Not really shardorne, it’s a bullcrap statement.
The test has to take place in a system familiar and local to the person committed to being the hearing test component. meaning their own system and one single change, and no pile of people standing around or in the room browbeating them with stares and so on. perfect known comfortable environment and system, one single change, with time to acclimate to the changes of each cable. To get to know what to listen for.
and, the first day, maybe one or two test tuns of 20 might be doable. Then, one or two more like that per day.
Properly handled, it could then likely go to 10-20 sets of 20 changes of a or b, and then under such a scenario a statistically significant number should arise.
A single run of 20 and asking for 18 right is not any known form of testing (of this sort of nature) that I’m aware of. It’s not a proper trial.
Repeating that set at least 20 times, begins to end up looking more like a proper trial. A valid trial. Statistically significant. Under this 400 individual tests in total, a result of 52-53% correct is statistically significant. And definitely considered valid. 1000 tests is better. At 1000 tests, 51% correct is valid.
Asking for 18 out of 20 to be correct, in a single run, is totally invalid. Not scientific at all.
Even though I think it can be pulled off.
And if he did ’nail it’ and pass the test, the non believers would never change their stance.
Not a hope in hell of that happening.
They would each have to do the test themselves with the same or different guy.
Thus the information gleaned is not transferable due to the people involved, people who will never accept the data.
Psychologically speaking...each skull will individually have to beaten off the floor until they each individually break down and deal with it. And they would still find a way to rise up and try and find a way back to their comfort zone by second guessing the results.
There’s no winning here. The problem is not the data, it’s the people involved.
The test has to take place in a system familiar and local to the person committed to being the hearing test component. meaning their own system and one single change, and no pile of people standing around or in the room browbeating them with stares and so on. perfect known comfortable environment and system, one single change, with time to acclimate to the changes of each cable. To get to know what to listen for.
and, the first day, maybe one or two test tuns of 20 might be doable. Then, one or two more like that per day.
Properly handled, it could then likely go to 10-20 sets of 20 changes of a or b, and then under such a scenario a statistically significant number should arise.
A single run of 20 and asking for 18 right is not any known form of testing (of this sort of nature) that I’m aware of. It’s not a proper trial.
Repeating that set at least 20 times, begins to end up looking more like a proper trial. A valid trial. Statistically significant. Under this 400 individual tests in total, a result of 52-53% correct is statistically significant. And definitely considered valid. 1000 tests is better. At 1000 tests, 51% correct is valid.
Asking for 18 out of 20 to be correct, in a single run, is totally invalid. Not scientific at all.
Even though I think it can be pulled off.
And if he did ’nail it’ and pass the test, the non believers would never change their stance.
Not a hope in hell of that happening.
They would each have to do the test themselves with the same or different guy.
Thus the information gleaned is not transferable due to the people involved, people who will never accept the data.
Psychologically speaking...each skull will individually have to beaten off the floor until they each individually break down and deal with it. And they would still find a way to rise up and try and find a way back to their comfort zone by second guessing the results.
There’s no winning here. The problem is not the data, it’s the people involved.