|
General Asylum General audio topics that don't fit into specific categories. |
|
|
|
General Asylum General audio topics that don't fit into specific categories. |
|
In Reply to: Re: it's a pretty good test but it has a big problem posted by PhilNYC on August 20, 2006 at 19:51:34:
Your idea is sound but it could be improved.In any test, there is a possiblity that the results could be due to pure chance. In a well designed test, you want to decrease this possiblity so that it is negligible.
You want a strongly defined pass/fail. Indeed if you see some peoples' hands go up immediately after the change, this is strongly suggestive that they really heard something. But why only half the group? And then, if other hands go up later, how long is this really considered a positive?
The whole test description is rather confusing. If 1/2 the people heard the change immediately, did you keep track if these were always the same 6 people or not? If they were, that indicates that different people may have a varying ability to hear this effect. If they weren't, the statistics start to look a lot more like just random guessing.
In a more controlled test, if the tweak is either ON or OFF, you can keep track of each person over 10 trials. Then if any particular person has gotten all 10 correct, you know that he had only a 1 in 1000 chance of doing that randomly. And that's pretty good evidence that this particular person is really hearing the tweak.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: it's a pretty good test but it has a big problem - tunenut 08/20/0621:02:50 08/20/06 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
