4.6 Data Analysis
When analysing data, qualitative researchers typically use text. The goal of qualitative data analysis is to assemble or reconstruct the data in a meaningful or understandable method that is transparent and rigorous while keeping ‘true’ to the participants’ stories.69 Although qualitative data analysis is inductive and focused on meaning, methods of data analysis vary in aim and have ontological and epistemological basis.70 There are two key approaches to analysing qualitative data – inductive analysis and deductive analysis.
- Inductive analysis involves coding data without trying to fit it into a pre-existing coding frame or the researcher’s analytic preconceptions.
- Deductive analysis is driven by theoretical interest and may provide a more detailed analysis of some aspects of the data. It tends to produce a less detailed description of the overall data.
Types of qualitative data analysis
There are different data analyses used in healthcare research, including content analysis, discourse analysis, thematic analysis, interpretive phenomenological analysis, narrative analysis and grounded theory analysis. The common ones include content analysis, discourse analysis and thematic analysis, and as beginner researchers, we will focus on these.
- Content analysis: is a method of unobtrusively investigating vast volumes of textual material to detect trends and patterns in words used, their frequency, their connections, and the structures and discourses of communication.71 It transforms qualitative input into quantitative data by quantifying words, messages or concepts and analyses the relationships between the concepts. The goal of content analysis is to explain the features of the document’s content by looking at who says what to whom and to what effect.72 Content analysis requires that the text be broken down into manageable codes for analysis. The analysis is conducted in the following stages, decontextualization, recontextualization, categorisation and compilation.73 In the decontextualization stage, the researcher must familiarize themselves with the data (read through the transcribed test) to understand the data before breaking it into smaller meaning units assigned codes.73 Recontextualisation involves checking that all aspects of the content have been addressed with regard to the aim of the study.73 The original text is reread with the final list of codes, and any missed relevant text is included. The codes (meaning units) are condensed in the categorisation stage, and themes and categories are identified. In the final stage – compilation, the analysis and writing-up process begin once the categories have been established.73 Abroms et al. 2011 used content analysis to analyse the content of 47 iPhone applications (apps) for smoking cessation and their adherence to the U.S. Public Health Service’s 2008 Clinical Practice Guidelines for Treating Tobacco Use and Dependence.74 The study found that apps identified for smoking cessation had low levels of adherence to key guidelines in the index.74
- Discourse analysis: investigates language in use instead of psychological factors such as attitudes, memories, or emotions.75 Language is studied in discourse analysis in terms of construction and function and is viewed as a tool for producing reality. Thus, discourse analysis investigates how particular concerns are produced in people’s narratives and the variety in these accounts to examine the link between language and social reality.76 Discourse analysis can be used in health care to analyse interpersonal communication processes between doctors or nurses and patients, interprofessional conversation and in-depth interviews about lay health beliefs.76,77 The article by Brooks et al. 2019, utilised discourse analyses as one of the analytical strategies to analyse the relationships and relationality with companion animals as therapeutic agents in the context of people’s wider social networks.78
- Thematic analysis: is a technique for finding, examining, classifying, and reporting themes in data collection.79 It involves identifying codes or units of analysis that emerge from the data. Thematic analysis is the most common form of qualitative data analysis.79 It offers a flexible analytic strategy that may be adjusted to suit the objectives of numerous studies, offering a detailed and intricate description of the data.79 Six phases have been identified in the thematic analysis and include familiarizing yourself with your data, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report.79 The process is iterative and reflective and evolves, constantly moving back and forth between the phases. For researchers early in their research career, thematic analysis offers a more approachable analysis, as it does not demand the in-depth theoretical and technological understanding that other qualitative techniques entail.79 An example is the study by Danielson et al. 2015, which used thematic analysis to identify common concerns facing pharmacy experiential education (EE) programs.80 The themes identified were site capacity, workload/financial support, quality assurance, preceptor development, preceptor stipends, assessment, onboarding, and support/recognition from the administration.80
