Data analysis: When it starts to make sense—also known as “breakthrough”

Yes, that title says what I want it to say. I’m hoping that by sharing this in my blog, it provides some hope for doctoral students groping through the dark depths of their data.

Admittedly, I have been procrastinating somewhat—a few months, actually. I have done the initial coding of my transcripts using discourse analysis techniques derived from the work of Gee (2010), Potter (1996), and some other significant authors. Then, I did some additional coding for some salient and potential phenomenographic categories that I saw emerging—some expected; some not so expected. I exported from Atlas-TI quotes according to the code categories. I have read through all of them in Word, highlighting and jotting down additional notes in on the pages themselves and in my extremely messy journal. I have plodded along in faith that something would one day make sense, and with an unspoken knowing that these steps would lead me to that place.

As a result of this seemingly blind journey, I started seeing categories of description emerge. And, I am getting a sense for the variation in experience associated with the categories of description.

During my last meeting with my supervisor in which she shared some newly discovered ideas from another student’s viva, I crossed my most profoundly deep threshold of understanding. I am now coding for some additional code-perspectives. And, I can see how discourse analysis strongly supports the study of liminality–sub-liminality, in particular. My supervisor and I discussed the idea of discourses and sub-liminality (hyphenated spelling is intended), and I found the relevant literature by the threshold-concept-gurus in my notes. (Thank gawd for the notes I took on my resources—invaluable.) Today, it feels as if it is just coming together.

At the time of writing this, I was in one of my favourite, sunny coffee-spots, and I was starting to code with these new codes. Suddenly, I saw significant meaning and patterns. I could see the connection between discourse analysis, threshold concepts, Harré’s Vygotsky cycle, and my data emerging as if it is the most natural thing in the world. Whoa. Breakthrough. Now, I must find more time to code.

Of course, my experience in writing my master’s thesis had taught me to expect these moments of lucidity to be punctuated with moments of feeling overwhelming out-of-control. Alas, these are natural ripples in the pond. Keep going.

 References:

Edwards, D. (1997). Discourse and Cognition (p. 368). London, UK: Sage Publications, Inc.

Gee, J. P. (2011). An Introduction to Discourse Analysis: Theory and Method (3rd ed., p. 224). New York, NY: Routledge.

Potter, J. (1996). Representing reality (p. 265). London, UK: Sage Publications Ltd.

Wooffitt, R. (1992). Telling tales of the unexpected: The organization of factual discourse (p. 217). Hertfordshire, UK: Harverster Wheatsheaf.

No Comments on Data analysis: When it starts to make sense—also known as “breakthrough”

42 Questions in Discourse Analysis

• Gee, J. P. (2011). An Introduction to Discourse Analysis: Theory and Method (3rd ed., p. 224). New York, NY: Routledge.

Gee lists 42 questions that a researcher can ask when conducting analysis on a given text or set of texts (p. 121). I will put them into a simple table.

In the table, each row represents what he calls a tool of inquiry. The columns represent the building tasks. To understand how to ask the questions, first one must understand the terminology:

Tools of inquiry

Gee refers to tools of inquiry as thinking devices.

Situated meanings – may also be referred to as utterance-token meanings (p. 63). Form = “morphemes, words, phrases, or other syntactic structures” (p. 64). Function = what the utterance is intended to say or cause. If form and function are not in balance, then we might question what is happening. “Situated meanings arise because particular language forms take on specific or situated meanings in specific different contexts of use” (p. 65). Gee notes that analysis is complex because context is always changing. But, we can view an utterance from the viewpoint of different contexts and potentially gain insights into the meaning of the interaction (p. 68).

Social languages – “different styles of varieties of language for different purposes” or different social situations (p. 28). Social languages can have their own “distinctive grammars” (p. 50). They help people recognize and create their social contexts.

Figured worlds – typical, often taken-for-granted, stories or simplified pictures of situations with “typical participants, activities, forms of language, and objects and environments” (p. 71). Holland (1998) defines them as “socially constructed realm[s] of interpretation” (cited in Gee, 2001, p. 71). They are not static. Figured worlds “mediate between the ‘micro’ (small) level of interaction and the ‘macro’ (large) level of institutions” (p. 76). People can use figured worlds to construct simulations of situations to help them understand or act in the given situation(s). People also use figured worlds to evaluate appropriateness of social activity in the world (p. 90). They can be nested, incomplete, inconsistent, and changing. (An examination of figured worlds can help to uncover taken-for-granted assumptions that guide social behaviours—review previous blog posting on Ian Hacking’s book, The social construction of what?”)

Intertextuality – the act of referring to or quoting texts outside of one’s immediate discourse. It is a “sort of cross-reference to another text or type of text” (p. 29). Words may be borrowed or switched from one social language and used in another (p. 58).

