How to Report Qualitative Data Analysis
Qualitative data analysis is the systematic examination and interpretation of material obtained from interviews, observations, documents, diaries, focus groups, open-ended responses, and other qualitative data sources.
In a scientific article, qualitative analysis should not be reduced to a single statement such as “the data were analyzed thematically.” Readers should be able to understand how the researchers moved from raw data to codes, categories, themes, concepts, or theoretical explanations.
What Should the Qualitative Analysis Section Explain?
Readers should be able to determine:
- Which qualitative analytical approach was used?
- Why was that approach selected?
- How were the data prepared for analysis?
- How were codes generated?
- How were categories or themes developed?
- Was the analysis inductive, deductive, or combined?
- How many researchers participated in analysis?
- How was software used, if applicable?
- How was the researcher's interpretive role addressed?
- How was analytical quality supported?
How Should a Qualitative Analysis Method Be Selected?
The analytical approach should be selected in relation to the research question, qualitative design, epistemological position, data type, and the kind of knowledge the study seeks to produce.
Examples include:
- Thematic analysis: may be used to identify and interpret patterns of meaning across a data set.
- Content analysis: may be used to examine data systematically through categories and content structures.
- Grounded theory analysis: may be used to develop a theoretical explanation of a social or behavioral process.
- Phenomenological analysis: may investigate the meaning and structure of lived experience.
- Discourse analysis: may examine how language and discourse construct social meanings.
- Narrative analysis: may examine how individuals organize and communicate experiences through stories.
What Is Thematic Analysis?
Thematic analysis is an approach to identifying, developing, and interpreting meaningful patterns within qualitative data.
It is not a single fixed method. Depending on the theoretical position of the study, thematic analysis may be implemented in more structured coding-reliability forms or through a reflexive analytical approach.
How Is Thematic Analysis Conducted?
The exact process varies, but commonly involves:
- Familiarization with the data
- Generating initial codes
- Examining patterns among codes
- Developing candidate themes
- Reviewing themes
- Defining and naming themes
- Developing the analytical narrative
These phases are not necessarily linear. Researchers may repeatedly return to the original data, codes, and developing themes.
What Is Reflexive Thematic Analysis?
In reflexive thematic analysis, the researcher is not treated as a passive observer but as an active participant in the production of analytic meaning.
Themes are therefore not viewed as objects simply “discovered” within the data. They are developed through systematic, theoretically informed, and reflexive engagement with the data set.
For this reason, measures of inter-coder agreement or kappa statistics are not necessarily required or methodologically appropriate in reflexive thematic analysis.
What Is Content Analysis?
Content analysis systematically examines textual or visual material through codes, categories, and patterns of meaning.
Depending on the study, approaches may include:
- Conventional or inductive content analysis
- Directed or deductive content analysis
- Summative content analysis
What Is Inductive Analysis?
In inductive analysis, codes and categories are developed primarily from the data rather than being determined entirely before analysis.
This can be particularly useful for exploratory research questions.
What Is Deductive Analysis?
In deductive analysis, data may be examined using concepts or codes derived from a theory, model, conceptual framework, research question, or pre-existing coding framework.
Using predefined codes does not necessarily mean that new meanings emerging from the data must be ignored.
Can Inductive and Deductive Coding Be Combined?
Yes. Some studies begin with predefined categories while also allowing new codes to develop from the data.
When this approach is used, authors should explain which parts of the analysis were deductive and which were inductive.
What Is a Code?
A code is a concise analytic label used to identify a meaningful segment of data.
Codes may be:
- Descriptive
- Interpretive
- Process-oriented
- Emotion-focused
- Conceptual
A useful code should carry analytical relevance to the research question rather than function merely as a generic topic label.
How Is Coding Performed?
Coding may be conducted line by line, by meaning units, by paragraph, or at another level appropriate to the analytical approach.
For example, the statement:
“After working here for a long time, I am no longer as surprised by patients' behaviors as I used to be.”
might be coded, depending on the study, as:
- Adaptation
- Emotional desensitization
- Adjustment to behavior
- Professional normalization
The appropriate code depends on the research question and analytical framework.
What Is a Codebook?
A codebook is an analytical tool containing code names, definitions, inclusion and exclusion criteria, and examples.
