It seems to me that much of the evaluations/criticisms/defenses of EA conflate the empirical phenomena (whatever you see on with the EA forum, whatever is happening at your local university group, CEA, EAGs, or whatever) with some broader idea of effective altruism (the philosophy stuff / possibly the normative stuff if you believe in that). I am uncertain how to define this “broader idea”, maybe I can call it some theoretical core, maybe I can call it Weber’s idealized type (a deliberately purified theoretical construct that no empirical instance fully matches), maybe I can say something about Wittgenstein and family resemblance (his refute to the demand that a concept must have a common essence), maybe we can each reference our own favorite theoretical definition of EA for the purpose of this point)

But I think never the less it is pretty important to distinguish the empirical manifestation of EA (the type of stuff you can do sociology on), and the theoretical core (to the extent that we can define some semi-coherent notion of such).