The Conceptualization of Data

Judith Holton in an article titled "The Coding Process and Its Challenges" within the Sage Handbook of Grounded theory sums up grounded theory and the coding process: "The conceptualization of data is the foundation of grounded theory development. The essential relationship between data and theory is the conceptual code" (266). It sounds simple, right? Who knew that in that little word "conceptualization" lay all the misery and mystery of research, as well as the source of so much disagreement.

This blog post won't examine many of Holton's interesting propositions about grounded theory as its own research paradigm or her thoughts about pre-conceptualization. Instead, I want to write for a few moments on the subject of description and conceptualization.  Holton states: "To understand the nature of classic grounded theory, one must understand the distinction between conceptualization and description. Grounded theory is not about the accuracy of descriptive units, nor is it an act of interpreting meaning as ascribed by the participants in a study; rather, it is an act of conceptual abstraction" (272). Holton attempts in her article to distinguish GT from qualitative research methodologies, and you can see her do it in this quote surrounding the issue of description and interpretation. While I think her differentiation of GT from qualitative research is interesting, I am more interested in her discussion about the movement beyond description toward conceptualization.

I have struggled with this issue of moving beyond description in my own research. Corbin and Strauss stress that naming and labeling is a first step of open coding, yet it seems more mysterious as to how to arrive at "codes" that are conceptual in nature and not merely descriptive. Holton warns that those of us trained in qualitative research may latch onto what she calls "descriptive coding" "with its capacity to portray rich detail, multiple perspectives, and the voices of lived experience" (272). I have certainly felt this descriptive pull and done my fair share of it. At times I feel as if I'm lingering on the "ground" level of description waiting for the wings of conceptualization to pop up on my back and enable me to fly. Perhaps this flight will still happen.

Reading Holton, however, provided me with one anchor for this conceptualization process that I think I might find helpful. She is admittedly a Glaserian "classic" grounded theorists, and I have spent more time studying and using Strauss and Corbin as my guide (though I consider Dey a neutral mediator), so I hope that I don't "mix my method" by incorporating some of Glaserian practice. But Holton brings up Glaser's notion of the "concept-indicator model." I was not familiar with it (or had forgotten it), but it make sense. Holton gives an example of a number of in vivo, descriptive codes. The researcher then looks at these codes and asks, "What concept might these indicate?" What concept might these descriptive items/features indicate or be described by in a more abstract sense?

I am seeking to delineate the categories, properties, and dimensions of my data, so which comes first? The property or the category? Are properties the "indicators" of a concept (and a concept is the heart of a category)? Or do you get your category first and them map out its properties and dimensions? It all seems  messy because I don't think this conceptualization process happens sequentially, nor can description ever be free of elements of conceptualization. Holton seems to acknowledge the messiness of this activity, and affirms that it is GT's methods that are most important to trust and follow through this process:

"The chaos is in tolerating the uncertainty and subsequent regression of not knowing in advance and of remaining open to what emerges through the diligent, controlled, often tedious application of the method's synchronous and iterative processes of line-by-line coding, constant comparison for interchangability of indicators, and theoretical sampling for core emergence and theoretical saturation" (273).

Chaos and uncertainty indeed. She calls GT a form of discipline, and she is right. I am getting more comfortable with this messiness and with the procedures of this methodology, and that includes becoming more at ease with the uncertainty. Interestingly, Holton brings up one warning that Glaser talks about related to when researchers begin to generate concept (true concepts!) from the data. They talk about an excitement that happens: "Captured by the imagery, or 'grab' (Glaser, 2001: 19-21) of the emerging concepts, they [the researcher(s)] switch their attention from abstraction to description" (273). It sounds like a possible danger is to jump back into the descriptive level to use the concept as an integrative descriptive tool too soon. I must remember not to do this.