A qualitative findings chapter turns hours of interviews into a few claims a reader can check: the pattern, the evidence in the participants' words, and what it means for the research question. Writing up qualitative findings well is mostly choosing what to leave out. EdCitation publishes this as the qualitative findings guide of How to write a dissertation.
We read each rule here at its source on 26 September 2026: the methods' own authors and a grounded theory design paper, the COREQ and SRQR standards, the UK Data Service and BERA on anonymity, and the dissertation guides of Southampton, Sheffield Hallam and Capella. Their disagreements are named.
Nothing in EdCitation writes a word of your dissertation. It finds the studies your themes will be compared with through Find sources, builds the method texts' references with Cite a source, and checks the chapter's list with Verify references, free.
What does a qualitative findings chapter have to do?
Show the patterns in your data, prove each with evidence, and say what each means for your question. Capella University (2024) calls the researcher's job digesting the raw data, so readers reach the results without reading every transcript yet can still see the words they rest on.
Braun and Clarke (2006) set the same standard for thematic analysis: tell the data's story convincingly, with enough extracts to show each theme, inside an argument that goes beyond description. SRQR asks for the main findings and for evidence, such as quotations, behind each abstract claim (O'Brien et al., 2014).
How you coded belongs in the methodology chapter; the findings show what the analysis produced.
How should a qualitative findings chapter be organised?
Around the unit your method produces, which is not always a theme.
| Method | What the chapter is built around | What its authors ask for | The usual mistake |
|---|---|---|---|
| Reflexive thematic analysis | Themes: patterns of shared meaning, each with a central organising concept | About 2 to 6 themes and subthemes for a dissertation or one doctoral chapter | Topic summaries, or interview questions used as themes |
| Interpretative phenomenological analysis | Group themes, each told through individual participants | A narrative that unfolds, with convergence and divergence between people | Losing the individual in the average |
| Grounded theory | Categories integrated around a core category | A storyline that presents the theory and its propositions | Categories reported without the theory that joins them |
| Narrative analysis | Narratives, grouped into a typology or read closely a few at a time | Case studies or vignettes for each type, or attention to how stories are told | Grouping stories until the deviant ones and their context vanish |
Themes, not topic summaries
Braun and Clarke (n.d.) separate a theme, which says something about shared meaning, from a topic summary, which gathers everything said about one subject; "Experiences of Y" and "Benefits of X" are their examples of topic-summary names. Codebook and framework approaches use topic-style themes deliberately; say which approach you used.
IPA, grounded theory and narrative analysis
Nizza et al. (2021) name four marks of a strong IPA write-up: an unfolding narrative, a vigorous account of experience, close reading of participants' words, and attention to where participants converge and diverge. Each quotation makes one point; the next moves the story on.
A grounded theory chapter presents a theory: Chun Tie et al. (2019) describe a storyline that joins the categories around the core category and yields a set of theoretical propositions. Riessman (2005) describes thematic narrative analysis as a typology illustrated with case studies or vignettes, and asks what happens to the deviant stories a typology leaves out.
What order should the themes go in?
In the order that answers the question best, stated at the start. Capella University (2024) names two common structures: one section per research sub-question, or each participant first and the shared themes after.
Should the themes be shown in a thematic map or a table?
Usually one or the other, near the start, so the reader holds the whole before the parts. Braun and Clarke (2006) use thematic maps to show themes, subthemes and the links between them, and their FAQ adds that maps drawn during analysis can go in a dissertation's appendices.
A table does the same job with more detail. Capella suggests an appendix table of themes with representative quotations, and SRQR allows judicious tables and figures (O'Brien et al., 2014). A useful opening table has three columns: the theme, its central idea in one sentence, and its subthemes. Grounded theorists often draw a model instead, which Braun and Clarke (n.d.) say they do not particularly advocate for thematic analysis.
How many quotations should I use, and how should I present them?
Enough to show each theme is shared, few enough that your analysis carries the chapter. No university read for this guide sets a number.
