This is exactly it: both analysis and the theory are structures one builds to make sense of the “reality.” We built a theoretical framework so impose (our own sense of) order on the chaos so that we’d know how to sort through them, and we interpret the data that we collect via analysis through the lens of the theoretical framework. But neither theory nor the analysis is the “truth” en toto, but abstractions and simplifications thereof that we have created for our “convenience,” based on our backgrounds, prejudices, and desires. If properly done, our “theory” and analysis will contain a sliver (or more than that) of the truth, but also our own baggage. Where we need “tolerance” and “diversity” in intellectual pursuits is not that we should humor other people’s beliefs “even if they are silly,” but because we know our theories and analyses contain our baggage, their theories and analyses contain their baggage. Once we have both our stuff and their stuff, which contain some things in common (hopefully, mostly “the truth”), we can triangulate and sort out what the common denominators are while leaving aside both sets of baggage. Perhaps somewhat the opposite of the popular conception of “diversity”: we are partly wrong, they are partly wrong, but if we are seeing the same thing, our perspectives will contain common shreds of the truth. But identifying where we are wrong AND where they are wrong is central. Much of data work in many fields today, though, seem to dispense with this triangulatory step.