Thematic analysis in nursing research provides a flexible way to interpret patterns across interviews, focus groups, observations and written accounts. However, credible analysis requires more than highlighting repeated words. Researchers must make clear methodological choices, code data systematically, develop meaningful themes and show how interpretations answer the research question. This guide explains the process step by step.
Nursing researchers often use qualitative data to understand experiences that cannot be reduced to numerical outcomes. These may include living with chronic illness, receiving compassionate care, managing clinical uncertainty or implementing a new service. Thematic analysis can illuminate these experiences while preserving context and complexity.
What is thematic analysis in nursing research?
Thematic analysis is a family of methods for developing and interpreting patterns of meaning across qualitative data. A theme is not simply a topic mentioned by several participants. It is an organised pattern that captures something important about the research question.
For example, interviewees may discuss staffing, time pressure, interruptions and documentation. Listing these as separate topics remains descriptive. A theme such as “safety work under operational pressure” could integrate them into an interpretive account of how nurses protect patients within constrained systems.
The method is popular because it can be used with different theoretical positions and types of qualitative data. That flexibility also creates responsibility. Researchers must specify which form of thematic analysis they are using and ensure that their procedures match its assumptions.
When should nursing researchers use thematic analysis?
Thematic analysis is appropriate when a study aims to explore shared meanings, experiences, perceptions or practices across a dataset. It can support questions about how patients experience care, how nurses implement policy, why an intervention is accepted or resisted, and how professional identities develop.
It may be suitable for:
- Semi-structured or in-depth interviews
- Focus-group discussions
- Open-ended survey responses
- Reflective diaries and field notes
- Documents, online discussions or visual materials
- Qualitative evidence within mixed-methods studies
The method should not be selected merely because qualitative data are available. If a study seeks to build a substantive theory, grounded theory may be more appropriate. If it focuses on the structure of personal stories, narrative analysis may fit better. Interpretative phenomenological analysis is more suitable for detailed idiographic exploration of lived experience in a small, relatively homogeneous sample.
Thematic analysis versus qualitative content analysis
| Dimension | Thematic analysis | Qualitative content analysis |
|---|---|---|
| Main aim | Develop patterns of shared meaning | Organise and interpret content into categories |
| Typical output | Interpretive themes with a central organising concept | Categories, subcategories and sometimes themes |
| Researcher role | Usually acknowledged as active and interpretive | Varies according to the selected approach |
| Frequency | Importance is not determined by frequency alone | May include more structured attention to content occurrence |
The distinction is not absolute because both methods contain variants. The important requirement is to name the chosen approach accurately and follow a coherent analytic process.
Choose the right form of thematic analysis
Researchers sometimes cite Braun and Clarke while applying procedures from coding-reliability approaches without explanation. This creates methodological inconsistency. Before coding, decide whether the analysis is reflexive, codebook-based or focused on coding reliability.
Reflexive thematic analysis
Reflexive thematic analysis treats the researcher as an active interpreter. Themes are developed through engagement with the data rather than discovered as objective entities waiting to be extracted. Coding can evolve throughout analysis, and differences between researchers are viewed as opportunities for reflection rather than errors that must be eliminated.
Codebook thematic analysis
Codebook approaches use a structured framework to support analysis, often in larger teams or applied research. A codebook can clarify definitions and coordinate work while still permitting interpretation. Researchers should explain how the framework was created, refined and used.
Coding-reliability approaches
These approaches commonly involve independent coders, a relatively fixed coding frame and measures of agreement. They are based on assumptions different from reflexive thematic analysis. Inter-coder reliability is not automatically a quality requirement for every thematic method.
Braun and Clarke’s guidance on good practice in thematic analysis warns against combining incompatible procedures without recognising their conceptual differences (Braun & Clarke, 2022). Methodological coherence is more valuable than adding every technique that appears rigorous.
Key decisions before analysing nursing data
A transparent methodology should establish the study’s philosophical position, analytic orientation and practical boundaries.
Inductive or deductive?
Inductive analysis is led primarily by the data, although no researcher approaches data without prior knowledge. Deductive analysis uses an existing theory, framework or question to guide attention. A study may combine both, but the relationship should be explicit.
Semantic or latent?
Semantic analysis concentrates on what participants directly express. Latent analysis examines underlying assumptions, ideas and social meanings. Neither is inherently superior. The research question and theoretical position should determine the depth of interpretation.
Experiential or critical?
An experiential orientation explores how participants understand their realities. A critical orientation asks how language, institutions and power shape those realities. For example, an experiential analysis may explore nurses’ accounts of burnout, while a critical analysis might examine how organisational discourse individualises a systemic staffing problem.
The six phases of thematic analysis in nursing research
The commonly used six-phase process described by Braun and Clarke (2006) is recursive rather than strictly linear. Researchers may return to earlier phases as interpretations develop. An audit trail should record meaningful decisions without falsely presenting analysis as a mechanical sequence.