University of Southampton Library (2026) suggests grouping the common and interesting responses and quoting two or three of each. Braun and Clarke (2006) call one or two extracts for a theme a sign of weak analysis: a reader who has not seen the data cannot tell a pattern from an anecdote. Capella wants only the few that best exemplify a theme, each standing for others.
Length, editing and layout
Short phrases sit inside your sentence in quotation marks; longer passages become a block quotation, as your style requires. The experienced researchers Corden and Sainsbury (2006) interviewed agreed that readers must always see where the participant's words end and the author's begin.
Editing varies. Southampton accepts deleting fillers such as "um" unless they matter to the analysis; narrative researchers keep hesitations because they interpret them, and some researchers could not say what their own dots meant, a pause or cut words (Corden & Sainsbury, 2006). State your convention once.
Participant identifiers and anonymity
Give every quotation an identifier that cannot identify the speaker. COREQ's item 29 asks whether each quotation was identified, for example by participant number (Tong et al., 2007); Southampton adds that details such as age or gender go in only as the research needs them.
Numbers or pseudonyms show how quotations spread across the group and stop a few articulate people dominating, while a string of descriptors can identify someone (Corden & Sainsbury, 2006). The UK Data Service (n.d.) calls this jigsaw identification: a rare job in a small place, a detailed timeline. For anonymisation it recommends consistent bracketed replacements such as [colleague 1], generalising detail rather than blanking it, and a log of every change.
The British Educational Research Association (2024) warns that anonymity may be impossible in a small community or a well-known institution, and asks researchers who anonymise or fictionalise to explain how and why.
How much description, and how much interpretation?
More than most first drafts contain. The first pitfall Braun and Clarke (2006) list is not analysing at all: extracts strung together, or comment that only paraphrases them. Their checklist asks for data interpreted, not described, and a balance of narrative and extract.
Here the guides differ. University of Southampton Library (2026) describes results as factual, without interpretation; Capella says findings interpret the data but leave other authors to the discussion. Interpreting your own data belongs in qualitative findings; comparing it with the literature is where universities part company.
Findings and discussion: one chapter or two?
Either, depending on the university. Capella keeps them apart: chapter four answers the question from the data, chapter five sets the answers against the literature. Sheffield Hallam University Library (n.d.) says qualitative results are often presented alongside the discussion, since they mean little without interpretation. Southampton allows both, combined as two sections of one chapter or as each finding followed by its discussion.
SRQR says the line blurs in qualitative work (O'Brien et al., 2014). Your handbook decides; the discussion chapter guide covers the comparison.
How do reflexivity and trustworthiness show in the findings chapter?
In the language, in the cases that do not fit, and in the limits. The methodology chapter describes the techniques; the findings show them working.
- Language. Braun and Clarke (n.d.) object to "themes emerged" in reflexive thematic analysis, because it implies meaning waiting in the data and a neutral researcher. IPA uses "emergent" in its own sense, and the example Sheffield Hallam reproduces says "two themes emerged": the rule belongs to one method.
- Divergent cases. COREQ's item 32 asks for diverse cases or minor themes (Tong et al., 2007). An honest negative case does more for credibility than a theme that fits everyone.
- Prevalence words. Braun and Clarke (2006) question what "the majority of participants" tells a reader; SRQR says frequency counts play a limited role unless they matter to interpretation (O'Brien et al., 2014). If you count, say what you counted.
- Your position. SRQR's item 6 asks for the researcher's characteristics and influence, item 19 for the limits of trustworthiness. Where your positionality shaped a theme, say so at that theme.
Which reporting checklist applies: COREQ or SRQR?
COREQ for interviews and focus groups, and SRQR for qualitative research of any design. COREQ has 32 items in three domains: the research team and reflexivity, study design, and analysis and findings (Tong et al., 2007). SRQR has 21 items (O'Brien et al., 2014). Neither is required unless your handbook makes it so, but both are free and are what a journal would ask. Run the chapter through them:
- Check that the chapter names each theme clearly and describes minor themes or diverse cases (COREQ 31 and 32).
