Thematic Analysis - Qualitative Research Design PDF

**Thematic Analysis -- Dr Chelsea Leadley -- 03.12.2024** **Qualitative Research Design** Methodology: a package of theory, method, and other design elements for doing research. Method: a process or tool used as part of research -- commonly to analyse or collect data. Things to think about in qu...

**Thematic Analysis -- Dr Chelsea Leadley -- 03.12.2024** **Qualitative Research Design** Methodology: a package of theory, method, and other design elements for doing research. Method: a process or tool used as part of research -- commonly to analyse or collect data. Things to think about in qualitative analysis: - Have I managed to capture the situation in a realistic manner? - Have I described the context in sufficient detail? - Have I managed to see the world through the eyes of my participants? - Is my approach flexible and open to change? **Reflexivity** Quantitative researcher's aims are for objectivity, but qualitative researchers are sceptical of this. We all have different viewpoints so are we observing objectively? We need to consider our own views and experiences: - Ami I sensitive to particular issues? - How does my experience affect my interpretation? - What assumptions do I hold about my topic? - Would someone else interpret this differently? **Thematic Analysis** **What is TA?** Braun & Clarke (2021) - At a very basic level, TA is a method for developing, analysing, and interpreting patterns across a qualitative dataset, which involves systematic processes of data coding to develop themes." It's more of a method than a methodology: thematic analysis is an umbrella terms for approaches to qualitative analysis that focus on themes. TA is not joined to any one particular theoretical framework. **Uses for TA in Qual Research** It can be used for a variety of research aims, looking for patterns in: - Individual experiences - Views and opinion - Behaviours and practice - The reasons people think, feel, or behave in a certain way - Identifying/exploring rules and social norms - Representations and constructions of social objects in certain concepts Research using TA should have a focus on what people say and the content of the language, not how people say things. **Types of TA** Reflexive TA (Braun & Clarke) -- approach to coding has some similarities to grounded theory: coding is an active and reflexive process. A close-up of a chart Description automatically generated Code reliability TA -- emphasises the measurement of coding accuracy or reliability and calculating inter-rater reliability (underpinned by the realist/positivist approach). Codebook approaches -- uses structured codebooks, but not concerned with reliability of codes. **Theoretical Foundations** **Philosophical**  **Ontology** The nature of reality/the social world, the 'what'. What is real/exists? What is your view of reality? What can you know? Realist position - One objective, stable reality that appears the same to everyone -- we do not influence this. Idealist position - Reality is made up of our ideas: A focus on mental/conscious phenomena Relativist position - Reality is relative to numerous frameworks and context, e.g. culture, history, experiences. It is entirely subjective, and no single reality exists. There are as many realities as people. **Epistemology** The theory of knowledge, the 'how'. How can I study the social world? What do you know and how can you know it? How can you learn about this reality? Positivist - Knowledge exists outside of the person and can be tested and measured: it is objective. The "scientific technique" - lots in common with the natural sciences: record, test, experiment, be objective. Relativist - Knowledge is specific to the social/cultural/historical context. Interpretivists - Humans are unique, people's actions are based on how they interpret the social world around them -- which varies depending on their experiences. Ontology + epistemology = paradigm **Combining the Ologies: Research Paradigms** Positivism -- one reality we can measure objectively. Critical realism -- one reality but we only know about it through interactions. Constructivism -- multiple realities that need to be interpreted to understand meaning. Pragmatism -- whatever works -- me writing all this fr. **Conducting TA** **Stages of TA** 1. Dataset familiarisation Read and re-read your dataset, make brief notes about any analytic ideas/insights you think of (specific and broad): Things of potential interest Ideas to explore further Your responses to data 2. Data coding Work systematically through the dataset, identify segments relevant to the research question, apply analytically meaningful descriptions, aka code labels, detailed, focused, aiming at capturing single meanings: range from semantic to latent meanings. Don't just copy the data. Include specific meaning to avoid potential for contradictory meanings. 3. Initial theme generation Start identifying shared patterns, group codes together that share core ideas/concepts and might be helpful in addressing the research question. You are generating themes, not identifying/finding them. 4. Theme development and review Assess initial fit of themes to your data, check themes make sense in relation to the coded extracts and the dataset, collectively do they represent the most important patterns in the dataset? What is the core focus/idea of the theme -- the central organising concept? 5. Theme refining, defining, and naming Fine-tune your analysis, it's an iterative process and is not linear, each theme should be clearly demarcated, be prepared to let it go. 6. Writing up Results and discussion combined, weave together a narrative using compelling and rich data extracts that convey your findings but don't force the data into what you want to say, stay focused on your research question, only included relevant themes as not all your data will be needed. A diagram of a system Description automatically generated with medium confidence **Myths and Misperceptions** - There is a singular method called Thematic Analysis - Thematic analysis is a generic way of analysing qualitative data - Thematic analysis is a descriptive method - There are no guidelines on how to do thematic analysis **Example of TA** **Shepherd et al. (2019)** Using social media for support and feedback by mental health service users: thematic analysis of a twitter conversation. Aimed to explore two things: - "The manner in which social media users with experience of mental disorder relate to each other and the social space during internet-based interactions." - "The potential role of resources such as Twitter for the provision of feedback on and engagement with mental health service user experience." Used the \#dearmentalhealthprofessionals to collect 515 tweets Majority of data was grouped into [four overarching themes](https://link.springer.com/article/10.1186/s12888-015-0408-y/tables/1): 1. The impact of diagnosis on personal identity and as a facilitator for accessing care 2. Balance of power between professional and service user 3. Therapeutic relationship and developing professional communication 4. Support provision through medication, crisis planning, service provision, and the wider society Twitter is a useful discursive space in which individuals with a MH disorder can share information, develop an understanding, and receive support. Mental health can be openly discussed on the platform from multiple perspectives. It can be a source of feedback to mental health service providers, specifically on attitudes of providers and their communication skills. **Evaluating Thematic Analysis** - / - TA works with primary or secondary data. - / - It's a highly flexible method that can be used with numerous theoretical frameworks. - X -- Some argue that it is too simplistic for experienced qualitative researchers. - 0 -- It is subjective: analysis is reliant on the experience of the researcher. - X -- It is often performed or explained poorly.