jimmysnyder said:
The brief one has more questions than the real one.
Keep in mind that the diagnostic criteria is not a questionnaire, nor designed to be self-diagnostic. A real diagnosis, with the criteria sheet from the CDC, may take hours. Just getting the results back may take a couple of months, for some. Then they have to interview parents, etc. Then it costs money. At least that's what those diagnosed as adults say.
Keep in mind the diagnostic sheet on the CDC's website isn't a questionnaire for patients, but rather it's diagnostic criteria. I guess someone can just tell themselves whether they fit the criteria, however there's always the possibility that patients won't know if the symptoms are severe enough to have something. Keep in mind that having symptoms, in self-diagnosis, is not enough. Everyone has symptoms. One must have it at the clinical impairment level for those areas. The Autism-Spectrum Quotient test is brief in the sense that an actual clinical diagnosis will take much time and cost something.
Consider that the AQ test has been tested across different cultures and people, results in Autism peer-review journals, and has been found to be a good brief assessment, even if not as accurate as a formal clinician diagnosis.
jimmysnyder said:
The nature of the questions would be expected to screen for autism. However, the questions require an kind of inward examination that my own son would find difficult and he's on the spectrum. There is also an unnecessary unevenness in the questions that leaves me wondering how much thought went into them. Here's a pair:
4 I frequently get so strongly absorbed in one thing that I lose sight of other things.
23 I notice patterns in things all the time.
Why the words 'frequently', 'tend to', 'usually', and 'often' in the other questions, but 'all the time' in question 23?.
It's not meant to be reader friendly for everyone on the spectrum. It's meant specifically for AS and HFA, rather than all with ASD. Even if they can't reflect on their emotions as well, and even if some questions are vague, when 80% of those diagnosed score above a 32 and only 2% of controls score above, how is that to be explained? Even if some questions are vague or not worded well, looking at forest from the trees, if it does a good job of telling the difference, don't you think that's what matters?