Discourses – linguistic and non-linguistic elements that combine into “characteristic ways of thinking, acting, interacting, valuing, feeling, and believing” (p. 28). They combine to produce a “socially recognizable identity” (p. 29). [Note: big “D” discourse.] On page35, Gee emphasizes the importance of recognizability. Discourses do not necessarily have clear boundaries; they can overlap, split, meld, die, mutate, etc. (p. 38).

Conversations – Gee uses this word with a big “C” to refer to “all the talk and writing that has gone on in a specific social group or in society at large around a major theme, debate, or motif” (p. 29). He adds later, “They are the products of historical disputes between and among different Discourses” (p. 56).

Building Tasks

For each building task, Gee suggests that a researcher can ask the following questions:

Given what the speaker has said or the writer has written, and how it has been said or written, what

• [things, people, practices, identity(s), relationships, connections, disconnections, sign systems, languages, social languages, ways of knowing]

in this context are relevant and significantand in what ways are they significant? How is the speaker or writer trying to

• [give significance to things, enact things, depict things, recruit things, use things, connect things, disconnect things, privilege things, disprivilege things]?

Note: From my perspective, the building tasks of identity, relationships, and connections can be very difficult to separate. In particular, I see identity as highly integrated with relationships.

Questions from combining the tools of inquiry with the building task

Here is how Gee phrases the questions from the table of 42:

Significance: “How are situated meanings, social languages, figured worlds, intertextuality, Discourses, and Conversations being used to build relevance or significance for things and people?” (p. 121).

He asks the same question for each building task.

So, my question is what elements of this do I take with me when I do my phenomenographic study? It is definitely helpful in expanding the way I will read the interview transcripts. Gee’s book is also helpful for me in developing my transcription protocols. Now onto Wetherell et. al.

5 Comments on 42 Questions in Discourse Analysis

Discourse analysis & a constructionist approach to phenomenography

Gee, J. P. (2011). An Introduction to discourse analysis: Theory and method (3rd ed., p. 224). New York, NY: Routledge. http://www.amazon.ca/gp/reader/0415585708

I am exploring the use of discourse analysis (DA) alongside the constructionistt phenomenographic methodology that I will be using on my PhD thesis. So, I have started with what has turned out to be a nicely written introductory text by James Paul Gee.

As my first love in academia was languages and linguistics, I feel very comfortable with Gee’s text. In fact, it feels much like a homecoming. I now understand why social constructionist philosophy has become, quite unintentionally, so significant in my doctoral work. In this book, Gee’s own philosophies about language are made clear:

In the broadest sense, we make meaning by using language to say things that, in actual contexts of use, amount, as well, to doing things and being things. These things we do and are (identities) then come to exist in the world and they, too, bring about other things in the world. We use language to build things in the world and to engage in the world building. (p. 16)

To an extent, I try to contain my excitement upon reading this until I can fully ascertain that Gee does not cross the line into critical realism—the idea that there is a reality out there, but we can only see it from our own limited perspective. (I know that this is an oversimplification.) The danger of this crossing could have been perpetrated in Gee’s treatment of the other “stuff” related to language such as non-vocalized behavours, appearances, enactments, social institutions, and props—the context of language (p. 35).  However, Gee appears firmly rooted in the social constructionist camp. Consider, for example, how he refers to knowledge:

. . . the physics that the experimental physicists “know” is, in large part, not in their heads. Rather, it is spread out (distributed), inscribed in (and often trapped in) scientific apparatus, symbolic systems, books, papers, and  journal, institutions, habits of bodies, routines of practice, and other people. Each domain of practice, each scientific Discourse . . . attunes actions, expressions, objects, and people (the scientists themselves) so that they become “workable” in relation to each other and in relation to tools, technologies, symbols, texts,  and the objects they study in the world. They are in sync. (p. 36)

In this example, a given discourse exists and creates and recreates itself including the scientists who engage in the discourse. This recognition of the cyclical relationship between language and perceptions of reality is, in my opinion, a cornerstone of social constructionism.  Later on the same page, Gee uses the word instantiation with reference to the power of language to create a constantly shifting and abstract world. This, too, dramatically buoys a social constructionist position. (Interestingly, Gee suggests reading Bourdieu, Clark, Engeström, Foucault, Hacking, Hutchins, Latour, Lave & Wenger, and Wittgenstein—all of whom have constructionist leanings or whose work has been foundational to constructionism.)

Before I outline how I will use DA to support my phenomenographic research, I will first post my notes on the main concepts of Gee’s book. But, that is for tomorrow.

1 Comment on Discourse analysis & a constructionist approach to phenomenography