First: How do I know that these results are results that I want? If I know clearly enough what I want, and I know clearly enough what I’m observing, then I think your question comes down to how to compare the two.

Every now and then, I’m tempted to put that comparison into the model explicitly, phrased like your question. But mostly I haven’t notice that the comparison per se is the problem. When people have trouble deciding whether they’re getting what they want, it’s usually (and maybe always) because they are unclear about either what they want or what they are getting.

Another way to interpret your question: I do I know that these results are my creation? That’s a much tougher question. And by “tougher” I mean that I haven’t the slightest idea how to answer it (that’s my criterion for toughness :-).