Types of Coding Processes
There are many types of coding processes and procedures. Procedures adapted from Strauss and Corbin’s (1990) grounded theory procedures have been adapted to provide a basic approach to coding qualitative narrative data. They identify three main steps in the coding process.
- Open coding
- Axial coding
- Selective coding
Step 1 - Open coding: This involves reading and re-reading the data (transcript, notes) and assigning labels to chunks of text – a phrase, a sentence, a paragraph – some unit of words.
Step 2 - Axial coding: The process of creating categories – relating the codes to one another, identifying groupings for the list of labels or codes generated in step one.
Step 3 - Selective coding: In this step, the categories that were identified in step two are related to each other – connecting the dots – and creating the “story line” for what the data are saying.
Constant Comparative Analysis
Constant comparative analysis: Throughout the process, the researcher attempts to “saturate” the categories identified in the above steps. This involves identifying areas in the data that represent the category. When there is no new information pertaining to that category (you are seeing the same concepts repeated), it is considered “saturated.” Continue to check the data to determine whether all of your codes / labels fit into a category. Do you need to revise your codes? Do you need to edit, add or combine categories? You should keep track of which chunks of data are assigned to which codes and categories.
University of Wisconsin Extension offers a nice concise description of a qualitative analysis process.
Qualitative Data Analysis Software
Qualitative data analysis software, such as NVivo, HyperRESEARCH and Atlas.ti, can be useful when you have quite a lot of narrative data to code, and thus many codes, categories and themes to capture, track and relate. These each have a learning curve; if you have large (many participants) and complex (lengthy, transcribed interviews), engaging an evaluation specialist with qualitative data analysis experience would be advised.