Quote:
|
Originally Posted by Wodgy
None of that is relevant.
What Factor was trying to say is that science proceeds by empirical verification. Disallowing empirical verification is the same thing as making untestable assertions; neither approach is in the realm of science. It's fine to question whether science is even useful or appropriate for assessing audio gear, but it's inappropriate to start calling unverifiable subjective assessment "science". |
Well if that is what factor was trying to say, perhaps he should have actually said it. What he actually said was promulgating the false notion that only dbt's are scientific. Depending upon the hypothesis being tested and the actual implementation of a single-blind or double-blind procedure, there is nothing inherently better about any type of experimental design the design is a function of the hypothesis to be tested.
No one is disallowing empirical verification. the point is that no experimental design, including dbt's ensures empirical verification. For example, if I could indeed detect a difference 100% of the time then it is also easy for me to make sure of a result of no siginificant difference because traditional hypothesis testing is set-up based upon controlling alpha, the type I error rate, not beta, the type II error rate. You then need to replicate the experiment with other listeners who may or may not alsso cheat.
My comments about other types of approaches to science is relevant because it is all too common on these and other forums to misrepresent anything other than a dbt as being unscientific. An observation can range from hearing a difference to measuring the a difference. Both are merely observations and both can be validly used in the context of a hypothesis test. You may put more credence in one versus the other or have other reasons to dismiss them, but neither is inherently more scientific than the other. Do not confuse the ability to measure something as the criteria for making a scientific test. The actual way something is measured has implications for the generalization of the hypothesis test (assuming that previous observations have led us to conclude that the type of measurement is valid for our hypothesis test).
I realize that most people only get exposed to the scientific method in general science courses, but do not assume that you are being taught the full range of how science actually operates in those courses. They are a simplification of the process, at best.
Oh. and BTW science actually proceeds by empirical falsification not empirical verification and if you don't realize that is much more than a semantic difference then yourscientific credibility is seriously in doubt.









