In my two recently published books, and blogs, I discuss the development of genuine experiential self-understanding as the basis of healing psychological pain and developing greater levels of psychological well-being, as well as developing psychologically healthy personal relationships and compassionately transforming society. In my books and blogs, I maintain that psychological self-help and/or psychotherapy is most effective when we engage in direct/unmediated empathic communion with our own feelings or with another individual, not mediated and distorted by any preconceived theoretical or interpretive model.

I realize I am late to the game with this response but hopefully it will still be of some use. The first thing to realize, and I say this because it can as a shock to me when I began my essays, is that there are numerous methods called “phenomenology” and while, in theory, they all rest upon the same philosophical principles they are all very different in how they process data and the goal of the analysis. I personally am using Amedeo Giorgi’s psychological phenomenological approach (Book: The Descriptive Phenomenological Method in Psychology: A Modified Husserlian Approach). In this approach one is extracting (not interpreting) the psychological meaning held within the lived experience of the participant. The procedure involves reading the single interview text as a whole, then breaking it into “meaning units”, and then transforming those meaning units in order to distill the psychological meaning. Once this is done then all of the transformed interviews are taken as a whole set of data and analyzed for key constituents (which for practical purposes could be thought of as themes though they are somewhat different). I personally feel that transforming the interviews as they are taken, before all of them are complete, might contaminate the interview process. Then, as this method is structured, each interview is examined first by itself and then as part of the larger group.

Another method that clusters themes somewhat as you describe is the Clark Moustakas method (Book: Phenomenological Research Methods). Still I think that the analysis is best reserved until after the data collection phase and even though Moustakas’ method uses theme clusters I believe that each interview is broken into themes within itself before being grouped with other interviews.    

These are only the two methods I am most familiar with, as I said there are many. It is important to know that each of these methods is self-contained and cannot be combined with one or more of the others so pick one and stick with it as my adviser puts it.  

Good luck.