Dr Judy Rose
This workshop provides an introduction to thematic coding and analysis of data from semi-structured interviews. Firstly, we define thematic analysis (TA) and distinguish it from other analytic methods commonly applied to textual data, such as content analysis and text mining. Themes can be identified as "focalised points' of meaning, when interviewees address the "heart of the matter' being discussed. Themes refer to relational ideas, rather than keywords, and consist of mostly latent (deep, unobvious) rather than manifest (surface, obvious) concepts. Themes may be coded via a series of steps which involve 1) being immersed in the data, 2) developing an initial "open' coding of themes, 3) revisiting coding to combine themes, then coding text to multiple themes, before 4) refining coding by checking theme alignment with theory. In practice, this process involved 1) labelling deductively-derived themes from prior review of relevant literature and from interview questions, and then 2) identifying inductively-derived themes to capture unexpected or new information emerging in an interview study conducted on big data linkage (Rose et al., in press).
The strategy we employed to code and analyse interview text involved deductive, inductive and abductive processes to organise and categorise the data into countable units; important for conducting thematic analysis in a mixed method framework. The countable units comprised of larger "discursive chunks' of spoken text that were coded at the scale of paragraphs, rather than words, phrases or sentences. Counting the occurrence of thematic codes (containing discursive chunks of text from each participant about an aspect of data linkage) found a slightly higher number of inductively-derived themes (39), compared to deductively-derived themes (27). This was not surprising given the relatively new area of big data linkage. Theme coverage was quantified to indicate relative "airtime' given to topics by all participant and then compared across themes. The thematic coding and analysis demonstrated in this workshop prepared the data for a mixed analysis approach, which is presented by A/Prof. Sama Low-Choy in the "Mixing Clustering in with Thematic Analysis of Interviews' workshop.
Format: This workshop will be delivered online during a 2.5-hour period, via Collaborate, with active learning activities, and Q & A.
Pre-requisites: This builds on the foundations provided by the six Mixed Methods Foundations workshops (MM1 - MM6). It is a companion to other RED workshops and seminars on specific mixed methods: "Mixing Clustering in with Thematic Analysis of Interviews', "Interpretive Phenomenological Analysis of Interview Data in QUAL-dominant MM', "Integrating Hermeneutic Phenomenology in QUAL-Dominant MM', and "Interviewing to Quantify Expert Knowledge with Uncertainty'.
Pre-reading activity, before attending: Onwuegbuzie, A.J. & Hitchcock, J. H. (2015). "Advanced Mixed Analysis Approaches", Chap. 16, In Hesse-Biber, S. N., & Johnson, R. B. (Eds.). (2015). The Oxford handbook of multimethod and mixed methods research inquiry. Oxford University Press.
Reference: Rose, J., Low-Choy, S., Katz, I., Homel, R. (in press) Enriching thematic analysis with clustering techniques: Applying mixed analysis to interviews about big data linkage. In Cameron, R., & Golenko, X. (Eds.). Handbook of Mixed Methods Research in Business and Management, Edward Elgar.