It may be particularly useful in structured analyses involving several coders.
A codebook may include:
- Code name
- Definition
- When the code should be applied
- When it should not be applied
- Example data segment
- Associated higher-order category
What Is a Category?
A category may be formed by grouping codes that share similar meanings under a broader conceptual structure.
Categories often organize what the data are about, whereas themes may provide a broader interpretation of what the pattern means in relation to the research question.
What Is a Theme?
A theme is an analytical pattern that captures something meaningful in relation to the research question.
A theme should not simply be a heading under which similar quotations have been grouped. It should communicate what the data reveal analytically.
For example:
Weak theme: Work Experience
More analytical theme: Repeated Exposure Normalizes Professional Responses
Is a Theme the Same as a Topic?
No. Labels such as “family,” “work,” “problems,” or “emotions” often identify general topic areas rather than analytical themes.
A theme should communicate a more specific pattern of meaning.
How Should Themes Be Distinguished?
Themes should be sufficiently distinct conceptually while contributing together to the overall interpretation of the study.
The same data segment may sometimes relate to more than one theme; qualitative categories do not always need to be mutually exclusive.
What Is a Subtheme?
Subthemes may be used to identify particular dimensions or variations within a broader theme.
However, excessive use of subthemes can fragment the analysis and reduce interpretive depth.
Data Analysis in Grounded Theory
In grounded theory, data collection and analysis often proceed concurrently. Analytical findings may guide subsequent data collection.
Depending on the grounded theory tradition, processes may include:
- Initial coding
- Focused coding
- Constant comparison
- Category development
- Theoretical sampling
- Memo writing
- Theoretical integration
Classical, Straussian, and constructivist grounded theory approaches do not use identical analytical procedures. Authors should identify the tradition they follow.
Are Open, Axial, and Selective Coding Required in Every Grounded Theory Study?
No. This terminology is associated particularly with certain grounded theory traditions and should not be treated as mandatory across all grounded theory approaches.
What Is Constant Comparison?
Constant comparison involves repeatedly comparing new data with earlier data, codes, and developing categories.
This process helps clarify the properties, boundaries, similarities, and differences of analytical categories.
What Is Theoretical Sampling?
Theoretical sampling refers to selecting additional participants, events, or data sources in response to the needs of developing analytical categories.
It is therefore not simply another term for purposive sampling. The emerging theory guides subsequent sampling decisions.
What Is Memo Writing?
Memos are analytical notes in which researchers develop thoughts about codes, categories, relationships, and emerging explanations.
Memo writing can support:
- Tracking conceptual development
- Exploring relationships among categories
- Documenting analytical decisions
- Developing theoretical explanations
Phenomenological Data Analysis
Phenomenological analysis seeks to understand how lived experiences are interpreted and how the structure or meaning of an experience may be described.
Phenomenology does not consist of one universal analysis method.
Researchers may use:
- Descriptive phenomenology
- Interpretive phenomenology
- Interpretative Phenomenological Analysis (IPA)
- Another clearly defined phenomenological approach
The analytical procedure should be consistent with the philosophical and methodological tradition selected.
What Is IPA?
Interpretative Phenomenological Analysis focuses on detailed examination of how individuals make sense of significant lived experiences.
Analysis is often conducted intensively at the individual-case level before patterns across cases are considered.
How Is Discourse Analysis Reported?
Discourse analysis may examine not only what participants say but also how language constructs realities, identities, and social meanings.
It should therefore not be reported as merely another method for extracting themes.
How Is Narrative Analysis Reported?
Narrative analysis may examine how individuals construct stories, sequence events, identify turning points, negotiate identity, and position experiences within broader social contexts.
Depending on the approach, emphasis may be placed on narrative content, structure, or performance.
How Is Document Analysis Conducted?
Policy documents, reports, institutional records, diaries, archival materials, and other texts may serve as qualitative data.
Authors should describe:
- Source of the documents
- Selection criteria
- Historical or institutional context
- Authenticity and credibility considerations
- Analytical method
How Are Interviews Prepared for Analysis?
Where interviews are audio-recorded, they should be transcribed in a manner appropriate to the research design.
