S.TRUEMAN PhD THESIS 2016

159

‘post-it’ notes for ease of reference to identify segments of data. Initially having manually identified codes the researcher, using an electronic copy of the transcripts, then matched them with data extracts thereby highlighting the code. All highlighted data extracts were then collated together within each code. 5.9.3 Identifying patterns and forming themes On completion of coding the researcher commenced identifying patterns (as opposed to pattern-matching) across the data and categorising the codes. This process was iterative; moving back and forth across the data and codes, reading and re-reading the data. While identifying patterns and making themes occurred simultaneously, this dissertation will describe the process separately. Reoccurring codes were identified which were noted for their regularity and any emerging patterns and relationships of connection (Agar, 1996; Saldana, 2003). Patterns are not just stable regularities but can be in varying forms (Hatch, 2002). Saldana (2009) outlines six possible ‘characteristics of patterns which guided the researcher in this study;

• similarity (things happen the same way) • difference (they happen in predictably different ways)

• frequency (they happen often or seldom)

• sequence (they happen in a certain order) • correspondence (they happen in relation to other activities or events) • causation (one appears to cause another)’ (p. 6) Identifying patterns across the data meant not just identifying the most frequent. While frequency is important, the researcher also interrogated and interpreted the identified patterns ‘to capture the different elements that are most meaningful for answering the