S.TRUEMAN PhD THESIS 2016

158

Miles and Huberman (1994) argue that coding is a complex analytical task. A researcher does not just label but also makes linkages. ‘[Coding] leads from the data to the idea, and from the idea to all the data pertaining to that idea’ (Richards & Morse, 2007, p. 137). This coding process is heuristic in nature; ‘an exploratory problem-solving technique without specific formulas to follow’ (Saldana, 2009, p. 8). The exploratory aspect of coding is what generates the cyclical nature of the process. The data was broken apart and deconstructed ‘in analytically relevant ways in order to lead towards further questions about the data’ (Coffey & Atkinson, 1996, p. 31). Achieving this process took a number of coding cycles whereby the researcher applied and reapplied different interpretations and approaches to coding the data to discover and uncover meanings in the data—a process of segregating, grouping, regrouping and relinking in order to consolidate meaning and explanation (Grbich, 2007). A consequence of coding and recoding was the refinement of initial codes through the process of rearranging and reclassifying. This refinement meant that some codes … were subsumed, relabelled, combined and even eliminated (Saldana, 2009). The process of coding ceased when there were enough codes ‘captur[ing] both the patterning and the diversity within the data’ (Braun et al., 2013, p. 211). Practical aspects of the coding process involved working manually and systematically through the entire data set, affording thorough and equal attention to each data item thereby identifying interesting aspects of the data, which related to answering the research question. Coding manually involved a combination of writing notes in the margins of the transcripts being analysed, significant and relevant aspects of the transcripts being highlighted using pre-chosen colours to denote similar meanings or concepts and using