AmberEyes wrote:
I'm starting to doubt the scientific validity of this test.
I would answer some of the questions very differently in different situations and the test doesn't take different styles of human communication into account. Some people feel connect better socially with some people than others. Sometimes superficial and friendly extraversion can ignore the deeper issues.
According to one study, there does seem to be a correlation between having an ASD and scoring higher on the AQ:
http://en.wikipedia.org/wiki/Autism_Spectrum_Quotient
That said, it isn't intended to be a diagnostic test. My scores tend to fluctuate between being borderline NT (around 34) and being firmly in the ASD range (41). My biggest problem is that there's no neutral "neither agree nor disagree" choice; for a few of the items, I tend to oscillate on whether I agree or disagree with the statements depending on how I feel when I take it.