Comment by waterhouse
Comment by waterhouse 2 days ago
I have encountered this definition of "spectrum", as a vector of numbers that go 0 to 100, rather than a single number that goes from 0 to 100 (which you call "continuum" IIUC).
But... I mean, if you asked 100 people what they think a spectrum means in this context, how many of them would think it meant "vector" rather than "real number"? I would guess fewer than 10. I consider myself a fairly well-informed nerd, but I think I had encountered many usages of "spectrum" describing a single trait for many years, and I think this is the second time I've ever encountered someone using the "vector" definition (the first one was also using it to describe autism). Has this linguistic battle already been lost? Does it improve clarity to call it a "spectrum" and insist on using the "vector" definition?
(I've personally been using the phrase "collection of imperfectly correlated traits")
I think the word "spectrum" is reasonable, as it implies a broad range. Or it's analogous to the rainbow with a variety of colours.
But what people consistently misunderstand is that there is a fundamental dichotomy at the diagnostic level. Speaking from the perspective of the DSM, which I prefer because it's at least concrete and has medical relevance in North America, you meet the criteria for Autism Spectrum Disorder ("on the spectrum") or you do not ("not on the spectrum").
In other words, the diagnostic criteria themselves do not constitute a spectrum, especially not a linear one. Maybe people are confusing this with the DSM's three levels of support needs.