It's a good idea to blur the lines to test the candidate's thinking, but I'd disagree that the correct answer can be (say) 60% correct. What about 30% correct? What about 90% correct? Just where is this line drawn? What chance does a candidate have in picking the "most correct answer" if the candidate's thinking has been skewed by an incorrect model?
The question must be set up so that any knowledgeable person will pick the correct answer. I agree that all questions should have answers that are defendable and correct, and I agree that some of the FAA questions are unfairly vague.
I did look to the FAA-published PHAK and AFH, and they support the theories to support the questions. Might I suggest that that is where the problem lies?
As far as the rest of the post goes, I think we agree to the same principles
