Earlier (p440) suggested that you choose your approach and try not to do both simultaneously - talking about whether to organize data by question or by case study.
I am not sure which way to organize mine. Well that's not true. It's obvious that i will organize it by question - my analytic framework. But having said that, I can see that there is a great richness in seeing the cases as case studies too - not full blow case studies, but as individual "systems" if you like where the purpose-practice-community is coherent and cohesive. I think that may be an interesting level of triangulation to compare the individual systems with the systems that are suggested from looking at the questions.
By the way, increasingly i am thinking that the questions about phase of development, institutional form and CoP competence are not very useful in this thesis, but more useful for the other expected output of this study - a report outlining possible steps for the creation of an AWARD alumni program. We'll see. There was a reason why i included them and that was because without them the system lacks an environment and lacks structural coupling and so is kind of floating in theory space.
Anyway, back to case studies, p449: "the analyst's first and foremost responsibility consists of doing justice to each individual case. All else depends on that." See steps for writing one on p450. "the credibility of the overall findings will depend on the quality of the individual case studies"