RESEARCH GUIDE
How to Analyze Qualitative Data
From Familiarization to Codes, Themes and Reporting
How to analyze qualitative data depends on the question and the method, but most approaches move from close reading of the material, to coding it, to building patterns, to reporting them with evidence from the data. No single sequence suits every method, and the logic they share is that each analytic decision stays visible and the interpretation rests on the data.
Qualitative analysis is the systematic interpretation of non-numerical material such as interview transcripts, focus group discussions, open-ended responses, documents and field notes. Its aim is to describe patterns of meaning, experience or practice and to support them with evidence from the data, not to estimate population quantities or test hypotheses. Knowing how to analyze qualitative data therefore means choosing an approach that fits the question and showing how each claim was reached. The qualitative coding process, in which segments are labeled and then organized, is a central tool in most approaches, though the analysis does not stop at the codes. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, supports qualitative analysis with a coding framework suited to the method, the building of themes and an audit trail, while the research question, the interpretation and the conclusions remain the researcher’s.
SOURCES OF MATERIAL
Types of Qualitative Data and How Each Is Approached
Analysis begins with what the material is, because the way it was produced decides what can be claimed from it and how closely it must be read.
Interview Transcripts One-to-one interviews give depth on individual accounts, so analysis keeps each account intact and attends to what the questions themselves invited.
Focus Group Discussions Group talk includes interaction between participants, so the analysis records who spoke and whether a view was challenged, shared or shaped by the group.
Open-Ended Survey Responses Short free-text answers are written to a fixed prompt, often in a few words, so the wording of the prompt shapes the replies and the coding stays close to what respondents wrote.
Documents and Records Policies, reports, case notes and archives were produced for another purpose, so their origin, audience and period are considered before they are read as evidence.
Observation and Field Notes Notes record what the researcher saw and interpreted, so the separation of description from reflection is kept visible within the notes themselves.
Visual and Digital Material Photographs, videos and social media posts carry context that text alone loses, and consent and identifiability need attention before any extract is quoted.
APPROACH COMPARISON
Thematic, Content and Framework Analysis Compared
Three approaches are used most often for interview and text data, and they differ in how coding is organized and what the analysis produces, as the table sets out.
| Feature | Thematic analysis | Content analysis | Framework analysis |
|---|---|---|---|
| Main aim | Patterns of meaning across a dataset | Systematic description of what the material contains | Comparing cases across a set of themes |
| Coding scheme | Evolving in reflexive versions, a codebook in structured ones | Categories defined in advance, from the data or both | A framework developed from the data and the aims |
| Role of counting | Rare, and prevalence is stated with care | Frequencies reported where sampling and the coding unit support them | Not the aim, as cells summarize each case |
| Typical output | Named themes supported by extracts | Category definitions with examples and, if suitable, frequencies | A matrix of cases by themes |
| Suits | Exploring experiences or perceptions | Large amounts of text or comparing sources | Team projects and policy or health questions |
| Rigor shown by | Reflexivity and a documented process | Clear category definitions and, for shared schemes, agreement checks | A traceable framework that returns to the cases |
KEY PLANNING DECISIONS
The Order of Work From Transcript to Theme
These qualitative analysis steps overlap and are revisited as understanding grows, so the order is a guide to the work and not a fixed procedure.
Prepare and Read Check transcripts against the recordings and read the whole set closely before coding, noting first impressions so later interpretation can be traced to them.
Code the Data Label meaningful segments with short codes, inductively from the data, deductively from a framework or both, keeping each code's meaning stable as coding proceeds.
Group Codes Into Patterns Cluster related codes into candidate themes or categories, and try more than one arrangement before settling, since the first grouping is rarely the most faithful to the data.
Test Against the Data Return to the coded extracts and the full dataset to see whether each pattern holds, and look for cases that contradict or complicate it.
Define and Name Write what each theme or category means and where its boundaries lie, so that a reader can see how it differs from its neighbors.
Keep an Audit Trail Record coding decisions and reflective memos as they are made, since a reader cannot judge an analysis whose development cannot be followed.
COMMON ISSUES
Weaknesses a Reviewer Looks For in Qualitative Analysis
A reader judges qualitative findings by the route from extract to claim, and these weaknesses are the ones that tend to break it.
Topic Summaries as Themes A heading such as communication names a topic and not a finding, so each theme is stated as a claim about the data that the extracts support.
