What to look for in open-ended coding of text

Post date: Nov 11, 2010 6:22:56 PM

[Much of this learned from Gery Ryan & Russ Bernard]

Open-ended coding refers to the process of "pawing" through text and highlighting and labeling particular bits. Typically, you know what you are interested in in a general way (i.e., what the research question is) but you don't know ahead of time what your variables or codes should be. This leads to a question: what to code for? You can't code for everything, so what, strategically, is likely to pay off? This page gives a number of things to look for in text that have proven useful in various settings. Of course, it is only a starting point. Ultimately it will be your research question and the data you have collected that will determine what you look for.

Variables of theoretical interest

• have in mind certain areas that are always important issues, or which are predictably interesting given the data

• In general, the following theoretical constructs are of interest in many settings

  • conflict

  • control/power

  • contradictions

  • legitimacy

  • environment

  • agency

  • obstacles to change

  • coordination

  • socially significant personal attributes

Salient words

• frequently mentioned words (not counting "a" "the" etc)

• drawing connections among frequently repeated words: what is salient to the respondent?

Non-mentions

• looking for what's missing. sometimes tells you what is most sensitive.

Clues to how text sources sees the world.

• • for both interviewing, which is real-time analysis, and analyzing texts

  • metaphors, schemas, scripts, binary oppositions

Metaphors

• • metaphors encode a lot of information, including perceived relationships among entities

  • marriage -- seeing marriages as journeys, trials, rocks,

  • organizations -- seeing organizations as families, machines, organisms, information processors, societies, ecosystems, etc.

Chains of binary oppositions

• • republican / democrat

Frames, schemas, scripts

• • see definitions of frames, schemes, scripts

  • crime (frame)

  • restaurant (script)

  • sexual harassment (both frame and script)

  • signaled by: "the"; implicit reference (things unstated but necessary for understanding)

    • John went to a restaurant. He asked the waitress for coq au vin. He paid the check and left

      • • the use of "the" when neither waitresses nor checks have previously been discussed signals the assumption of a shared schema or script. The speaker assumes the listener expects a waitress and a check to be present in a story about a restaurant visit.

    • John wanted to do well on the exam, but his pen ran out of ink and his pencil broke. He tried to find a pencil sharpener, but there wasn't one in the room. Finally he borrowed a pen from another student. By then he was so far behind he had to rush, and the teacher took off points for poor penmanship

      • • to understand this story, the listener needs to know some basic stuff -- otherwise the final sentence is a complete non-sequitur

Semantic relations & connectors

• • cause & effect

  • similarity

  • association

  • part of

  • signaled by connector words like

    • • causes: because, since, as a result

      • conditionals: if, instead

      • taxonomic categories: a, kind of

Domains and categories

• • lists

  • attributes

  • relations

Taxonomies

• • dendrograms of part whole relationships

transitions

• • pauses, changes in tone, recurrent phrases signal thematic shifts. So helps to identify themes

Example

• Jehn and Doucet use a number of these techniques in their papers. For a summary, see this handout