HiFi WigWam Power Cable Test
Reply #13 – 2005-10-06 21:28:45
So what would pass muster around here then how would you carry out tests ?[{POST_SNAPBACK}][/a] Here are guidelines that a blind listening test should comply with : [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=16295]http://www.hydrogenaudio.org/forums/index....showtopic=16295[/url] And here is an example of such a blind test. The only drawback is that it's not double blind : http://www.hometheaterhifi.com/volume_11_4...ds-12-2004.html The main problem of the WigWam power cable test is that the objective choice that the listeners had to make (difference or no difference) was not about a random parameter. Thus we can't evaluate their expected rate of success, called "p", if the cables make no sonic differences. In order to analyze the listeners preference and see if they are significant, I've given a rating for each cable : the standard one is rated 2, the other ones are rated 1 if they are judged inferior, 2 if they sound the same, 3 if they sound better. Which gives the matrix :VDH YEL MAM SOS 25 STD 1 3 3 3 3 2 1 3 1 3 3 2 2 1 1 2 1 2 And I ran a Tukey HSD analysis on it :FRIEDMAN version 1.24 (Jan 17, 2002) http://ff123.net/ Tukey HSD analysis Number of listeners: 3 Critical significance: 0.10 Tukey's HSD: 1.940 Means: SOS YEL 25 STD MAM VDH 2.67 2.33 2.33 2.00 1.67 1.33 -------------------------- Difference Matrix -------------------------- YEL 25 STD MAM VDH SOS 0.333 0.333 0.667 1.000 1.333 YEL 0.000 0.333 0.667 1.000 25 0.333 0.667 1.000 STD 0.333 0.667 MAM 0.333 ----------------------------------------------------------------------- Even with the loose 0.10 threshold, the data are not significant at all. Answers are just like if they were random. Otherwise, it would be stated in the report.