Qualitative research often produces rich and detailed information from interviews, focus groups, observations, and other sources. However, collecting this information is only the beginning. Researchers must also make sense of the data and identify the patterns, ideas, and experiences that are relevant to their research questions.
One of the most widely used approaches for analysing qualitative data is thematic analysis. Thematic analysis involves systematically examining qualitative data to identify, analyse, and report recurring patterns or themes within the dataset.
Although the process may appear straightforward, conducting a rigorous thematic analysis requires careful attention at every stage. A commonly used approach involves six key steps: familiarising yourself with the data, generating initial codes, developing themes, reviewing themes, defining and naming themes, and producing the final report (Braun & Clarke, 2006; Byrne, 2022).
1. Familiarising Yourself with the Data
The first step in thematic analysis is becoming thoroughly familiar with the data.
For interview-based research, this usually begins with transcribing the audio recordings. Transcription converts recorded interviews into written documents that can be systematically reviewed and analysed. Researchers may assign alphanumeric identifiers to transcripts to maintain an organised dataset. For example, the transcript from the first participant can be labelled P1, the second P2, and so forth.
Once the interviews have been transcribed, the researcher should read and re-read the transcripts carefully. This repeated reading is important because it allows the researcher to develop a deeper understanding of participants' experiences and begin noticing potentially important patterns.
Where appropriate, participants may also be given an opportunity to review their transcripts through member checking. This can help confirm whether the written transcript accurately reflects what they said during the interview.
During this stage, researchers should take notes about interesting observations, recurring ideas, unexpected responses, and possible relationships between different parts of the data. Qualitative analysis software such as NVivo can also be introduced at this stage to organise and manage the dataset.
The goal is not yet to identify final themes. Instead, the researcher is building a detailed understanding of the data before beginning formal coding.
2. Generating Initial Codes
The second step involves generating initial codes.
Coding involves assigning labels to sections of data that appear meaningful or relevant to the research questions. A code might represent an idea, experience, behaviour, perception, or issue expressed by a participant.
For example, if several participants describe difficulties accessing healthcare because of transportation problems, sections of their responses might initially be coded as transport challenges, distance to healthcare facilities, or limited transportation options.
Coding should be conducted systematically. One practical approach is to work through the transcripts one document at a time, carefully examining each participant's responses and assigning relevant codes.
Researchers may use open coding, particularly when they want themes and concepts to emerge from the data rather than imposing predetermined categories on participants' responses. Open coding encourages the researcher to remain attentive to what participants are actually saying and can help reduce the influence of assumptions made before analysing the data.
NVivo and other qualitative data analysis software can be useful during this stage because they allow researchers to organise coded sections, compare responses, retrieve related passages, and manage large datasets more efficiently.
3. Developing Initial Themes
After the initial coding has been completed, the next step is to begin developing themes.
At this stage, the researcher looks across the codes and considers which ones appear to be related. Similar or connected codes can be grouped together to form broader categories or potential themes.
For example, codes such as high transport costs, long distance to facilities, and lack of public transport may point towards a broader theme such as Barriers to Accessing Healthcare.
This stage goes beyond simply identifying individual ideas. The researcher begins examining the relationships between different codes and asking what they collectively tell us about the research questions.
Not every code will necessarily become part of a final theme. Some may be too narrow, occur infrequently, overlap with other codes, or ultimately prove irrelevant to the research questions.
4. Reviewing and Refining the Themes
The fourth step involves reviewing the themes that have been developed.
At this point, researchers need to determine whether the proposed themes accurately represent the underlying data. A theme should have sufficient supporting evidence and should tell a meaningful story about an important aspect of the dataset.
Researchers may need to:
• Check whether enough data support each theme.
• Combine themes that overlap substantially.
• Separate themes that contain several distinct ideas.
• Remove themes that are not sufficiently supported by the data.
• Revisit the original transcripts to ensure that important information has not been overlooked.
• Refine the codes and themes where necessary.
This stage is particularly important because initial themes are not necessarily final themes. Qualitative analysis is an iterative process, and researchers may move back and forth between the transcripts, codes, and themes as their understanding of the data develops.
5. Defining and Naming the Themes
Once the themes have been reviewed and refined, the researcher needs to clearly define and name each one.
