Should You Expect Qualitative Research to Be Representative?

Does a qualitative sample need to represent the population?

You open a qualitative paper and find interviews with 18 participants. Perhaps they were recruited purposively from one hospital, school, community, or profession. Your quantitative instincts may immediately raise a familiar objection: how can 18 people represent thousands?

That is a reasonable question, but it may be asking the study to accomplish something it was never designed to do. In much qualitative research, researchers deliberately select participants because they can illuminate a phenomenon, experience, process, or perspective. The central issue is therefore not automatically whether the sample statistically resembles a population.

The harder appraisal question is whether the researchers' claims remain proportionate to what their sampling strategy and data can actually support.

Usually no, but it depends on what the study claims

Representativeness is only meaningful in relation to what researchers want to infer

Statistical representativeness has a specific purpose

In its familiar statistical sense, representativeness concerns whether a sample adequately reflects a defined population on characteristics relevant to the inference being made. Probability sampling provides a formal basis for making population estimates when the design, response process, and analysis support them.

That logic is essential when a study asks questions such as, "What percentage of university students use generative AI each week?" or "What proportion of nurses report burnout?" If researchers want to estimate prevalence in a population, who gets sampled and with what probability becomes fundamental.

Many qualitative questions are different. Consider: "How do first-generation doctoral students experience academic belonging?" The researcher may want to understand how belonging is experienced, negotiated, threatened, or reconstructed. Selecting participants who have relevant experiences may be more useful for that question than constructing a miniature statistical replica of all first-generation doctoral students.

Purposive sampling follows a different logic

Purposive sampling deliberately selects people, settings, documents, events, or other cases because they are expected to be informative about the phenomenon being studied. The point is not that every member of the population has a known probability of selection. Researchers instead make theoretically and substantively informed decisions about which cases can contribute useful evidence.

This is why judging a qualitative study simply by asking whether its participants are demographically proportional to a larger population can be misleading. A study may deliberately recruit people with unusual experiences, contrasting positions, particular expertise, or especially relevant exposure to the phenomenon.

For appraisal, a more productive question is whether the participants were appropriate for the qualitative question. A carefully selected group of relevant participants may provide much stronger evidence for a focused qualitative question than a larger but poorly targeted group.

Diversity and representativeness are not the same thing

A sample can be diverse without being statistically representative. Researchers may intentionally seek variation in age, professional role, experience, location, institutional type, or another characteristic because they expect those differences to reveal contrasting experiences or conditions.

For example, a study examining teachers' experiences implementing an educational technology might recruit teachers from urban and rural schools, experienced and novice teachers, and schools with different levels of technical support. That variation could help researchers examine how implementation differs across circumstances. It does not establish that the resulting proportions mirror the teacher population.

Conversely, demographic resemblance alone does not guarantee a strong qualitative sample. A group that happens to look population-like on age and sex may still omit the people whose experiences are most relevant to the research question.

Do not replace representativeness with a simplistic sample-size rule

Once statistical representativeness is removed as the default criterion, another temptation appears: perhaps qualitative adequacy can simply be judged by the number of participants. That does not solve the problem.

The adequacy of a qualitative sample depends on factors such as the research question, specificity of the sample, study design, quality of the dialogue or observations, analytical strategy, heterogeneity of the phenomenon, and the claims researchers intend to make. Ten highly relevant participants in a narrowly focused inquiry can serve a different analytical purpose from 50 participants recruited for a broad exploration.

Numbers matter, but their meaning comes from the design around them.

What matters is the relationship between the sample and the claim

This is where appraisal becomes more demanding. A qualitative study does not become immune from sampling criticism merely because statistical representativeness was never intended.

Suppose researchers interview 12 volunteers from one highly resourced private university and conclude that "university students prefer AI-assisted assessment." The problem is not simply that 12 is a small number. The conclusion has expanded far beyond the sampling frame and evidence. The study might credibly describe how those participants understood or experienced AI-assisted assessment, but a population-wide claim about what university students prefer requires different evidence.

Qualitative research can support forms of inference beyond the immediate cases

Rejecting statistical representativeness as a universal requirement does not mean qualitative findings are useful only for the exact individuals studied. Qualitative researchers may develop concepts, explanations, mechanisms, typologies, or theoretical propositions that help readers understand related situations.

One important idea is transferability: readers consider whether findings developed in one setting may illuminate another sufficiently similar setting. That judgment requires detailed reporting of the participants, circumstances, setting, and boundaries of the findings. It is not an automatic leap from a small sample to everyone else.

The question of whether qualitative findings may transfer to another context therefore deserves separate appraisal. Representativeness and transferability solve different inferential problems.

Some qualitative studies may pursue broader forms of generalization

It would also be too crude to say that qualitative research never generalizes. Qualitative traditions differ in their epistemological assumptions, sampling strategies, analytical aims, and kinds of inference. Case studies may seek analytical insights from a bounded case. Grounded theory may develop concepts or theoretical explanations whose applicability is explored across contrasting cases. Some large qualitative projects may deliberately include substantial variation across sites or participant groups.

Accordingly, the defensible question is not "Is qualitative research representative, yes or no?" It is "What kind of inference is this study trying to make, and does its design justify that inference?"

What a non-representative sample can still tell you

Common mistakes when judging representativeness

Judge the sampling logic before demanding representativeness

When reading a qualitative paper, begin with its research question rather than its participant count. Ask what the researchers needed to learn and which people, cases, or settings could realistically provide that evidence.

Then inspect the chain from question to sampling to conclusion. The strongest qualitative studies make that chain visible. They explain why particular participants were recruited, provide enough contextual detail to understand who contributed the evidence, and keep their conclusions within defensible boundaries. Transparent reporting frameworks such as SRQR and COREQ can help you identify information that should be available when evaluating qualitative work.

Representativeness is therefore not a universal quality criterion for qualitative research. A better appraisal asks whether the researchers sampled deliberately, collected sufficiently informative evidence, analyzed it rigorously, and made claims consistent with what those choices permit. Those broader considerations contribute to whether the qualitative study is ultimately convincing.

Before criticizing a qualitative study for being non-representative, check:

Questions about representativeness in qualitative research

Do not demand a representative sample unless the intended inference requires one

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.