Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Double blind? But how? (Read 6100 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Double blind? But how?

From wikipedia:

"In a double-blind experiment, neither the individuals nor the researchers know who belongs to the control group and the experimental group."

How is it possible to call an ABX-test double blind then?

(I'm sure it's possible, but I fail to see the connection)



The wikipedia entry:
http://en.wikipedia.org/wiki/Double_blind#...le-blind_trials

Double blind? But how?

Reply #1 –
Well, looks like I just found the answer (by coincidence really =):

From ITU BS.1116-1:

"Blind test

A test in which the only source of information for the subject about the trials is the stimuli.

Double blind test

A blind test in which there is no possibility of uncontrolled interactions between experimenter and the listening test."

Double blind? But how?

Reply #2 –
Again, we can't take everything we read on wikipedia word for word as credible source of information.  The wikipedia quote is correct, double-blind "neither the researchers nor the individuals know about it". If it doesn't come from stimuli then it's usually given by someone else, who is neither.  ABX in this context is slightly different. Usually controlled ABX experiments are performed in medical tests though.
budding I.T professional

Double blind? But how?

Reply #3 –
In audio codec ABX tests, the control group is the set of "X" stimuli that were assigned to the original file by the randomization, and the experimental group is the set of X stimuli that were assigned to the encoded file.
When you perform an ABX test with your computer, you are the individual, and the computer is the researcher. There are no uncontrolled interactions between you and the machine. Since the computer is not human, it cannot give you hints about what trials have got X in the experimental group, and what trials have got it in the control group.
Therefore the test is double blind

Double blind? But how?

Reply #4 –
Tip: If Wikipedia disagrees with someone, ignore Wikipedia.

Double blind? But how?

Reply #6 –
Quote from: status_brun on
Quote from: tgoose on

Tip: If Wikipedia disagrees with someone, ignore Wikipedia.


hehe.. yeah..


I guess I'm missing something.  Explaining why the wikipedia entry isn't complete in the context of a computer-administered ABX test is hardly ignoring wikipedia.  If there was an HA lab where guys in white coats administered the listening tests, wikipedia is an entirely accurate way of describing what would need to happen for those tests to be double blind.

Double blind tests are designed to address the issue that come up years ago when some scientists thought that they had found a (I believe) horse that supposedly could do math.  They would show the horse an addition problem and it would tap out the correct answer.  It turned out that the horse was reacting to cues from one of the researchers and it would stop when it reached the correct number of taps based not on .  The horse made the association between responding to that cue and being given a reward.

Double blind? But how?

Reply #7 –
Quote from: Danimal on
Double blind tests are designed to address the issue that come up years ago when some scientists thought that they had found a (I believe) horse that supposedly could do math.  They would show the horse an addition problem and it would tap out the correct answer.  It turned out that the horse was reacting to cues from one of the researchers and it would stop when it reached the correct number of taps based not on .  The horse made the association between responding to that cue and being given a reward.



That horse would be the famous 'Clever Hans'

http://skepdic.com/cleverhans.html