Introduction to Abductive Thematic Analysis
Qualitative research analysis is an interpretive process that is taken up by a researcher to work in such a way that some relevant meaning is assigned to the dataset and it starts to make sense. With the help of data, it becomes possible to dive deep into the social world, getting close to the true meaning and understanding of the participants in the research process and at the same time presenting a solid description of the phenomenon. For a qualitative researcher, it is important to not present the data in elaborate pages, rather he is expected to synthesise and condense the data into meaningful information so that it becomes easy for the reader to comprehend, visualise and rationalise the theoretical and practical implications of the findings.
Thematic analysis is an extremely popular method used for analysis of qualitative data. It captures the pattern in raw data and compiles the data into meaningful themes. The technique is specifically famous for the flexibility it offers and finds its application in both inductive as well as deductive techniques of research. Flexibility does offer a lot of advantages and popularity to thematic analysis but because of this trait, its application is being done with a lot of “mishmash” of different approaches and divergent philosophies. This has played a detrimental role in the credibility of thematic analysis.
Social Sciences research is broadly divided into three categories, deductive, abductive and inductive. The deductive research analysis is theory driven and works on positivist methodologies. It usually tests phenomenon objectively. The inductive approach is exploratory in nature and breaks away from prior assumptions to create innovative theoretical assumptions that are done through interpretive methods. The Abductive technique is somewhere between the inductive and deductive technique. It finds its inception somewhere in the philosophical realm of pragmatism. Unlike the other two approaches, this approach is neither driven by data nor driven by hypothesis. It rather conducts parallel and equal engagement with empirical data and extant theoretical understanding
When a researcher feels that abductive thematic analysis is going to be the right approach for his research, he has to go with a certain mindset. He cannot enter the domain with an open mind as the influence of theoretical understanding would set parameters to what they are looking for since the beginning. It puts a halt on the discovery of arbitrary results or abstract findings that may be insignificant to the research question that has been established.
Likewise, abductive research is a misfit when it comes to adopting empirical data within the theoretical understanding of concepts by the means of simplified testing. This is so because the key purpose of abductive research is to not discover a singular objective truth. It rather aims to find the most logical solution and explanation for a phenomena.
By applying abductive thematic analysis, a researcher intends to heuristically examine the breakdowns where the empirical data is different from what is expected on the grounds of the present theoretical understanding. It brings on the surface the gaps that have been left unrevealed by the extant theoretical frameworks on account of empirical findings. When these identified gaps by abductive technique are innovative and unexpected, the creativity of the abductive researcher should be such that he should be able create theory out of it that is an improved version of the existing theory and generates understanding based on contextual empirical material. Abductive thematic research gets its adjectives of being recursive and iterative in nature as new theoretical development is culled out in areas where already phenomenon are explained adequately by extant literature. The theoretical contributions of abductive research have a degree of generalizability as the theorisation of the raw data moves away from the individualised social setting, as it has links to existing knowledge and understanding
A Step by Step guide to using Abductive Thematic Analysis
1) Transcription and Familiarisation: The researcher collects notes in form of audio recordings and field notes. He has a choice to do the transcription in full after the data collection phase is over. However, it is advisable to take up transcription and data collection in parallel as it becomes apparent where full details of points is required and the researcher can adopt to a different collection method or seek clarification. NVivo software tool works very well in transcription followed by analysis. It gives very handy search and visual display benefits. If you have a smaller and manageable dataset, the same analysis can be achieved by doing the transcription in MS word and applying colour coding on similar or related terms. The colour highlighting tool can be used here.
Researcher should be transparent about the process and ensure to transcribe narratives authentically. He can either record the actual speech of the participants or denaturalise the transcription by improving the grammar and syntax so that the comprehension for the prospective reads becomes easier. Many times researchers use the auto transcription tools of MS Teams/You Tube. This is done because of paucity of time but it is advisable to not deviate towards outsourcing this step because doing it manually can help the researcher to go through the length and breadth of the corpus at this stage. Still, if you decide to use a tool to do the transcription for you, you may listen to the recordings again to ensure accuracy while making reflective notes and highlighting the areas of interest.
2)Coding: The process of coding compresses the huge chunk of qualitative data by sorting and highlighting sentences and paragraphs that are based on characteristics related to each other. Coding is result oriented when it is done in various rounds as just one round rarely succeeds in the identification of all codes.
