A Grounded Theory Update: Coding and Memos
Initial Stage of Grounded Theory Coding
Coding the data using the initial stage of the Grounded Theory process, known as Open Coding or Initial Coding, has progressed substantially since the previous update. In fact, I’ve actually completed the task of coding through the first set of data during the previous week, which I had not expected but has put me ahead of schedule!
Just as a brief reminder, Open Coding or Initial Coding refers to identifying concepts within the data and the use of codes that summaries or describes the meaning or characteristics of that particular data segment, and therefore identifies these concepts. You could say that coding gives data segments an identity that you can refer to time and time again as you progress through your coding, depending on the characteristics that you identify and interpret within each data segment. You are essentially making practical, empirical observations of the data, and interpreting that data to mean something that is of value or in some way contributes towards characterising the phenomenon of interest that you are exploring. Whilst all your codes and code-data segment matching is an interpretive process, it is also objective as all codes are grounded in the data, particularly with the process of comparisons between data segments for similarities of characteristics. I shall be talking more about this in my future short blog series of Grounded Theory from next week.
At the time of writing this blog post, I think I have about twenty or more different codes that I have used across the whole data set, and this is actually a reduction on the amount produced during previous coding sessions. What I am increasingly discovering within the grounded theory approach is the influential impact and role of context on my interpretations, and perhaps the way that I should be interpreting and coding the data, and identifying the appropriateness of code-data segment matching. What is assigned a particular type of code in one context would be coded as something completely different in another context. This appears to be the nature of exploring learning processes and phenomena using grounded theory: the understanding and acquirement of knowledge regarding the development and process of learning differs between contexts. With collaborative learning for example, the collaborative activities, processes and communication shifts and moulds what is happening within the data as time progresses, and can illuminate different patterns at different times depending on the context; depending on what is being dealt with at the time. It is simply not a case of observing a particular process and thinking that it’s always universally understood because learning processes and phenomena have a nuanced existence that is shaped and moulded by events, happenings, actions and others within collaborative situations.
Therefore, as a grounded theory researcher, when you are exploring learning phenomena, the context that envelopes or provides the basis for the learning process is able to mould and shape this learning process over time, yet grounded theory enables you to identify the nuanced existence and subtle differences between the characteristics of similar concepts. Beyond reading the textbooks on grounded theory, the biggest learning curve and learning experience of my application of grounded theory has been trying to understand the importance of context and the way in which this really impacts my interpretations and observations of the data. I’m still learning now. I’m still wondering and questioning if I have really coded everything correctly even though I have checked through things several times during the past week and have altered the coding where I feel necessary.
Along with coding, I’ve also been writing plenty of memos. Memos is a technique of grounded theory that helps you to build your theory by capturing all of your thoughts about the development of your codes, what you have observed, the similarities and differences that you find between coded segments, and the comparisons between different coded segments e.g., their characteristics and contrasts between similar and different concepts and what makes those data segments really what they are.
Additionally, all this information contributes valuable insights and input into your theoretical sensitivity and theoretical awareness of the data, as well as developing theoretical sampling. Theoretical sampling is a qualitative sampling method that determines what to sample next (e.g., what information or data you need next) based on the emerging theory: the observations and questions derived from the data and the codes all guiding and directing the next set of data to pick up and analyse. I shall be talking about this more either during the short blog series of grounded theory or at some point early next year.
Focussing on rewriting the memos shall be the focus of the rest of the week in an attempt to communicate my ideas more clearly, to tidy them up a bit, and to reduce their number and organise them into something that makes a bit more sense. The set of memo writing sessions just completed involved writing a memo page (in some cases several pages) per code, within which each data segment coded with that respective code was explained and compared to previous segments in order to identify and locate subtle differences between each segment, leading in some cases to identification of potential categories (which are basically a combination of various codes and provides the core of the theory) and categorical properties and dimensions.
Writing a memo per code worked fine for a while, and the potential categories identified so far are suitable although these obviously need to be re-examined continuously (shall talk more about categories next month) but what I have realised is I have been taking these data segments out of their context and trying to explain them as standalone entities. As I went deeper into the data I began to realise that data segments can be logically connected, therefore trying to explain them independently of each other was becoming an increasingly difficult task. I found myself referring to these logically connected data segments in order to provide a contextual explanation for the data segments and their difference between other similarly coded data segments.
What I shall do next is rewrite the memos and add more details about the context. Instead of writing about each data segment as stand alone entities, I shall now write about complete units of logically connected data segments. This way, I can break the unit down into constituent segments and attempt to explain them individually and then discuss their relationship to each other as part of that unit. Doing it this way, I think I can then explain the meaning of individual segments without losing its contextual meaning and relationship with other segments. And, I can compare data segment to data segment, and data unit (a series of logically connected data segments) to data unit. It makes sense, um, well, currently in theory……..
What’s The Aim Then?
At the conclusion of the week I aim to have a complete coded first set of data (shall be rechecking again), a full set of rewritten memos and an updated theoretical framework. This will then, as far as I am currently aware of, bring grounded theory work to a conclusion for the year. I shall send everything off to the supervisor for feedback and guidance, and up to the Christmas holiday I shall work on the first literature review chapter, and write the blog series on Grounded Theory!
Plenty to come; watch this space (or just read the blog!)