Reflexivity in analysis

The last time I blogged, I was expressing some reluctance at the prospect of engaging in the next stage of my PhD analysis. If I was honest (actually, I think you will find I was) then I'd be saying the work ahead of me didn't seem as interesting as the previous phase: the last set of conversations with participants, where we talked about cartoons and 'maps' I had created based on the first phase of analysis of their talk about professional identity and social interactions.   Since then, I've got on with that next block of work and I've also pulled together some thinking about what's getting me 'stuck'. Before I share some of those insights, I'd like to say that I'd never 'got there' without the pig headed determination to just persist. I realise some PhD researchers 'give up' at several points in their journey - I understand why this might happen - but I'm convinced at the value of just putting the work in. Thanks to my wife, without whom I couldn't do this; we have four children and I have a busy job. Right, that said, on to my recent insight into my own research journey.

Put simply, I've realised that what felt frustrating was the sheer...impossibility of the task ahead of me. I don't mean the analysis, I don't even mean the writing of the thesis; but the sort of questions I was asking. I have realised that what 'worked' for me up to this point were not 'researchable' questions, but were philosophical wonderings. Ironically, I spend a lot of my time advising my students about forming research questions...which I think I'm good at! What I had were a sort of general 'place markers' which talked about the sort of focus I had, but talked in general terms about 'relationships' between things. However, I was asking questions that great philosophers and sociologists spent their lives asking!

I'm not bitter. It's all part of the research journey. We all are living with our 'best' articulation of our research questions. In a flexible research design, it's no problem, but it calls for researcher reflexivity - our ability to think about our thinking, to be conscious about how we are positioned in the research process and how we interact with it. I try to do that through blogging, journaling and talking - in person and through my twitter network (as @ianrobsons).

So; I was living with research questions that had done their job in directing and sensitising my inquiry - they had taken me as far as they could. I wasn't about to throw them out, but they needed re-articulating. This had contributed in my sense of 'stuckness'. I only realised this because I had pushed on with the analysis. This is a lesson in the value of persisting (to a point) when you are not 100% sure. As I immersed myself in what I sometimes thought of as boring (let's be honest) raw data, I felt the distance between those very real narratives and the rather abstract 'questions' that were directing my studies...or rather, had become less useful in directing my studies.

I have realised what may seem quite obvious, which is that there is a reciprocal relationship between the questions and the process of analysis. Analysis benefits from the focus and direction good questions gives, but questions are re-shaped in the light of the demands of the process of analysis (if analysis begins to get unfocused or not productive in some way). The key seems to be aware of the 'balance' between the two. If things are 'working' then I'm usually OK, but if I'm stuck then a re-assessment of one of the two is usually needed. Either questions need refining, or the process of analysis isn't informing the question. I'm sure that's a simplification, but let's be honest, this is a blog post not a thesis so short is good.

So, I've written a note to myself that will direct a re-drafting of the research question. I'm thinking of practices, representations, transformations and all sorts of more specific things I can actually ask my data about. I've realised that the job of my thesis is to move forward what 'we' know about the topic not reinvest or define the world. In my case, it's about the narrative construction of professional selfhood for leaders in early childhood services, and I'm a little bit more equipped to examine my data on that now, thank you very much.

Are we there yet? (more narrative data analysis)

Completing a PhD as a part time student involves more than a little self motivation and management of expectations. You become very aware of what motivates you, and what is most likely to slow you down. Following my last post, I have completed (insert massive relief here) all of the transcription of the audio data from my study. Amazingly, I have also written, edited and had accepted the first article in an international journal using some of the data, on a subject which is not the 'central' question of the thesis. I should be really pleased with myself. Perhaps the fact everyone in my house is recovering from chest infections, then flu, then another set of chest infections is taking the edge off things. The article is a big confidence boost, though.

Perhaps it's a good thing I can bask in the glow of the article and post transcription relief: I am facing up  with the last stages of my analysis-interpretation. This has involved going back to 'the plan' established last summer, which I've followed so far. In summary, this has involved:

• Coding all of the narrative data into one of my two main categories: 'self talk' and 'interactional context'.
• Coding within each of these categories for each participant.
• Creating two different 'maps' (previously discussed) - one relating all the themes of the 'self talk' (what I call 'how the story hangs together') and the other identifying 'patterns' of connections between codes in each category. 
• Creating cartoon images to complement this whole process, including a cartoon 'telling' of how I see each participants overall identity narrative. 

Of course, this list doesn't give you all the detail you need to clearly understand this whole process...the point is that I have already begun an awful lot of 'analysis', and have involved participants in that. I have produced some nice images, and had some wonderful conversations...except I have not 'answered' the questions I set out to answer. Yet. Are we not there yet?

Perhaps I am just at another of those points where it is possible to flag, to give up, to feel tired of it all. I understand from friends and colleagues that this is common enough. The last two days have seen me 'going back in' to the data and it has certainly felt the opposite of glamorous. I think I've just come from a very intense phase of creating and talking about the data at quite an abstract level - patterns, connections, reflections. I am now creating a set of memos of all of my data artifacts - transcripts, maps, cartoons, session photographs. This is legwork - legwork that will help me compare and link, annotate and begin to theorise, but the experience of going through the transcripts and summarising them felt...like starting again.

I take encouragement from my understanding of the process. Thankfully, I have been keeping a research journal and writing a draft analysis chapter for the thesis. In fact, I had forgotten I'd written some of that, so it was good to read these when I sat down; before the depression set in. In short, the process I am following is heavily influenced by Paul Ricoeur's approach to hermeneutics - the interpretation of texts. I won't go into details, other than to say it is a process that involves moving from explanation to understanding and back. To get to 'understanding' you need to explain, and explain better. Understanding starts with a structural understanding, of sorts, and then moves through the process of interpretation (reflecting, comparing, immersing and so on). Of course, I'm going to write a lengthy section of my thesis saying a lot more than that, but suffice to say that I'm back with the explanation. Can you hear the sound of me stamping my feet?

I guess I forget the detail at my peril. Like many people, I like the big picture, but those insights into creative transformations between action and talk will come from building up layers of insight - and those need gathering from each of those transcripts, cartoons, maps and session photographs I have in front of me. What I am learning is that I have to ignore the quiet voice that says 'there's nothing amazing here' - because it is from the very 'ordinary' accounts of participants that I will begin to identify some amazing patterns. Eventually.