Dabbling with Discourse Analysis

Introduction and Revisiting Memoing

Discourse Analysis (DA) is a qualitative research principle and practice with strong links to Constructivist Grounded Theory, and I'm going to have a go at blending some of what it it and what it does with my CGT memoing and coding, but before I get lost in the weeds of Discourse Analysis, it’s worth backtracking a little and talking some more about memo writing as the first of those links. 

Writing memos is an activity that is essential to Grounded Theory and is  the fundamental process of researcher/data engagement that results in a ‘grounded’ theory. It’s also the methodological link and the distillation process through which I hope to transform data into theory. I've been scribbling memos throughout all my interviews, and basing their content around things that instinctively make me go 'oooh, that's interesting!'  

Writing memos means I can make a (tentative?) start on interpreting data. Through sorting, analysing, and coding the ‘raw’ data in my memos, I hope to discover emergent social patterns, and by writing memos continuously throughout the research process, will explore, explicate, and theorise these emergent patterns. “It is the methodological practice of memo writing that roots the researcher in the analyses of the data while simultaneously increasing the level of abstraction of his/her analytical ideas.” (Charmaz 2006).  

These early memos are speculative and can lack coherence and connection to one another, but that’s okay. They record interpretations and incipient patterns emerging from the concrete realities of the social worlds of research sites, reflect the social lives that they interpret and the interactional social positions of me as researcher and my respondents. They are messy and incomplete, and filled with undigested theories and nascent opinions. Some of my ideas are represented in fragmented phrases, weird diagrams, half sentences, and frankly illegible scribbles. That said, whatever works is just fine in a memo, as it is merely the account of me talking to myself. Clarity and integration will (hopefully) come with the expanding analysis. 

What Memos Do: 

• Provide a means for me to engage in and record intellectual conversations with myself about the data 

• Clarify processes by explaining and defining properties and characteristics 

• Allow me to gain the analytical distance that enables movement away from description and into conceptualization 

• Record research and analytical progress, as well as thoughts and feelings, about data and directions for further collection and/or analysis 

• Distinguish between major and minor codes and categories 

• Maintain a ‘storehouse of analytical ideas’ (Strauss & Corbin 1998: 220) available for sorting, ordering, re-ordering, and retrieval 

• Do what people in research situation probably cannot do, that is, identify patterns and their properties for both general and specific situations (Strauss & Corbin 1998) 

• Facilitate the generation of theory (this is the bit I'm scared won't happen)

• They also contribute to the generation of data, and generating data is what grounded theory is all about

Memos serve as the fundamental link between data and emergent theory, and that’s well-worth putting in bold type. They record an idiosyncratic and creative process of theory development, and as well as being central to Grounded Theory research, memo writing is a tool used in Discourse Analysis. And DA can complement my CGT research by strengthening and contextualising emerging theories.  

What's Discourse Analysis, then?

Discourse Analysis studies written and spoken language in relation to its social context. It aims to understand how language is used in real life situations and examines the linguistic content and the way language is used in each context to convey meaning in different social settings. Instead of simply focusing on the literal meaning of words and sentences, DA explores the deeper meaning behind how language is used in specific social contexts. It analyses the intended meaning behind the words and phrases within a specific context, rather than just their definition, frequency, or surface-level themes. So, it should help me to understand underlying values, assumptions, and intentions embedded in language. DA will also give me a lens through which I can view the many functions of language, including shaping societal narratives and opinions. Importantly, as far as my study goes, I can use DA to look at how language curates and maintains my participants’ social identities, shapes their social and cultural values, constructs and maintains their social realities, and how language can be used by participants to persuade, inform, and entertain. But most importantly, I want to know how the language my participants use helps them to describe their experiences of playing OWRPGs.

The key concepts of DA are context, statements, structure, and - vitally in my research - the shaping of reality. I’ll be using (playing with) a ‘language in use’ model, focusing on technical aspects of language such as how grammar, syntax, and phonology are used to create meaning. Key to this approach is the recognition that language does not simply reflect a pre-existing social reality but actively constructs it.  

Key concepts related to this method include the context of the situation (the specific circumstances surrounding language use, including the relationships among participants, communicative purposes and the mode of communication, be it spoken or written), register (the variety of language used by specific people in specific contexts and situations; for example while a non-gamer would refer to a game as being of ‘poor quality’ or ‘unreliable’, a gamer would refer to the game as being ‘janky’), genre (be it spoken interviews or extant written reviews of gameplay experiences). Intertextuality looks at the way in which one text incorporates, references, or alludes to other texts, be it directly or indirectly and highlights the interconnectedness of texts, and how meaning is constructed through these connections. That may also come in handy for my literature review(s).

A Dual Coding Approach

The second link between both methodologies and methods is coding. I’m going to take a ‘dual coding’ approach at the initial analysis stage, though at the moment I'm not sure how I'll be going about this from a logistical standpoint (note to self: transcripts formatted as a column of text in the centre of pages set to landscape orientation, with CGT coding carried out in the space to the left of the transcript text, and DA coding on the right? That could work?)

When it comes to thesis-writing time, I need to write a rationale for this, and critique it. I'll need to write about why I’ve chosen this approach, and why I did not choose my original approach: unlike grounded theory, DA is usually preceded by a hypothesis. Feeling tied to this, I initially wanted to use DA to prove my hypothesis that there is a meta unintended consequence linked to providing a safe space to talk about gaming. However,  this would have turned the research into a mixed methods study which runs the risk of 'muddying' the theoretical-discovery waters. And while Glaser would say ‘You can’t do GT this way! It’s not pure enough!’ Charmaz says ‘Do what you want but justify it.’  I’ll be dual coding interview transcripts and focusing on the structure and patterns of spoken interaction. Like CGT, I’ll be using interview transcripts to examine how conversations markers such as pauses and specific phrases contribute to conversational flow and meaning. Coding ‘Call for Participants’ survey responses may also be a good idea, as these are rich narratives in themselves. Again, memoing and dual coding will work in parallel to strengthen any theories that are emerging. Or at least, that's the plan and I hope it works.

And Finally...

...I'll finish with an admittance. This post has been an info dump of 'stuff' about CGT and its links to DA, and it's full of acronyms, and it's probably not very coherent or cohesive, and for that I apologise. However, writing this has given me a place to come back to and make more sense of at some point, and the inclusion of another methodology, even if only to a small extent, is a big enough process to warrant a post. And talking of trying to make sense of where all these things crossover, and how they relate to each other (and by 'things' I means grounded theory, discourse analysis and phenomenology, which I've written about previously here), I felt the need to try and visualise this crossover, so I came up with an as yet unnamed model thingy: