Qualitative analysis · Lesson 3 of 3

Saturation, triangulation, reflexivity

Rigour here is a visible trail: memos, negative cases, and an honest account of how categories were built.

9 min read · Free — no account needed to read

Saturation is reached when new material stops producing new categories. It is a claim, and claiming it requires evidence: how many cases produced nothing new, and were the later ones deliberately chosen to be different? Saturation asserted after six homogeneous interviews on a broad topic has not been demonstrated.

Memoing is writing notes to yourself throughout analysis — why a code was created, where a category came from, what puzzled you. Memos are the analytic trail: the record that lets somebody follow how you got from transcripts to findings. Without them the process is genuinely unreproducible, and the reader is asked to take your word for it.

Triangulation compares across sources, methods, analysts or theories. When they converge, confidence rises. When they diverge, that is data — the divergence usually means the two are capturing different aspects, which is often the most interesting finding available.

Multiple coders improve dependability, and the number reported matters less than what was done about disagreement. Agreement statistics suit deductive coding with a fixed codebook; for interpretive work the useful practice is coding independently, discussing divergences and refining definitions — and reporting what changed.

Reflexivity and positionality mean stating who you are in relation to this research and how it shaped access, what people told you and what you noticed. This is not confession. A former nurse studying nurses gets candour a stranger would not, and may take for granted what a stranger would see. Both matter to a reader, and only you can report them.

Negative cases — material that contradicts your emerging account — are the strongest evidence of rigour available. Seeking them deliberately, and reporting what you did with them, is what distinguishes analysis from illustration.

See it happen

Nineteen participants describe a process as frustrating. One describes it as straightforward.

The read that arrives on its own

Nineteen out of twenty is a clear pattern; the one is an outlier and not worth the space.

The same situation, with this idea

You treat the one as a negative case and look at what is different. They joined before a reorganisation and still use the old workflow. That single case turns "the process is frustrating" into something much better: the frustration is attributable to a specific change rather than to the task, which is checkable and actionable. The outlier was the most informative interview in the set.

After this, you can

go looking for the case that does not fit, because it usually sharpens the finding.

The takeaway

Rigour here is a visible trail: memos, negative cases, and an honest account of how categories were built.

Easy to take the wrong way

That rigour means imitating quantitative standards. Inter-rater statistics on interpretive work can perform objectivity while the real question — can a reader follow your reasoning — goes unanswered.

Try this

Find one case in your material that does not fit your emerging account, and write what it changes.