Possible approaches include:
- Verbatim transcription
- Cleaned verbatim transcription
- More detailed transcription including interactional features
The chosen approach and the treatment of relevant non-verbal features should be described where necessary.
How Can Transcription Accuracy Be Checked?
Transcripts may be compared with audio recordings, selected sections may be independently checked, or another verification procedure may be used.
Where automated transcription software is used, authors should state whether the transcripts were reviewed manually.
Can Artificial Intelligence Be Used for Transcription or Coding?
Automated transcription or AI-supported tools may be used for technical assistance in some qualitative studies.
Authors should report:
- Which tool was used
- Its purpose
- How outputs were checked by researchers
- How confidentiality and data security were protected
- Who made the final analytical decisions
Use of AI to suggest codes or themes does not transfer methodological or ethical responsibility away from the researchers.
Do MAXQDA, NVivo, or ATLAS.ti Perform the Analysis?
These programs can assist with data organization, coding, retrieval, comparison, memoing, and visualization.
They do not independently produce the scientific interpretation of a qualitative study. Analytical meaning remains dependent on the researcher's methodological decisions.
The software name and, where relevant, version may be reported.
What Should Be Reported When Several Researchers Analyze the Data?
The roles of each researcher should be described clearly.
For example:
- Did two researchers code the complete data set independently?
- Did a second researcher review only selected data?
- Was a codebook developed collaboratively?
- How were differences discussed?
- How were final themes agreed or developed?
When Is Inter-Coder Reliability Appropriate?
Inter-coder agreement may be useful in structured content analyses where a predefined coding framework is applied consistently across a data set.
If Cohen's kappa or a similar coefficient is used, report:
- Which data were evaluated
- Number of coders
- Statistic used
- How the value was interpreted
However, high inter-coder agreement is not a mandatory quality criterion in every qualitative methodology.
What Is Reflexivity?
Reflexivity involves systematic consideration of how researchers' positions, experiences, assumptions, and relationships may influence the production and interpretation of knowledge.
Researchers may describe:
- Professional position
- Relationship with participants
- Previous experience in the research setting
- Analytical assumptions
- Decision-making processes
Is It a Problem If the Researcher Is an Insider?
Not necessarily. Being a member of the organization, profession, or community being studied may create both advantages and methodological challenges.
Insider knowledge may improve contextual understanding, while prior assumptions and power relationships may influence data collection and interpretation.
These issues should be addressed reflexively rather than hidden.
How Should Saturation Be Addressed?
Saturation can have different meanings across qualitative methodologies.
Instead of simply stating that “saturation was reached after 15 interviews,” authors should explain:
- How saturation was defined
- What analytical level was considered
- Whether additional interviews produced new information
- How and when the decision was made
Is Saturation Required in Every Qualitative Study?
No. Saturation is important in some methodological traditions but is not a universal requirement for justifying sample adequacy across all qualitative approaches.
Reflexive thematic analysis and some phenomenological traditions, for example, may justify sample adequacy differently.
How Should Negative or Deviant Cases Be Handled?
Data that do not fit a developing theme or explanation should not automatically be excluded.
Negative cases may help:
- Clarify theme boundaries
- Develop alternative explanations
- Produce a more nuanced analysis
What Is Triangulation?
Triangulation refers to examining a phenomenon through multiple sources, methods, researchers, or theoretical perspectives.
Possible forms include:
- Data triangulation
- Researcher triangulation
- Method triangulation
- Theoretical triangulation
Triangulation is not mandatory in every qualitative study and should not be added mechanically simply to claim greater “validity.”
What Is Member Checking?
Member checking involves seeking participant feedback on data, interpretations, or preliminary findings in some qualitative studies.
Possible forms include:
- Summary checking at the end of an interview
- Transcript review
- Feedback on preliminary interpretations
- Discussion of themes
Member checking should not be treated as universally mandatory across all qualitative methodologies.
What Is an Audit Trail?
An audit trail documents methodological and analytical decisions throughout the research process.
Records may include:
- Versions of coding frameworks
- Theme revisions
- Analytical memos
- Sampling decisions
- Analytical meeting notes
Such documentation can strengthen transparency and traceability.
How Should Participant Quotations Be Used in Qualitative Findings?
Participant quotations may provide evidence supporting an analytical theme, but quotations should not substitute for interpretation.
