Well, perhaps I should have worded it differently. I think what I meant to say is that observation of possibly connected events leads to formulating a theory of the connection between those observed events and that an empiric approach to proving/disproving that theory is then worked out. It's always possible that there's a flaw in the logic anywhere along the line, but it makes a big difference if you have an observational opportunity as opposed to not having one.
My observational experience was initially limited to four individuals (my own children) but quickly expanded, in a sense, through talking to other parents of disabled (often autistic spectrum disordered) children. I guess you could say I started with an "anecdotal cluster" of my own four children.
I'm not sure what point you're contesting.
Your argument does not improve with verbosity.