- Check that every quotation is identified, for example by participant number, and none can identify its speaker (COREQ 29).
- Check that the data shown and the findings claimed are consistent, with nothing asserted that no extract supports (COREQ 30; SRQR 17).
- Check that the main findings are presented as interpretations, themes or a model, in line with the method (SRQR 16).
- Check that researcher characteristics and trustworthiness techniques are reported somewhere (SRQR 6 and 15; COREQ 1 to 8).
- For reflexive thematic analysis, add Braun and Clarke's 15-point checklist, updated on their FAQ page.
- Run the reference list through Verify references before the draft goes to your supervisor.
A worked example: one theme written up weakly, then well
The example is ours, and the study, participants and quotations are invented: an imagined reflexive thematic analysis of interviews with student nurses about belonging on placement.
The weak version
Theme 2: Experiences of placement. Many participants talked about their placements. Some felt welcome and some did not. "The staff were nice" (P1). "It was hard at first" (P2). "Some days I felt in the way" (P5). "My supervisor was really good" (P9). This shows placement experiences varied, and themes emerged around support.
The name is a topic summary; "many" counts nothing; the quotations are strung together; the comment paraphrases; "themes emerged" hides the analyst.
The better version
Theme 2: Belonging on loan. For most students in this (invented) study, welcome on the ward was real but conditional: given when they were useful, withdrawn when they were not. Seven of the ten described being treated as staff on busy shifts and as spare on quiet ones. S4 (second year) put it plainly: "When it's short-staffed I'm one of the team. When it's quiet I'm the student again." The move from "one of the team" to "the student again" suggests that usefulness, not presence, earned belonging. S7 (third year) described the same condition as a tactic: "I learnt to keep moving. If you stand still you look spare." Not every account fitted. S9 (first year), supervised by one nurse for the whole placement, felt included whatever the ward's pace, which suggests a stable relationship may remove the condition rather than meet it.
The name carries a claim; the count says what was counted; each quotation does different work; the analysis says what the words mean; the divergent case sets a limit. Comparison with published studies follows in a combined chapter, or waits for the discussion.
How EdCitation helps with the qualitative chapter
For this chapter's references EdCitation is the best tool, since each one is looked up in the publisher's record, never written, and a qualitative chapter leans on method texts cited from memory and comparison studies found in a hurry. We ran each free tool on 26 September 2026.
Find sources takes a topic, or the claim you want supported, across about 300 million published works. Our claim search, "nursing students feel they do not belong on clinical placement", returned 12,051 results, led by a 2009 paper in the Australian Journal of Advanced Nursing, "The duration of clinical placements: a key influence on nursing students' experience of belongingness", then two qualitative studies of clinical placement in BMC Nursing (2005 and 2019). Read each before it goes near your theme.
Cite a source takes a DOI, title, PubMed ID or ISBN and returns the entry in any of six styles, screened for retraction. From DOI 10.1191/1478088706qp063oa it returned, in APA 7, a correct entry: Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa. From ISBN 9781473953239 it returned "Braun, V., & Clarke, V. (2021). Thematic analysis: A practical guide. SAGE Publications.", with the title in sentence case, as APA 7 asks. Check one thing by hand: the publisher's page gives October 2021 while the authors' own FAQ cites the book as 2022, so take the year from your copy's copyright page.
Verify references checks a pasted list or an uploaded paper. We gave it Braun and Clarke's 2006 paper, the COREQ paper with its year changed to 2009, and an article we invented. Verdicts, word for word: the 2006 paper verified ("The DOI resolves to this record and the title matches."), the altered COREQ entry doubtful ("A record with this title exists, but it was published in 2007, not 2009."), and the invented one not found ("No publisher's or registry's record matches this reference."). A source it cannot reach is shown as "could not check", never as not found.
