Description
Tracking new codes per interview in a table and curve so the decision to stop sampling is based on evidence rather than intuition.
Saturation is routinely claimed but rarely demonstrated. Saturation monitoring operationalises it: after each interview the analyst records new codes, revised codes and new themes, then plots new codes per interview. Guest, Bunce and Johnson found most codes in a homogeneous sample emerged within about twelve interviews, while Hennink and colleagues distinguished code saturation (hearing it all) from meaning saturation (understanding it all), which takes longer. Saturation is judged per category: two or three consecutive sessions adding no new codes on core categories, after theoretical sampling toward expected variation. The decision and its evidence are documented, giving stakeholders a defensible answer to 'why this many participants?'.
Attributes
- Discipline
- Research Design & Rigor
- Loss type
- Latent
- Stage
- Discovery
- Data
- Qualitative
- Effort
- Days
Steps
- After each interview: update a tracking table. Columns: interview number, new codes generated, codes revised, new themes identified.
- Plot the curve of new codes per interview. Saturation = the rate of new code generation approaches zero.
- Saturation is indicated when: (a) 2–3 consecutive interviews produce zero new codes on core categories AND (b) you have used theoretical sampling toward expected variation.
- Category-specific saturation: Track saturation per major category. Continue theoretical sampling until all major categories are saturated.
- Document the saturation decision with specific evidence: how many sessions produced no new codes, what theoretical sampling was done.
Sources
- Rocking Qualitative Social Science
- Sampling and Choosing Cases in Qualitative Research
- How Many Interviews Are Enough? (Guest, Bunce & Johnson)
- Code Saturation Versus Meaning Saturation (Hennink, Kaiser & Marconi)
Also known as
- Code saturation tracking
- New-codes curve
- Saturation table
Categories
- Research Operations
- Research Design & Rigor