Explainer · Plain-language

What is external validity?

External validity is the degree to which a study’s findings generalise beyond the study itself — to other people, settings, times, and conditions than those originally tested.

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.

Will the finding travel?

External validity asks whether a result observed in one study would also appear elsewhere — with different participants, in a different setting, at a different time, or under everyday rather than experimental conditions. A finding from undergraduate volunteers in a laboratory may not generalise to older adults in a clinic; an effect seen this year may not hold a decade later. The dimensions of generalisation are usually described as population validity (to other people) and ecological validity (to other settings and real-world conditions). High external validity means there is good reason to believe the finding extends beyond the narrow circumstances in which it was found.

What strengthens and threatens it

External validity is helped by representative sampling, realistic and varied settings, and replication across different groups and contexts. It is threatened when the sample is unusual or self-selected, when the setting is highly artificial, or when the experimental task bears little resemblance to real behaviour. Over-reliance on convenience samples is a well-known limitation across many fields. Crucially, no single study proves generalisability; confidence grows as findings are reproduced across diverse populations and conditions. This is one reason replication and open, reusable data — supported by FAIR principles — matter for cumulative, generalisable knowledge.

The trade-off with internal validity

External validity often stands in tension with internal validity. Tightly controlling a study to secure a clean causal claim — fixed conditions, a narrow sample, an artificial task — can make it less like the messy real world, weakening generalisability. Loosening controls to study realistic conditions can let confounds back in, weakening the causal claim. Neither is simply "better"; the right balance depends on the research question. A useful sequence is to establish an effect with strong internal validity, then test how widely it generalises through replication in varied real-world settings.

Key facts

At a glance

Definition: degree findings generalise to other people, settings and times
Two facets: population validity and ecological validity
Strengthened by: representative samples, realistic settings, replication
Threatened by: convenience samples and highly artificial conditions
Question: does the finding hold beyond this study’s circumstances?
Distinct from: internal validity, which concerns the cause-effect claim

Common misconceptions

What people often get wrong

Often heard: A result proven in a rigorous lab experiment automatically applies to the real world.

Actually: Strong internal validity does not guarantee external validity. Artificial conditions or an unrepresentative sample can limit how far a tightly controlled finding generalises.

Often heard: A single large study is enough to establish that a finding generalises.

Actually: Generalisability is built through replication across different populations, settings, and times. One study, however large, only samples a limited slice of the conditions a finding might face.

Often heard: External and internal validity rise together — improving one improves the other.

Actually: They often trade off. Tight control that boosts internal validity can reduce realism and external validity, so researchers must deliberately balance the two.

Going deeper