I am coming to the conclusion, however, that the rule of "if it doesn't fit the model, it is not data, and if it IS data, it damn well WILL fit the model" is not going to relax on this thread, and similar ones we have had before.
Amos, the protocol of science is that, "If the data do not fit the model, the model may need adjusting, or the data may need refining or expansion."
One does not throw the baby model out with the bathwater just because a stain is observed to have escaped the washing process.
Leaving my rather fanciful metaphor, what we are involved with here, at least in part, is the question of "What constitutes data?" The great tradition of a number of centuries, which has served the cause of knowledge admirably, is that data must be measurable and reproducible--which is to say, verifiable.
Hearsay reports of individual personal experiences are extremely difficult to investigate and verify. If they fly in the face of the model, they are indeed not data.