A good theme should have a clear focus and should communicate the central idea that the theme represents. The researcher should be able to explain what the theme is about, what aspect of the data it captures, and how it contributes to answering the research questions.
Theme names should therefore be concise but meaningful. Rather than using vague labels, researchers should choose names that give the reader an immediate sense of what the theme represents.
Each theme should also be clearly connected to the supporting data. Participant quotations can be particularly useful at this stage because they provide direct evidence of the experiences and perspectives represented by the theme.
However, quotations should not simply be presented without explanation. The researcher should interpret the quotations and explain how they support the argument being made.
6. Writing the Thematic Analysis Report
The final step is presenting the findings in a clear and coherent report.
A good thematic analysis report does more than list themes and participant quotations. It should tell a clear story about what the data reveal and explain how the findings answer the research questions.
For each theme, the researcher should explain the central idea, present relevant evidence from the dataset, and interpret what the findings mean in relation to the study objectives and existing literature.
Participant quotations can help illustrate important findings, but they should be used selectively. The strongest reports combine participants' own words with the researcher's interpretation and analytical explanation.
The final report should therefore provide a concise but convincing account of the patterns identified within the data and demonstrate how those patterns contribute to answering the research questions.
References
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://sci-hub.ru/10.1191/1478088706qp063oa
Byrne, D. (2022). A worked example of Braun and Clarke’s approach to reflexive thematic analysis. Quality & Quantity, 56(3), 1391-1412. https://doi.org/10.1007/s11135-021-01182-y
Frequently Asked Questions About Thematic Analysis
What is thematic analysis?
Thematic analysis is a qualitative data analysis method used to identify, organise, analyse, and interpret patterns or themes within a dataset. It is commonly used with interview transcripts, focus group discussions, open-ended survey responses, and other forms of textual data.
What are the six steps of thematic analysis?
A commonly used six-step process involves:
• Familiarising yourself with the data
• Generating initial codes
• Developing initial themes
• Reviewing and refining themes
• Defining and naming themes
• Producing the final report
The process is not always strictly linear. Researchers may return to earlier stages as they develop a deeper understanding of their data.
What is the difference between coding and thematic analysis?
Coding involves assigning labels to meaningful sections of qualitative data. Thematic analysis is the broader analytical process through which researchers use codes and other analytical work to identify and interpret patterns or themes within the dataset.
In simple terms, coding is one important part of thematic analysis rather than a separate replacement for it.
Can NVivo be used for thematic analysis?
Yes. NVivo is widely used to organise and manage qualitative datasets. It can help researchers store transcripts, create and organise codes, retrieve coded passages, compare data, and develop and review themes.
However, NVivo does not perform the analytical thinking for the researcher. The researcher remains responsible for interpreting the data, deciding which codes are meaningful, developing themes, and explaining what the findings mean.
How many themes should a qualitative study have?
There is no fixed number of themes that every qualitative study should have. The appropriate number depends on the research questions, the size and richness of the dataset, and the patterns identified during analysis.
The focus should be on developing meaningful, well-supported themes rather than trying to reach a particular number.
Should qualitative research include participant quotations?
Participant quotations can be valuable for demonstrating how themes are grounded in the data. However, quotations should be selected carefully and accompanied by appropriate interpretation.
A strong thematic analysis does not simply present a collection of quotations. It explains what the quotations demonstrate and how they contribute to answering the research questions.
Is thematic analysis suitable for a dissertation or thesis?
Yes. Thematic analysis can be appropriate for undergraduate dissertations, master's theses, doctoral research, and other qualitative studies, provided that the method is suitable for the research questions and is applied systematically and transparently.
Researchers should clearly explain how the data were collected, coded, analysed, and developed into themes so that readers can understand how the findings were reached.
Need Help With Qualitative Data Analysis?
At EviMine Research & Analytics, we support students, researchers, and organisations with qualitative research and data analysis. Our services can assist with interview transcription, qualitative coding, thematic analysis, NVivo-based analysis, interpretation of findings, and academic research support.
Whether you are working with a few interviews or a large qualitative dataset, we can help you organise your data, identify meaningful patterns, develop themes, and present your findings clearly and professionally.
Ready to discuss your research project?
At EviMine Research and Analytics, we provide professional support across the research process. Visit our Research & Analytics Services page to learn more about how EviMine can support your research:
https://evimine.com/services/