Phase 1: Become familiar with the data
Familiarisation begins during data collection and transcription. Read transcripts repeatedly, listen to recordings where appropriate and note early observations. Correct transcription errors while protecting participant meaning. Record reactions, questions, contradictions and possible connections in a reflexive journal.
Do not rush this phase. Software can retrieve text efficiently, but it cannot replace close engagement with participants’ accounts. Consider what is said, how it is expressed, what appears uncertain and how the context influences meaning.
Phase 2: Generate initial codes
A code is a concise label attached to a meaningful segment of data. Code inclusively enough to retain context. Work systematically across the full dataset and allow codes to change as understanding deepens.
Suppose a participant says, “I knew the patient was deteriorating, but I waited because the team was overwhelmed and I did not want to overreact.” Possible codes include “recognising deterioration”, “hesitation to escalate”, “team workload” and “fear of appearing alarmist”. The codes should reflect the analytic purpose rather than merely shortening the sentence.
Good coding practices include:
- Keeping the research question visible without forcing data into it
- Coding both expected and surprising material
- Allowing one extract to receive several codes
- Recording why codes are added, merged or renamed
- Preserving contradictory and minority accounts
Phase 3: Develop candidate themes
Review the codes and explore how they combine into broader patterns of shared meaning. A theme needs a central organising concept. It should explain a significant aspect of the dataset rather than function as a storage folder.
Create provisional maps showing relationships between codes, themes and subthemes. Some codes may form themes, while others remain contextual or are discarded because they do not contribute to the research question.
Phase 4: Review and refine themes
Test each candidate theme against its coded extracts and the complete dataset. Ask whether the extracts form a coherent pattern, whether themes are sufficiently distinct and whether the overall analysis tells a convincing story.
At this stage, a theme may be divided, combined, redefined or removed. Search actively for data that complicate the interpretation. A credible analysis should not erase participants whose experiences differ from the dominant pattern.
Phase 5: Define and name themes
Write a concise definition explaining the scope, central concept and contribution of every theme. Identify what is included and excluded, how subthemes relate, and how the theme answers the research question.
A name such as “communication” is usually too broad. “Speaking up in an unreceptive hierarchy” conveys an analytic idea and hints at the theme’s argument. Avoid catchy labels that obscure meaning.
Phase 6: Write the analysis
Writing is part of analysis, not simply the final presentation stage. Select vivid extracts that demonstrate the pattern, then interpret them. Do not expect quotations to speak for themselves.
A strong thematic section normally:
- Introduces the theme’s central organising concept.
- Explains how the pattern appears across the dataset.
- Uses carefully selected quotations as evidence.
- Interprets each extract in relation to the theme.
- Considers variation or contradiction.
- Connects the interpretation to the question and literature.
A worked nursing example
Imagine a study asking how newly registered nurses experience escalation of patient deterioration. Interviews produce codes such as “trusting intuition”, “seeking senior confirmation”, “fear of criticism”, “workload delaying response” and “supportive medical review”.
A descriptive result might group these under “barriers” and “facilitators”. A more interpretive analysis could develop the theme “Escalation as a negotiated professional risk”. This theme explains that escalation is not experienced as a simple protocol step. Participants balance clinical concern against hierarchy, credibility and team capacity.
Possible subthemes could include “borrowing confidence from senior nurses” and “calculating the social cost of speaking up”. Quotations would illustrate these patterns, while the discussion would connect them to psychological safety, professional accountability and patient-safety systems.
How to improve trustworthiness and rigour
Rigour means producing an analysis that is coherent, transparent, well supported and appropriate to the selected method. It does not come from adding procedural checkboxes.
Maintain reflexivity
Researchers should examine how professional background, assumptions and relationships affect interpretation. A nurse researching colleagues may recognise clinical nuances but may also normalise practices that an outsider would question. Reflexive notes can document these influences and how they informed decisions.
Keep a decision trail
Retain versions of code lists, theme maps, analytic memos and definitions. The trail should show how the analysis developed. It need not imply that another researcher would produce identical themes.
Use rich supporting extracts
Choose extracts that demonstrate the theme and preserve enough context for interpretation. Avoid using only short fragments or unusually dramatic quotations. Represent the range of relevant accounts.
Seek conceptual coherence
The research question, theoretical position, chosen thematic approach, coding practice and quality criteria should fit together. For reflexive thematic analysis, researcher subjectivity is a resource to examine, not a bias that can be removed through consensus coding.
Report data handling ethically
Explain consent, anonymisation, secure storage and access. Remove names and indirect identifiers from quotations. A rare job role, location or clinical incident may identify a participant even when a pseudonym is used.
Using NVivo or other software
NVivo, ATLAS.ti and MAXQDA can help organise data, retrieve coded extracts, compare cases and visualise relationships. They do not perform the intellectual work of thematic development. Researchers remain responsible for decisions about meaning, relevance and theme boundaries.
