Coding and theming get you to a set of findings. This guide covers what happens next: checking that those findings actually hold up, deciding how to show them, and writing them up in a way that lets other researchers trust — and if needed, scrutinize — your conclusions.
Checking your themes actually hold up
Before writing anything, it’s worth going back through the full dataset and asking whether each theme genuinely fits every segment coded under it, or whether a theme is really two different patterns wearing one name. This step is sometimes skipped under deadline pressure, but it’s usually where the biggest revisions happen — a theme that looked solid across ten interviews can turn out to only fit six of them once you look closely.
Qualitative researchers use different vocabulary for this than quantitative researchers, and it’s worth knowing both. Lincoln and Guba (1985) proposed credibility, transferability, dependability, and confirmability as the qualitative equivalents of internal validity, external validity, reliability, and objectivity. Tracy (2010) later reframed quality more broadly around eight markers, including worthy topic, rich rigor, sincerity, and meaningful coherence — useful as a checklist when you’re not sure your write-up is actually saying enough.
Two practical techniques come up constantly in methods sections:
- Member checking — going back to participants with your interpretation to see if it resonates with their own experience.
- Negative case analysis — actively looking for data that contradicts your theme, rather than only for data that confirms it.
Choosing how to show your findings
Different findings call for different formats, and using the wrong one for what you’re trying to say is a common way for otherwise-solid analysis to read as weaker than it is.
Direct quotes
Still the backbone of most qualitative reporting — a well-chosen quote does more to convince a reader than a paragraph describing what participants said. The skill is picking quotes that are representative, not just the most quotable ones.
Frequency tables
Useful when a reviewer or reader will reasonably ask “how common was this?” — a simple table of how many participants or documents a code appeared in adds a layer of transparency, without pretending qualitative frequency means the same thing statistical significance does.
Network and relationship visuals
When the finding is about how codes relate to each other — which ideas co-occur, which themes sit under which broader category — a network diagram or hierarchy communicates that structure far faster than prose can. This is especially useful when a project has enough codes that the relationships between them are themselves part of the finding.
Case-based summaries
When the unit of interest is a single participant, organization, or document rather than a cross-cutting pattern, a case summary — walking through one case in depth — can carry more explanatory weight than a table ever would.
Writing the methods section
Whatever format you choose for the findings themselves, the methods section carries the weight of establishing trust. At minimum, it should be specific about: how data was collected and from whom, which analytical approach was used and why (see the previous guide in this series for the thematic analysis / content analysis / grounded theory distinction), how codes were developed — deductively, inductively, or both — and what steps were taken to check the analysis, such as member checking, a second coder, or negative case analysis.
Vague methods sections (“data was analyzed thematically”) are one of the most common weaknesses reviewers flag in qualitative submissions — not because the analysis itself was necessarily weak, but because there’s no way for a reader to evaluate it without knowing how it was actually done.
Closing the loop
Good qualitative reporting does the same thing good coding does: it keeps every claim traceable back to real evidence. A reader should be able to follow the line from a theme in your findings section, back through your coding decisions, to the actual words a participant used. That traceability — more than any particular format — is what makes qualitative research credible.
Reporting in Byleron QDA: every figure and table you build stays linked to the underlying coded data, so exporting a network diagram or frequency table for your paper doesn’t mean losing the connection back to the original evidence behind it.
References
- Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic Inquiry. Sage.
- Tracy, S. J. (2010). Qualitative quality: Eight “big-tent” criteria for excellent qualitative research. Qualitative Inquiry, 16(10), 837–851.