In the first round of coding, the researcher is able to establish a link between raw data and the cognitive interpretation of data in mind of the researcher. In order to extract the best out of the first round and to extrapolate as much semantic meaning as possible from raw data, it is suggested that each and every point is included as a code in the first round. As the researcher progresses to the second round, the coding becomes more selective as al the codes that could be put under a single head are consolidated and the ones that are insignificant are nit repeated. It is a heuristic process and the researcher has to develop an in-depth understanding of the different patterns and relationship in the data. Before starting the third round of coding, it is suggested to start a code book , develop it and test it.
3) Building the code book: Thematic analysis have made code books popular as they provide clarity and a strong structure to the coding process. In this process, as a prerequisite, you must have a label for each code. The label that you produce needs to be short, concise and remain in vicinity of raw data to avoid an over the periphery jump on conceptual clarity. After this, a definition is offered that highlights the key features of each code and also depicting the. story that each code is exhibiting. Each code works on two criteria. One is “when to use” and the other is “when not to use”. The “when to use” criteria explicitly details under what condition the code relates to a piece of text and the “when not use” criteria tells when there is a potential overlap between two codes. There are contrasting views on the application of the codebook and its usage. The advocates of the reflexive and the inducive form of thematic analysis have rejected the positive application of the codebook. They feel that the codebook is only useful to check. The skill of coding of the researcher and has little or no accuracy in its objectivity. However, he codebook has its own set of unique advantages as it provides an opportunity to the researcher to reflect if he is satisfied with the labelling, terminology and definition of the code book. The biggest advantage of a codebook is that it gives an increased level of verifiability to the research as external readers are able to witness each of the steps taken up in the process of coding. Also when the researcher gets back to the corpus after a break, he is able to connect the dots that why a certain section of data was included under a certain code?
4) Establishing Themes: in the case of abductive thematic analysis, themes and codes are distinct from each other. Codes are precise, specific and concise while themes are complex and include within them a multitude of codes in order to explain a phenomenon. Hence, development of themes begins by looking at relationships between the different codes and sorting them in such a way that the story behind the data can be explained in a systematic manner. Thus, it can be said that, a group of codes that are able to effectively portray a phenomenon can be labelled as a theme. The label that is given to a specific theme should be of simple terminology that is easy to retain, digest while also capturing the essence of the respective theme explicitly.
5) Creating the theory: Till this stage, the data has been examined as codes and themes. At the theorising stage, you explain the relationship and the story between your themes and the entire dataset. It is one of the critical stages of abductive thematic analysis and this is what distinguishes it from other guides to qualitative analysis. The abductive thematic analysis the clustering and the explanation associated with the themes should be guided and not determined by the existing body of theoretical understanding. The deductive theory works in in the reverse fashion where you test data by fitting knowledge into existing theoretical frameworks.
6) Comparison of Datasets: Quantification of data is not necessary for qualitative thematic analysis. When quantification of qualitative data in intended, field work becomes mandatory and the participants need to be given equal opportunity to answer the same question and are faced with similar environmental situations. However, for abductive thematic analysis when we are comparing datasets, two questions can be asked out of qualitative data.
a) Are some of the themes present in only one data set
b) If the theme is found present in datasets from both groups , is the expression of the theme distinct in both the groups.
Most common dataset comparisons that are being found used in abductive thematic analysis are gender, age and differences amongst case studies. Of course, any other comparison that is specific to research can also be undertaken.
7) Display of Data: Data display is associated with improving the data and making it visually more appealing. This does not need data to be quantified in any way. It is associated with reducing the qualitative data in quantitative metrics to improve the visual look of the data. By displaying the abductive thematic analysis, the researcher is able to ascertain that his theoretical contributions are truly representative of raw data. Amateur researchers are often confused about how to create visual display of their data. However, it is relatively simple to construct data displays with basic arrows, using shape and arrow tools that can be found on MS word or similar kind of software. The purpose of qualitative figures is successfully achieved when they are accessible to anyone so researchers should use simple language and not technical words or jargons of any kind.
8)Writing: After having completed the first 7 steps, the findings have to be written. Each theme should have a heading. Also, for each theme there should be a theoretical explanation in detail talking about how theory is linked to empirical data along quotations from raw data to offer empirical evidence for the purpose of theorisation. Quotations chosen by the researcher should be in such a way that they offer a compelling support to the theme. The. should also have the trustworthiness as the veracity of raw empirical data is difficult to check. Finally, the researcher should ensure that the findings give a thick description of the context, participants and the social setting. Write with a persuasive tone so that the reader gets convinced that your findings are substantial and have significant implications.
Leave a Reply