Explainer · Plain-language

What is generalisability?

Generalisability is the extent to which conclusions drawn from a study sample can be applied to the wider population or to other contexts beyond the specific participants studied.

CASRAI plain-language explainers — clear answers to recurring research-administration questions

The step most authors miss

Doing CRediT right? Don’t stop at the statement.

A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.

Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.

From the sample to the population

Researchers almost never study an entire population; they study a sample and hope the findings apply more widely. Generalisability is the strength of that leap. It rests largely on how the sample was drawn: a sample that fairly represents the target population — ideally through random or otherwise systematic selection — supports confident generalisation, whereas a skewed or self-selected sample does not. Generalisability also depends on matching conditions: results obtained under specific circumstances generalise best to similar circumstances. The clearer and more representative the sampling, the more defensible the claim that "what we found here holds there too".

Generalisability and external validity

Generalisability is the practical core of external validity — the two terms are often used closely together. External validity is the broader judgement about whether findings extend to other people, settings, and times; generalisability most often emphasises the population dimension: can we extend from this sample to that population? Both improve with representative sampling, realistic conditions, and replication across diverse groups. And both must be weighed against internal validity, since the controlled conditions that secure a clean causal claim can narrow the range of situations to which a finding readily applies.

It is not always the goal

Not all research aims to generalise in the statistical sense, and assuming it must can be a mistake. Much qualitative research seeks deep, contextual understanding of a particular case rather than broad generalisation; here scholars speak of transferability — providing enough rich detail about the context that readers can judge whether findings apply to their own setting. Even in quantitative work, a finding need only generalise as far as its intended use requires. Good practice is to state plainly the population and conditions to which results are meant to apply, and to avoid over-claiming beyond what the sample can support.

Key facts

At a glance

Definition: extent findings extend from a sample to a wider population
Depends on: representative, adequately sized, well-drawn samples
Related to: external validity (its population-focused core)
Qualitative analogue: transferability, judged through rich context
Risk: over-generalising beyond what the sample supports
Good practice: state the population and conditions results apply to

Common misconceptions

What people often get wrong

Often heard: A bigger sample size, on its own, makes a study generalisable.

Actually: Size reduces random error but does not fix an unrepresentative sample. A large but skewed sample generalises poorly; representativeness matters as much as size.

Often heard: Every study should aim to generalise to a broad population.

Actually: Many studies, especially qualitative ones, seek deep understanding of a specific case. There the goal is transferability through rich context, not statistical generalisation.

Often heard: Generalisability and internal validity always improve together.

Actually: They can trade off. The tight control that strengthens a causal claim can narrow the conditions a finding applies to, so broad generalisation and strong internal validity must be balanced.

Going deeper