference, however even the word "significant" carries with it operational definitions when you get into the science of it, and I'm still asserting (at the moment without solidly-based opposition) that the differences are very likely not going to be determined to be significant in the end, logically, or experimentally...
According to his own conclusions as well though, he finds in favour of the null hypothesis, that wood type in fact "does not contribute significantly to the sound of the amplified instrument".
I have no intention to turn this thread/post into a master class on experimental design, so I'm not going to break down the flaws point by point. Any social science undergrad who's done a basic stats class can do that easily.
Salman, I'm starting to think we're arguing for different aspects of the same point. We on the 'dark side' as you've put it, all concede that there is in fact a difference that can be assumed to be due to wood (if everything else is controlled for), however we also assert that the difference is negligible and likely not easily discerned by the human ear in a properly designed experiment.
So far, no one has been able to put together a coherent argument that proves (factually or logically) that our 'dark side' viewpoint is in fact wrong or incorrect. Most seem still to come over to this side after thinking about this quandary at length, and I think more are still on their way over.
On its face, it sure seems like wood would make a significant difference, however even the word "significant" carries with it operational definitions when you get into the science of it, and I'm still asserting (at the moment without solidly-based opposition) that the differences are very likely not going to be determined to be significant in the end, logically, or experimentally...
