The thread starts: I've noticed on this forum a segment that demands blind testing as a means of verifying reports of audibility with "scientific" evidence. The reason for demands of blind tests is to rule out the false positives that sighted listening is known to be prone to. It has long been recognised that blind test could well be prone to false negatives & ITU standards recommend inclusion of controls to cater for such known possibilities. I have suggested, on a number of occasions, the inclusion of internal controls in blind tests but have never seen anyone here including these controls within their tests.
Recently, I was told that Foobar ABX (once a gold standard for blind testing but no longer) had been tested for false negatives but when asked for the links to these tests I was met with evasion. I have not found any evidence myself in my searches. I'm wondering why people who proclaim to want a "scientific" approach are so reluctant to submit their tests to these internal controls & reluctant to verify that the tests they clamour for are actually of value in revealing false positives?
So, let's for a moment switch from this overarching focus on false positives & consider false negatives. What are people's view on how important it is to ensure a test doesn't suffer from false negatives? What evidence is there of the use of such internal positive & negative controls in the tests that are often suggested that will provide evidence?
Please click HERE to continue reading