Coding Data: At What Point Should I Know What the Heck I'm Doing?

(this is a bit of me rambling...just trying to articulate for myself why this article has impacted me)

An article I read about 2 weeks ago has now become hugely influential to my project: Godley, Carpenter, and Werner's (2007) " 'I'll speak proper slang': Language Ideologies in a Daily Editing Activity."

Initially reading this article, it was the conceptual framework that was most loudly "whispering my ear." I had been struggling for a few weeks to figure out exactly how I wanted to organize and relate the central ideas of my study--language attitudes, code-switching, AAE, SE, identity. What immediately struck me was that I was that I had a very similar aim to Godley et. al: we both wanted to analyze perceptions and beliefs about standard and non-standard dialects. Godley et. al articulate this as a question of "language ideologies," and as I read, I decided that I did too. Language ideologies, as they define it, is "the assumptions about language that are collectively held by a particular group of people within a particular sociocultural and historical context" (p. 104). They go on to say later how our social identities are inherently embedded in these language ideologies as they are a part of our "particular sociocultural and historical context." I'm now in the process of figuring out how to tinker with this idea of "language ideology" as a means of tying together those ideas from my study I listed above.

Equally important to me was how they were able to codify these seemingly intangible features of "language ideology." These researchers had 3 coding phases. Initial coding was open coding, looking for emerging themes and patterns. Through this initial coding, the researchers arrived at the focus for analysis--language ideology. In the second coding, the researchers decided to code for content of ideology (what opinions were being expressed) and source of ideology (who/what was expressing them). In the third and final coding, the researchers wanted to look at the relationship between the source and the content of the ideologies and used discourse analysis codes to look at this relationship.

A few things I found useful about this:
1) the researchers really didn't know what they were looking at or for until the end of the first round of data analysis. Reading this came as a huge relief to me. As I'm beginning to collect data, I start to feel that I'm in a bit of a fog of different thoughts, hunches, concepts, and questions. It can feel like I'm a bit lost, and in reading this article, I started to feel like that might be beneficial to me.
2) I really liked the way they divided up the 2nd and 3rd phases in coding b/t what they were looking for and then the relationship between these. I think too often I jump ahead to that "relationship between" different concepts. It was a nice model to see how certain topics can be looked at discretely.
3) It's so clear in reading their coding process how these codes are tied to their definition of language ideology--what is believed (content) by a particular group (source). I admire the clarity of that...and maybe I can get there down the road.

thanks for indulging the rambling a bit :)