Explainer · Plain-language

What is internal validity?

Internal validity is the degree to which a study establishes a trustworthy cause-and-effect relationship between variables, free from the influence of confounding factors and alternative explanations.

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.

Confidence in a causal claim

Internal validity is about cause and effect inside the study itself. When researchers say a treatment "caused" an outcome, internal validity is the strength of the ground beneath that claim. It is high when the design rules out alternative explanations — when you can be confident the manipulated variable, rather than something else, produced the result. A confounding variable, which influences both the supposed cause and the effect, is the classic threat: if the treatment group also differed systematically in some other way, you cannot tell which factor mattered. High internal validity means such confounds have been anticipated and controlled.

Threats and how design counters them

Many recognised threats can undermine internal validity: confounding variables, selection differences between groups, participant dropout (attrition), maturation, history (outside events during the study), testing effects, and instrumentation changes. The most powerful general defence is the randomised controlled experiment: random assignment distributes unknown confounds evenly across groups, a control or comparison group isolates the treatment’s effect, and blinding limits expectation bias. Holding conditions constant, standardising procedures, and pre-registering the analysis plan further protect the causal inference. The more of these safeguards are in place, the harder it is to explain the result by anything other than the cause under study.

Internal versus external validity

Internal and external validity answer different questions and often pull against each other. Internal validity asks whether the cause-effect claim is sound within the study; external validity asks whether that finding generalises to other people, settings, and times. A tightly controlled laboratory experiment maximises internal validity but, by being artificial, can weaken external validity; a real-world field study may generalise better yet admit more confounds. Strong internal validity is usually the priority for a causal claim — a result that does not hold up internally cannot meaningfully generalise — but researchers must consciously balance the two.

Key facts

At a glance

Definition: degree a study supports a trustworthy cause-and-effect claim
Core threat: confounding variables and alternative explanations
Other threats: selection, attrition, maturation, history, testing effects
Best defence: randomised controlled experiment with blinding
Question: did the cause, and nothing else, produce the effect?
Distinct from: external validity, which concerns generalising the result

Common misconceptions

What people often get wrong

Often heard: A study with a large, representative sample automatically has high internal validity.

Actually: Sample size and representativeness relate to generalisability (external validity). Internal validity depends on controlling confounds, so a large study can still have poor internal validity.

Often heard: Internal and external validity are basically the same idea.

Actually: They are distinct. Internal validity is about a sound cause-effect claim within the study; external validity is about whether that finding generalises to other people, settings, and times.

Often heard: A statistically significant result proves the study has high internal validity.

Actually: Significance can arise even when a confound, not the treatment, drove the effect. Internal validity depends on design that rules out alternative explanations, not on the p-value alone.

Going deeper