|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
96.239.98.82
In Reply to: RE: Nobody In Audio Does True DBT .......... posted by Todd Krieger on December 03, 2014 at 09:09:29
I think you are not using that information correctly. The reference clearly refers to a "double-blind, placebo-controlled study."Two separate things. There is no necessity for a separate group to test a placebo in audio as there is in medical/physiological studies. In the latter, exposure to the DUT can change the response to the placebo and vice versa. One cannot untake a drug or procedure.
Also, double-blind tests do not restrict the type of testing. They only require, by definition, that neither the subject nor the tester be aware of the identity of the DUTs until the conclusion of the test.
Kal (Assoc.Prof. of Neuroscience and Physiology at the referenced instituion)
Edits: 12/03/14Follow Ups:
...but Kal, as you know, true DBTs were designed in England in the 1950s as a way to test new drugs.
They have been very poorly adapted to audio component difference testing.
Science requires a valid test to have all of the variables controlled except the one being tested for.
Since the subject's critical listening skills, test taking abilities, listening experience and even mood can affect the results, isn't it more of a listener test than an audio component difference test?
Individual subject variables are noise if there is a sufficient number of subjects.Let's agree that the DBT is only one aspect, albeit a critical one, of experimental design.
Edits: 12/03/14
...but I have never seen an audio component DBT with enough subjects or enough trials to make it anywhere close to scientific.
Just parlor games verging on pseudoscience.
The problem is that the cost of doing a proper study is very high and I cannot think of an institution that can fund it............and cares enough about the topic to do it. ;-)
"There is no necessity for a separate group to test a placebo in audio as there is in medical/physiological studies."
They do placebo-controlled studies with other endeavors, including automotive testing. The key is the test subject not being aware there is anything being compared to. Once the test subject become aware that a comparison might be taking place, there will be bias. (He might then think he might be evaluating a placebo. This information should not be given to him.)
"In the latter, exposure to the DUT can change the response to the placebo and vice versa. One cannot untake a drug or procedure."
Exactly.
From the perspective of the test subject, he might be sent an audio component for evaluation. He evaluates it, then sends it back, with his impressions. The key is that he has no idea that the item might be a placebo product, and he has no idea that there is such a comparison taking place. (Same goes for the individual who packed and shipped the item.) He just knows he received a product to evaluate, and he evaluates it. He might even become interested in the product if it were to come to market, even if it turns out he evaluated a placebo. (Although if he liked what turned out to be a placebo, he might not like the real product.)
"Also, double-blind tests do not restrict the type of testing. They only require, by definition, that neither the subject nor the tester be aware of the identity of the DUTs until the conclusion of the test."
True.... But once either individual becomes aware that a comparison is taking place, all bets are off.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: