Qualitative Sample Justification Guide

When to use it

When you plan a study with individual interviews or focus groups and have to say how many people you will interview and why. And again during fieldwork, to record how much new information each interview adds.

It is the fill-in version of the qualitative sample size tool: the same questions and the same method, without the part that calculates. The tool builds the paragraph and does the arithmetic for you; this guide is for taking the reasoning into your research plan, your field notes, or the conversation with whoever is asking you for a number.

Before you use it

It will not give you a number. Malterud and colleagues warn that their model is not a checklist to calculate N, and that its five dimensions trade off against each other (Malterud et al., 2016, p. 1756). That is why the guide does not add up the marks: it asks you to write down the reason behind each one, which is what later supports the justification.

The model was developed for individual interview studies (Malterud et al., 2016, p. 1758). With focus groups, skip the five dimensions and fill in only the empirical reference and part 2.

Part 2 assumes a codebook-based analysis. Braun and Clarke (2021) consider saturation generally coherent with that kind of thematic analysis, but not with reflexive thematic analysis. If that is your approach, stay with part 1.

The document

Study: [study name] Lead: [person or team] Data collection: [individual interviews / focus groups] Date: [date]

Part 1. Before fieldwork: justify your sample

Mark where your study sits on each dimension and write down why. The dimensions trade off against each other: they are not added up.

1. Study aim. How broad is the aim of your study?

• Narrow: one specific task or moment
• In between
• Broad: a whole experience or several contexts

Why: [ ]

2. Sample specificity. How close are your participants to the experience you are studying?

• Dense: precise recruitment criteria, seeking variation within the experience
• In between
• Sparse: broad criteria or convenience recruitment

Why: [ ]

3. Established theory. Does your study build on an existing theory or framework?

• Yes: a specific framework guides the interview guide and the analysis
• Partly
• No: the study is exploratory

Why: [ ]

4. Quality of dialogue. How do you expect the interview conversations to go?

• Strong: the interviewer knows the topic and has interviewing practice
• I can't anticipate it yet
• Weak: little interviewing practice, or a topic that is hard to talk about

Why: [ ]

5. Analysis strategy. How will you analyze the data?

• In depth, case by case: narrative or discourse analysis
• A mix of both
• Cross-case: shared patterns across participants, as in a thematic analysis

Why: [ ]

Empirical reference. In Hennink and Kaiser's (2022) systematic review, most of the 16 tests with interviews reached saturation between 9 and 17 interviews, and the four comparable tests with focus groups between 4 and 8 groups. It is a starting point, not the number for your study.

Initial estimate: [number] [interviews / focus groups]

Justification paragraph: [Summarize the five dimensions and the initial estimate. End by saying that the sample size will be revisited during fieldwork.]

Part 2. During fieldwork: assess saturation

Method by Guest, Namey and Chen (2020). Record interviews in the order you conducted them, and count codes at a single level of your codebook. A new code is one that did not come up in any earlier interview.

Base size: [ ] 4 (recommended) [ ] 5 [ ] 6

Run length: [ ] 2 interviews [ ] 3 interviews

New information threshold: [ ] 5% or less [ ] 0%

Codes in the base (sum of the first [4 / 5 / 6] interviews): [ ]

Each run adds up the new codes of its interviews and divides them by the codes in the base. Runs overlap: they move forward one interview at a time. The first run at or under the threshold closes the assessment.

Saturation at interview: [number], reported as [number]+[run length].

Sentence for the report: With a base size of [ ] interviews and runs of [ ], the [ ] new information threshold was reached at [ ]+[ ] interviews (Guest, Namey and Chen, 2020).

Check before you close

• Does each dimension have its reason, not just the mark?
• Does the initial estimate read as an estimate, and does the paragraph say it will be revisited in the field?
• Are the interviews recorded in the order they were conducted?
• Did you count codes at a single level of the codebook?

What part 2 does not tell you

Meeting the threshold does not guarantee that saturation was reached: Guest and colleagues compare the threshold to a p-value, a transparent convention and not a proof. The method was tested with inductive thematic analysis on narrowly defined questions, and its use with other epistemological perspectives is untested. For a more conservative assessment, use runs of 3 or the 0% threshold.

Grounding

• Malterud, K., Siersma, V. D., & Guassora, A. D. (2016). Sample size in qualitative interview studies: Guided by information power. Qualitative Health Research, 26(13), 1753–1760. The five dimensions in part 1.
• Guest, G., Namey, E., & Chen, M. (2020). A simple method to assess and report thematic saturation in qualitative research. PLOS ONE, 15(5), e0232076. The method in part 2.
• Hennink, M., & Kaiser, B. N. (2022). Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Social Science & Medicine, 292, 114523. The empirical reference.
• Braun, V., & Clarke, V. (2021). To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qualitative Research in Sport, Exercise and Health, 13(2), 201–216. When part 2 does not fit.