Selective Quotation Vivid extracts can misrepresent the dataset, so quotations are chosen to show how common and how varied a pattern is, and contrary cases are reported.
Counts Without a Rationale Frequencies of codes are descriptive and depend on the sampling and the coding unit, so they appear only where they help the argument and are not read as prevalence.
Rules Borrowed From Another Method A fixed codebook or an agreement statistic suits some approaches, while reflexive ones treat coding as interpretive, so rigor measures follow the method chosen.
Software as the Analyst Coding programs organize and retrieve segments but do not decide what the data mean, so automated suggestions are drafts the researcher accepts or rejects.
Method Cited, Not Followed Citing a published approach while skipping its stages, such as the review of candidate themes, invites challenge, so the write-up shows how each stage of the chosen method was carried out.
REPORTING
Showing the Analysis in a Qualitative Report
A reader needs to see how the findings were reached, and reporting guidelines for qualitative work list what to include.
Qualitative studies are commonly reported with COREQ for interviews and focus groups or with SRQR for qualitative research more broadly, and APA 7 includes JARS-Qual for qualitative manuscripts. The report names the analytic approach and its source, describes how codes were developed and by whom, explains how themes were reviewed against the data and states how the researcher’s own position was considered. Quotations carry an anonymized participant label, and the findings keep the participants’ words distinct from the researcher’s interpretation.
QUALITATIVE METHODS
Choosing a Qualitative Method for the Question
The question decides the method, since patterns across a set, a developing theory and a single group's experience point to different approaches, and each method page sets out its own standards.
Thematic Analysis
Patterns of meaning across interviews
Content Analysis
Defined categories applied to text
Framework Analysis
A matrix of cases by themes
Grounded Theory
Theory built by constant comparison
Mixed Methods Research
Integrating text and numbers
Qualitative & Mixed Methods
The full set of approaches
ANALYSIS SERVICES
Where Analytical Support Enters a Qualitative Study
Support can sit alongside the researcher's own coding, from the choice of approach to the write-up of the findings, with a final check at the end.
Qualitative & Mixed Methods Analysis
A coding framework suited to the method, systematic coding of your material and the building of themes, with every interpretive judgment remaining yours.
Research Planning & Study Design
The choice of approach, the sampling logic and the plan for analysis, settled before the first interview is transcribed.
Results Reporting & Interpretation
Theme tables and methods wording organized to COREQ, SRQR or the standard your journal names, so each claim traces back to extracts.
Independent Statistical Review
A completed qualitative analysis is read for the trail from extracts to themes, the fit between method and claims, and any counts that overreach.
FREQUENTLY ASKED QUESTIONS
Questions About How to Analyze Qualitative Data
The line between a code and a theme, when analysis should start, short free-text answers and the researcher's own standpoint each come up early in a qualitative project.
What is the difference between a code and a theme?
A code labels a segment of data, while a theme is a pattern of meaning built from several codes that answers part of the research question. Codes are numerous and mostly descriptive, themes are fewer and interpretive, and the step from one to the other is where most of the interpretive work happens.
Should analysis begin during data collection or afterward?
In most qualitative approaches it can begin early, because reading the first interviews shows whether the guide is working and what to ask next. Grounded theory requires analysis alongside collection, whereas other approaches often hold formal coding until most of the material is gathered. Either way, early impressions are kept as memos and are not treated as findings.
Can open-ended survey responses be analyzed like interviews?
Partly. Short free-text answers lack the depth of an interview, so they are usually coded into categories, which suits content analysis or a structured form of thematic analysis. Counts can be reported if the respondents were sampled in a way that supports them, and where respondents wrote at length a more interpretive reading becomes possible.
Should my own views be set aside during analysis?
In reflexive approaches the researcher’s perspective is treated as part of the analysis and is made visible and not removed, usually through memos and a reflexivity statement. Some phenomenological traditions ask for a deliberate suspension of assumptions. Which stance applies depends on the method chosen, and the report says which was taken.
How do you protect my research data and confidentiality?
Interview transcripts and other qualitative materials often contain identifiable details, and your data and project details are handled confidentially. Project information is not published, and researcher names and institutions are not used publicly without permission. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, such as a protocol, an analysis plan or a dataset, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Talk Through Your Analysis Plan
Before Coding Begins
Coding is easier to set up well than to repair, so the approach, the coding unit and the form of the audit trail are worth settling while the transcripts are still being prepared.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014