For the whole uploaded draft, References from a file looks up each reference and pairs each in-text citation with its entry; a free account gets three checks a month, and Pro ($8 a month) lifts the limit. Check your paper reads your handbook into a checklist, free, and Theoretics QA (Max, $24 a month) judges the draft against it, rewriting nothing. Pricing has the plans. None of it writes a theme or a sentence of your analysis.
Quick questions
How many themes should a qualitative dissertation have?
For reflexive thematic analysis, Braun and Clarke suggest about 2 to 6 themes and subthemes for an undergraduate or master's dissertation or one doctoral chapter. Grounded theory produces categories and a core category instead.
How many quotations should each theme have?
No university read for this guide sets a number. Southampton suggests two or three per point; Braun and Clarke call one or two for a whole theme weak.
Can I write that themes "emerged" from the data?
Not in reflexive thematic analysis, whose authors say it hides the researcher's part. IPA uses "emergent" in its own sense; follow your method's vocabulary.
Should findings and discussion be separate chapters?
It depends on the university: Capella separates them, Sheffield Hallam often combines them, Southampton allows both. Your handbook decides.
Can EdCitation check the references in my findings chapter?
Yes. EdCitation's Verify references checks a pasted list or uploaded paper free, marking each entry verified, doubtful or not found. It never writes any of the chapter.
References
- Braun, V., & Clarke, V. (n.d.). FAQs. Thematic Analysis, University of Auckland. https://www.thematicanalysis.net/faqs/
- Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
- British Educational Research Association. (2024). Ethical guidelines for educational research (5th ed.). https://www.bera.ac.uk/publication/ethical-guidelines-for-educational-research-fifth-edition-2024-online
- Capella University. (2024). Qualitative dissertation chapter guide (Version 2.4). Office of Research & Scholarship. https://assets.capella.edu/campus/doctoral-programs/QualitativeDissertationChapterGuides.pdf
- Chun Tie, Y., Birks, M., & Francis, K. (2019). Grounded theory research: A design framework for novice researchers. SAGE Open Medicine, 7, 2050312118822927. https://doi.org/10.1177/2050312118822927
- Corden, A., & Sainsbury, R. (2006). Using verbatim quotations in reporting qualitative social research: Researchers' views. Social Policy Research Unit, University of York. https://www.york.ac.uk/inst/spru/pubs/pdf/verbquotresearch.pdf
- Nizza, I. E., Farr, J., & Smith, J. A. (2021). Achieving excellence in interpretative phenomenological analysis (IPA): Four markers of high quality. Qualitative Research in Psychology, 18(3), 369–386. https://doi.org/10.1080/14780887.2020.1854404
- O'Brien, B. C., Harris, I. B., Beckman, T. J., Reed, D. A., & Cook, D. A. (2014). Standards for reporting qualitative research: A synthesis of recommendations. Academic Medicine, 89(9), 1245–1251. https://doi.org/10.1097/ACM.0000000000000388
- Riessman, C. K. (2005). Narrative analysis. In Narrative, memory & everyday life (pp. 1–7). University of Huddersfield. https://eprints.hud.ac.uk/id/eprint/4920/
- Sheffield Hallam University Library. (n.d.). Presenting qualitative data. Dissertations and research projects. https://libguides.shu.ac.uk/researchprojects/presenting-qualitative-data
- Tong, A., Sainsbury, P., & Craig, J. (2007). Consolidated criteria for reporting qualitative research (COREQ): A 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care, 19(6), 349–357. https://doi.org/10.1093/intqhc/mzm042
- UK Data Service. (n.d.). Anonymisation for text data. https://ukdataservice.ac.uk/learning-hub/data-producer-support/preparing-data-for-sharing-and-reuse/anonymising-data-and-protecting-participants/anonymisationfortextdata/
- University of Southampton Library. (2026, June 18). Results and discussion. Writing the Dissertation. https://library.soton.ac.uk/writing_the_dissertation/results_discussion