Manual analysis can be appropriate for a small dataset. Spreadsheets, tables, printed transcripts and sticky notes may support transparent coding. Choose tools according to the size of the dataset, team arrangements and required security rather than assuming software makes analysis more rigorous.
Students needing specialist software support can review our nursing NVivo data analysis service. The page has a separate transactional purpose; this guide remains focused on teaching the method.
Common thematic analysis mistakes
Treating interview questions as themes
Headings that repeat the interview schedule usually organise responses by topic rather than analyse patterns. Themes should develop an insight that cuts across questions where appropriate.
Using frequency as the only test of importance
A common idea is not automatically meaningful, and a less frequent account may reveal an important safety or equality issue. Judge significance in relation to the research question.
Confusing codes with themes
Codes label segments; themes organise multiple codes around a central concept. A list of brief labels such as “stress”, “communication” and “support” is unlikely to constitute a developed thematic analysis.
Claiming themes simply emerged
This phrase hides the researcher’s interpretive role. Explain how themes were constructed through coding, comparison, reflection and review.
Mixing incompatible quality procedures
Do not claim reflexive thematic analysis and then present inter-coder agreement as proof that subjective interpretation was eliminated. Select procedures consistent with the chosen approach.
Reporting quotations without analysis
A string of participant quotations remains largely descriptive. Explain what each extract reveals and how it contributes to the theme’s argument.
How to report thematic analysis in a nursing dissertation
The methodology chapter should justify the method, name the specific approach and explain philosophical and analytical choices. Describe the dataset, transcription, familiarisation, coding, theme development, reflexivity, software and ethical safeguards.
The findings chapter should present themes logically with definitions, interpretation and supporting extracts. Include a thematic map or summary table when it improves clarity. The discussion should compare interpretations with existing evidence, address alternative explanations and explain implications for nursing practice.
Avoid repeating the same findings in both chapters. Findings establish the analytic account; discussion locates that account within wider scholarship and practice.
Thematic analysis in nursing research checklist
| Area | Check |
|---|---|
| Purpose | The method fits the research question and data. |
| Approach | The form of thematic analysis is named and justified. |
| Orientation | Inductive/deductive and semantic/latent choices are clear. |
| Coding | The full dataset is examined systematically. |
| Themes | Each theme has a central organising concept. |
| Reflexivity | The researcher’s interpretive role is addressed. |
| Evidence | Claims are supported by contextualised extracts. |
| Ethics | Consent, confidentiality and secure handling are explained. |
| Reporting | The findings answer the research question coherently. |
Frequently asked questions
How many interviews are needed for thematic analysis?
There is no universal number. Sample adequacy depends on the research question, population, depth of data, analytic approach and project scope. Justify the sample rather than relying on one numerical rule.
Can thematic analysis be used for open-ended survey responses?
Yes, provided responses contain enough detail for meaningful interpretation. Very short answers may support a more descriptive content analysis instead.
Do two researchers need to code every transcript?
Not always. Independent coding and agreement suit some codebook or reliability approaches. Reflexive thematic analysis does not require consensus coding as proof of objectivity.
Can themes be decided before data collection?
A deductive study may begin with theoretical concepts, but final themes still require engagement with the dataset. Predetermined headings should not be misrepresented as findings developed from analysis.
Is NVivo required?
No. NVivo organises data but does not create a credible interpretation automatically. Manual analysis can be suitable when it is systematic, secure and transparent.
What is the difference between a theme and a subtheme?
A theme captures a broad pattern with a central organising concept. A subtheme highlights a distinctive aspect within that pattern without becoming independent of the main theme.
Need support with qualitative nursing analysis?
Nursing Dissertation Service can help students clarify their analytic approach, review coding logic, refine themes and strengthen the connection between findings, literature and nursing practice. Support is aligned with the research question, university guidance and ethical responsibilities.
Explore qualitative nursing dissertation support.
Related Nursing Guides
- How to Conduct Thematic Analysis in a Systematic Review: Nursing Guide
- How to Appraise Qualitative Research for a Nursing Dissertation
- Nursing Dissertation Help by Stage: Topic, Methods, Analysis & Editing
Conclusion
Thematic analysis in nursing research can produce rich explanations of patient, caregiver and professional experience. Its flexibility is valuable only when researchers make coherent choices and report them transparently. Select the appropriate approach, engage deeply with the data, develop themes around shared meaning, practise reflexivity and support interpretations with contextualised extracts. This creates findings that are methodologically credible and genuinely useful to nursing knowledge and practice.
References
- Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
- Braun, V., & Clarke, V. (2022). Toward good practice in thematic analysis: Avoiding common problems and be(com)ing a knowing researcher. International Journal of Transgender Health, 24(1), 1–6. https://doi.org/10.1080/26895269.2022.2129597
