None of the ADHD self report surveys I've seen appears to have done in consultation with a reasonable number of actual ADHDers representing the entire spectrum of symptoms and co-morbidities.
It's long occurred to me that asking a person with issues relating to memory how frequently they forget things is only going to get somewhat reliable answers when the person has decided to designate the subject as a compelling topic of interest.
But even if a person does manage to diligently note down their symptoms for a sufficient amount of time, having been a certain way their whole life, how are they supposed to know what "rarely", "sometimes" and "often" mean?
Sure, these things are mostly used as a basis for a conversation with a doctor, but a good doctor can be a rare privilege these days.
Despite concerns*, I do think it would be nice to have better, open licence (like PHQ-9 vs GAD-7), tools for self-diagnosis and more consistent diagnoses.
* Screening tools can lead some people to be denied the care they need because their symptoms don't fit a typical pattern, their symptoms are masked, etc (this is a whole different rant).