Some thoughts on Grounded Theory

I want to write a couple of things right now while they are in my mind about Grounded Theory.




Generating Theory rather than Verifing It

I have had a modernist premise about research that it should be about validating theories--research should prove something. We "generate knowledge" by seeing if something is "so" and we do that by testing it. In addition, I think I have had a sense of "how I thought things were"--my own assumptions and logical deductions about reality (or in this case, about reflection)--and I wanted to see if they were right.

Thus, my largest adjustment has been to step back and shift into theory generation mode. I believe I am making this adjustment ok, but I am pinched by an aspect of my proposed research design that seems to be all about verification in this modernist sense.

Here's the flow of my research design (much condensed):
--do grounded theory comparative analysis to generate a theory of rhetorical reflection
--from this analysis, create a code or coding rubric
--get coders (not me) to use this code to analyze data to see if the code holds up
based upon a) what they find i.e. do they see what I see
and b) what their inter-rater reliability is
--somewhere somehow in here also is the idea of using data mining in the TOPIC vault to further "triangulate" my "findings"

My big question is whether this is too much. As I read about grounded theory, they assert that theory generation is enough for a study, and that it has its own kind of verification. If done well as far as the process, it should have a level of fit and workableness that precludes the need for verification.

"theory must fit the situation being researched, and work when put to use" (Glaser and Strauss 3).

They seem to imply that if grounded theory has been done well and fully then these aspects of describing and explaining the phenomenon will be almost self-evident. Hence, testing in the traditional way is almost redundant.

I'm still wrapping my mind around the issue of verification and its place in theory generation, and I am awaiting guidance from Rich on this question, but I did find this passage on verification this morning.

I note looking at this passage that the word "validation" is used rather than verification! Hmm...

This is a passage from Strauss and Corbin's Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory (1998) in a short section at the end of a chapter on Selective Coding (the end point of the grounded theory coding process):

"When we speak of validating, we are not talking about testing in the quantitative sense of the word. This can be left to future studies, if desired. What we mean by 'validating' is the following. The theory emerged from data, but by the time of integration, it represents and abstract rendition of that raw data. Therefore, it is important to determine how well that abstration fits the raw data and also to determine whether anything salient was omitted from the theoretical scheme" (159)

They propose two ways to validate the scheme:
1--a high-level comparative analysis (i.e. look again at the data)
2--member checking the scheme with the subjects

What they don't propose is to having OTHERS do the comparative analysis using the theory. We shall see. I think this second phase of my research design is still in flux and I may not need to define it absolutely ahead of time. What I think they are urging is that it is important to check how well the theory "fits" and how well it "works."

One other thing I found--the fittingness of grounded theory to study reflection and writing. It is good for studying a process, especially a social process.

Ian Dey in his Grounding Grounded Theory (1999) says this about grounded theory:
"One of the distinctive aspects of grounded theory is it firm location in an interactionist methodology. Hence grounded theory is oriented to explicating 'basic social processes' in dynamic terms--or, to put it crudely, how actions have consequences" (63).

Strauss and Corbin (1998) in talking about axial coding talk about the importance of relating structure to process. They say we must relate structure and process: "Because structure or conditions set the stage, that is, create the circumstances in which problems, issues, happenings, or events pertaining to a phenomenon are situated or arise. Process, on the other hand, denotes the action/interaction over timeof persons, organizations, and communities in response to certain problems and issues. ... Process and structure are inextricably linked, and unless one understands the nature of their relationship ..., it is difficult to truly grasp what is going on" (127).

Now that I look at this quotes I am not sure they are the ones I thought of, but I still think they are pointing to the prominence grounded theory puts on process and how that fits with studying reflection within the process of writing.

OK. Enough for this morning.

Within one day, I read in two different books on grounded theory about how it is a particularly good method for analyzing a process