How Do You Judge Whether the Qualitative Sample Was Adequate?

How many participants are enough in qualitative research?

It is one of the first numbers readers notice: 8 participants, 15 interviews, 32 participants, perhaps 60. The temptation is to decide immediately whether that number looks too small.

Unfortunately, qualitative sample adequacy does not work very well by numerical instinct. A sample of 10 can be defensible for one question and seriously inadequate for another. Even the familiar claim that researchers continued interviewing "until saturation" does not settle the issue unless the paper explains what saturation meant and how it was assessed.

The task is therefore not to find a universal minimum. It is to judge whether the sample generated enough relevant information to support the study's particular analytic purpose and claims.

Adequacy depends on information, not a universal sample-size threshold

Sample adequacy is a methodological judgment, not a magic number

Qualitative researchers still need to think carefully about sample size. What differs from many quantitative studies is the logic used to determine sufficiency.

Malterud, Siersma, and Guassora proposed information power as one way to reason about qualitative interview sample size. Their central argument is intuitive: the more information relevant to the study that a sample holds, the fewer participants may be required. They identify the study aim, sample specificity, use of established theory, quality of dialogue, and analysis strategy as factors affecting information power.

This framework helps move appraisal away from questions such as "Is 15 enough?" toward the more defensible question: "Enough for what, with whom, and given what kind of data and analysis?"

A narrow question may require less information than a broad one

Consider the scope of the research aim.

A tightly focused study examining how experienced emergency nurses in one hospital interpret a particular handover procedure asks for a relatively bounded form of knowledge. A study asking how nurses across different specialties, career stages, institutions, and healthcare systems experience digital transformation contains considerably more potential variation.

The second study may require a larger or more strategically varied sample because the analytic territory is broader. The number alone still does not determine adequacy, but the breadth of the question changes what the sample must accomplish.

Highly specific participants can provide substantial information power

Participant relevance matters alongside participant count.

A small group selected because every participant has extensive direct experience of a narrowly defined phenomenon may provide considerable information. A larger group whose connection to the phenomenon is weak may provide less.

This is why sample adequacy should be considered only after asking whether the participants were appropriate for the qualitative question. Twenty poorly matched participants do not become an adequate sample merely by outnumbering ten highly relevant ones.

Data richness changes what a sample can support

Two studies can recruit the same number of participants and obtain radically different amounts of useful information.

Ten lengthy, focused interviews with participants able and willing to discuss the phenomenon in depth may support a substantial analysis. Ten short interviews dominated by generic answers may not. Malterud and colleagues explicitly include quality of dialogue in their information-power framework.

This does not mean longer interviews are automatically better. Duration is a crude proxy. The real question is whether the data provide the detail, variation, examples, contextual information, and interpretive possibilities required by the study.

Sample adequacy and depth of qualitative data collection are therefore closely related but distinct. One concerns how much relevant evidence the sample collectively provides; the other examines the richness and depth of the evidence generated.

The analysis strategy affects how much data you need

Not every qualitative analysis asks the same thing of a dataset.

An intensive case-oriented analysis may require substantial engagement with a relatively small number of cases. A study attempting to compare patterns across several participant groups needs enough material within and across those groups to support meaningful comparison. A project seeking broad thematic variation may require different sampling than one investigating a narrowly specified experience.

The adequacy judgment must therefore remain connected to what researchers actually do analytically rather than treating "qualitative research" as a single method.

Saturation is useful, but the word alone proves very little

Saturation is frequently invoked to justify qualitative sample size, but the concept has multiple meanings and is not applied consistently across qualitative methodologies.

Hennink, Kaiser, and Marconi demonstrate one reason the term requires precision. In their methodological study of 25 in-depth interviews, they distinguished code saturation, when the range of thematic issues had been identified, from meaning saturation, when researchers had developed a richer understanding of the dimensions and nuances of those issues. In that dataset, code saturation occurred at nine interviews, while meaning saturation required 16 to 24.

The numbers from that study are not universal thresholds. Their importance is conceptual: hearing no new topic is not necessarily the same as understanding existing topics deeply enough.

Not every qualitative methodology should be forced through saturation

Saturation has become so familiar that it is sometimes treated as a compulsory quality marker for all qualitative research. That is too broad.

Malterud and colleagues note that saturation is closely associated with particular methodological traditions and has been used inconsistently. The appropriate justification for sample sufficiency should fit the study's methodological assumptions and analytic purpose.

A paper should therefore not be penalized merely because it avoids the word saturation. The more important question is whether the researchers offer a coherent rationale for why the amount and composition of their data were sufficient for what they attempted to claim.

Subgroups create additional demands on the sample

A total sample can appear substantial until you examine how it is divided.

Suppose a study interviews 24 participants and compares faculty, students, administrators, and technical staff. Six participants per group may or may not provide enough information for the comparisons the researchers make. If those groups are further divided by institution, discipline, or experience, the evidential base behind particular comparisons may become thinner still.

This does not mean every qualitative subgroup needs a predetermined minimum. It means the sample should be assessed at the level where the claims are made.

Recruiting more participants does not automatically repair weak data

Increasing sample size can add information, variation, and opportunities to test developing interpretations. It can also produce more of the same superficial material.

There is therefore a point at which "more participants" becomes an inefficient response to a different problem. If interviews are poorly designed, participants are inappropriate, or researchers fail to probe consequential issues, collecting another 20 interviews may expand the dataset without meaningfully strengthening it.

Likewise, a very large qualitative dataset can create analytic difficulties if the researchers lack the time or resources to engage adequately with the material. More data are useful only when they contribute to the inquiry and can be analyzed with sufficient depth.

Look for a justification rather than a ritual number

Vasileiou and colleagues systematically examined sample-size practices in interview-based qualitative health research and highlighted persistent weaknesses in how sample sufficiency was justified. Their analysis reinforces a practical appraisal principle: the authors should explain why their sample is adequate in relation to the study rather than leaving the number to speak for itself.

These are considerations, not equations. You cannot assign points to each column and calculate the correct qualitative sample size. They help you examine whether the authors' sampling decisions fit the intellectual demands of the study.

Why 12 interviews might be enough for one study but not another

This comparison reveals why asking "Is 12 enough?" is methodologically incomplete. Twelve is neither adequate nor inadequate in isolation. The adequacy judgment emerges from the relationship among the research question, participants, information obtained, analytic strategy, and scope of the claims.

Common mistakes when judging qualitative sample size

Replace “How many?” with “Enough for what?”

When evaluating a qualitative sample, record the participant count, but do not make it your verdict.

Instead, examine what the researchers are trying to understand, how specifically the participants relate to that phenomenon, how much relevant information the data appear to contain, what variation the researchers seek to capture, and how demanding the intended analysis is.

This approach also keeps sample adequacy in proportion when deciding whether the qualitative study is convincing overall. Sample size is one part of the evidential architecture. It cannot compensate for inappropriate participants, shallow data collection, or an analysis that does not adequately engage with the material.

Check whether the qualitative sample provided enough information

Questions about adequate sample size in qualitative research

An adequate sample provides enough relevant information for the claims being made

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.