Structuring Your PhD - the three pillars of a thesis

There are two trinities in your PhD ... one is the three ologies (see "being clear about methodology" in a separate post) ... the other is the relationship between the literature, the method and your empirical material.

Imagine three separate documents set out on a table, labelled as follows ...

A. My Literature Review
B. My Method Statement
C. My Empirical Work

The vast majority of PhDs will have chapters on these three themes, but for the moment, imagine them set out next to each other going from left to right ... A, B, C.

The ideal is that they each (i) make sense in and of themselves and (ii) link smoothly to the others. In summary each needs to do the following.

Document A. Your literature review needs to set out the broad territory that your PhD will sit within. This often entails setting out different schools of thought within the one broad area e.g. the econometric and the social views of strategy that can be seen in the research literature and that delineates between the research of people such as Michael Porter in the former case and Henry Mintzberg in the latter. Gradually, having set out your broad area and reviewed its genesis and development, your review needs to work through to the specifics of your interest. As you get deeper into the literature, you move from schools of thought to specific prior work or studies. Hopefully allowing you to get to a clear and articulate research question that you have demonstrated is not answered in the existing literature.

Document B. Your method statement follows a similar pattern to document A in that it typically moves from the general to the specific. First, you are looking to establish that you are aware of the range of choices facing any researcher and that you can set these out. This does not mean setting out a table that says quantitative vs. qualitative but rather, it implies being much more sophisticated than that. See the separate post on methodology again. Having set out your ontological and epistemological position, you need to move to the specifics of your research design. That means that you should set out the mechanics of how you'll execute the research. For example, you might be looking at blockages to growth within small, family owned businesses and you might be looking to find some firms that have been stuck at one level of turnover for a number of years, another that seem to have grown at a steady pace over the same timeframe and third group that were stuck for a while but somehow broke out and began to grow again. Within these three sample groups you might then want to look into owner-manager motivations, etc. Be specific. Say what you're going to look for, how you'll gather the data, what you'll do once you've got the data and how you'll know if you've answered the question you've set yourself. Of course, now Document A (which concludes with a research question) needs to articulate seamlessly with Document B which sets out a method for answering that question.

Document C is your empirical material. It needs to set out the original or secondary data that you've gathered. It then needs to say what you've done to the data perhaps in terms of coding, categorising, comparing or counting things within the data. Finally, having made all of that clear, it needs to set out your analysis of the data. Now, Document C needs to be the actual execution of the proposed research design set out in Document B.

Most texts on how to do doctoral research suggest simply that you need to do A, then B, then C in a nice, neat, logical and linear process. The lived experience of conducting research is however, a little more messy. Consider what happens when the data you gather doesn't actually answer the question you've set. Or the method you've decided upon turns out not to be workable. Most doctoral researchers experience something more like the need to reverse engineer the connections between documents A, B and C. You might find yourself thinking "given that this is the answer that I can get from my data (in document C), what might a good question look like (in document A) and how would I best describe the method that I use to get there (in document B)." Though this may seem a little odd, or perhaps even disingenuous, it is the lived experience of most researchers that unexpected things happen along the way. Your PhD will take you at least three years and if there are no hiccups along the way in methodological terms, then perhaps someone will publish something very similar to your own work just as you enter the final stages of your PhD. Hence, you find yourself with the need to go back and establish a slightly different contribution than the one you'd be thinking of.

The key thing is that the readers of your eventual thesis, and in particular your examiners, can see some cohesion between your take on the literature, your views on method and the empirical work you've undertaken. This structural integrity is the thing examiners home in on. Perhaps because PhDs are written over years not hours or days, it is not unusual to spot big inconsistencies. Take a single sheet of A4 paper and try to capture in a few sentences the key messages from your equivalents of documents A,B and C. Be clear with yourself the links that you need to make and always, always work backwards from the end of your PhD study to double check that the three pieces stick together neatly. In the early years of your studies the pieces can move about dramatically as you tweak or abandon early plans. In the latter years, the pieces begin to settle into particular locations and take root there. You still need to revisit the relationships between them however, to help signal to the examiners that you've thought through how your PhD "hangs together".