A useful structure is:
Theme → Analytical explanation → Subtheme where relevant → Illustrative quotation → Interpretation of the quotation
How Many Quotations Should Be Used?
There is no fixed number of quotations required for each theme.
Quotations should:
- Represent the theme meaningfully
- Avoid unnecessary repetition
- Reflect relevant variation where useful
- Support the analytical claim directly
How Should Quotations Be De-Identified?
Where participant confidentiality must be protected, names may be replaced with participant codes.
Examples include:
P1, P2, P3
or:
Participant 7, woman, 12 years of professional experience
Descriptors should not inadvertently make participants identifiable.
Can Qualitative Findings Be Quantified?
Some forms of qualitative content analysis may report frequencies of codes or categories.
In thematic and interpretive studies, however, the importance of a theme should not necessarily be judged solely by how many participants mentioned it.
Qualitative significance and statistical frequency are not the same concept.
How Should Qualitative Analysis Be Written in the Methods Section?
A practical sequence is:
Analytical approach → Methodological/theoretical basis → Data preparation → Coding → Theme/category development → Researcher roles → Software → Quality/reflexivity procedures
Example Thematic Analysis Paragraph
Example: Interviews were transcribed verbatim and reread before analysis to support familiarization with the data. The data were examined using an inductive thematic analysis approach aligned with the research question. Initial codes were developed from meaningful data segments, after which related codes were organized into candidate themes. Themes were reviewed in relation to the complete data set and research question, refined, and named. Analytical memos were maintained throughout the process to document the development of themes and key interpretive decisions.
Example Reflexivity Statement
Example: The researcher's professional experience in the study setting supported contextual understanding but was also recognized as a potential source of prior assumptions. Reflexive notes were therefore maintained during data collection and analysis, and pre-existing expectations were compared continuously with interpretations emerging from the data.
Example of Reporting a Multi-Coder Process
Example: An initial coding framework was independently applied to a subset of the data by two researchers. Code definitions were compared and refined to develop a shared codebook. Coding then continued across the full data set, with analytical differences discussed during research meetings.
This example is appropriate for coding-consensus approaches. It should not be treated as a required procedure in methodologies such as reflexive thematic analysis.
How Can the Qualitative Results Section Be Structured?
A useful structure is:
Main theme → Analytical explanation → Subthemes/patterns → Participant evidence → Connection to the research question
A table listing themes alone is not sufficient. The main text should explain what each theme means analytically.
Do Not Confuse Methods and Results
How coding and theme development were conducted belongs in the Methods section. The themes produced by the analysis belong in the Results section.
Detailed comparison of themes with previous literature generally belongs in the Discussion.
Common Mistakes in Qualitative Data Analysis
- Stating only that “content analysis was performed” without describing the process
- Treating thematic analysis and content analysis as interchangeable labels
- Failing to identify the methodological basis of the analytical approach
- Reporting codes as though they were themes
- Using generic topic labels as themes
- Using participant quotations instead of analysis
- Judging theme importance only through frequency counts
- Making the researcher's analytical role invisible
- Suggesting that MAXQDA or NVivo independently performed the analysis
- Calculating inter-coder kappa in every qualitative study regardless of methodology
- Treating member checking as universally mandatory
- Claiming saturation without explaining how it was assessed
- Excluding negative cases because they do not fit the developing interpretation
- Conducting coding in grounded theory without progressing toward theoretical development
- Calling any thematic analysis “phenomenological analysis” merely because the study is phenomenological
- Failing to disclose AI-supported transcription or analytical assistance
Check Before Submission
- Is the qualitative analytical approach clearly identified?
- Is the approach aligned with the research question and qualitative design?
- Are transcription and data-preparation procedures described?
- Is it clear how codes were generated?
- Is theme or category development explained?
- Is the inductive, deductive, or combined orientation clear?
- Are researcher roles in analysis described?
- If software was used, is its role described accurately?
- Is reflexivity addressed where relevant?
- If saturation was used, is its assessment explained?
- Are quality procedures consistent with the methodological approach?
- Do quotations support rather than replace the analysis?
- Were negative or deviant cases considered where relevant?
- Is participant confidentiality protected in quoted material?
