Once again, we see that there is no understanding of what the proof of a universal negative requires in a scientific setting, and especially when probability is involved.
I notice your multiple attempts at intimidation, and your relentless Gish Gallop here, but you still haven't shown any understanding of what "proof" means in a scientific setting, let alone how one would have to prove a universal negative.
Note, UNIVERSAL negative. Some negatives can be proven, but that's not germane to any subjective test.
The problem is simple. Let us take a test (sighted, dbt, whatever) with the following results:
10 of 10 answers were right. The choice is binary, A or B. Now, what is the probability of that happening by pure chance? What's the chance it wasn't? If it wasn't, what's the subject's actual chance of a right answer? Is it 1? No, it's not. You don't know exactly, and you can't know exactly. Ever.
Ok, now we have an answer of 6 of 10, correct. What is the probability of that arising from pure chance? Yes, pretty big, but does that prove anything absolutely? No, it does not.
What is the probability of that arising from a very faint detection mechanism that just barely works (i.e. a threshold effect)? It's possible, but how do you separate that from pure random results? How many tests must you run?
Now, let us say that the chance of a right answer for a very hard threshold effect is .51, that's not random, that does mean that one can detect something, sometimes, i.e. that the negative is not PROVEN.
Now, how does one run a test detect that, and to rule out a false negative with 100.00% assurance?
Ok, now we go to 50.001% right answers. How does one prove that's not happening?
Look up what type 1 and type 2 error are before you disgrace yourself any further. Your entire set of protestations is completely ignoring the real, actual problem at hand. Again, stick to the law, and let the scientists stick to science, ok?
I notice your multiple attempts at intimidation, and your relentless Gish Gallop here, but you still haven't shown any understanding of what "proof" means in a scientific setting, let alone how one would have to prove a universal negative.
Note, UNIVERSAL negative. Some negatives can be proven, but that's not germane to any subjective test.
The problem is simple. Let us take a test (sighted, dbt, whatever) with the following results:
10 of 10 answers were right. The choice is binary, A or B. Now, what is the probability of that happening by pure chance? What's the chance it wasn't? If it wasn't, what's the subject's actual chance of a right answer? Is it 1? No, it's not. You don't know exactly, and you can't know exactly. Ever.
Ok, now we have an answer of 6 of 10, correct. What is the probability of that arising from pure chance? Yes, pretty big, but does that prove anything absolutely? No, it does not.
What is the probability of that arising from a very faint detection mechanism that just barely works (i.e. a threshold effect)? It's possible, but how do you separate that from pure random results? How many tests must you run?
Now, let us say that the chance of a right answer for a very hard threshold effect is .51, that's not random, that does mean that one can detect something, sometimes, i.e. that the negative is not PROVEN.
Now, how does one run a test detect that, and to rule out a false negative with 100.00% assurance?
Ok, now we go to 50.001% right answers. How does one prove that's not happening?
Look up what type 1 and type 2 error are before you disgrace yourself any further. Your entire set of protestations is completely ignoring the real, actual problem at hand. Again, stick to the law, and let the scientists stick to science, ok?
