Skip to content

Psychometric assessment

A psychometric assessment is a standardised instrument that measures a psychological attribute, such as a behavioural pattern or an ability, in a way that can be scored, compared and checked for consistency. What separates it from a questionnaire is the evidence behind it: reliability, validity for a stated use, and fairness across groups.

Reliability is the freedom of a measure from measurement error, and it is the first thing an instrument has to demonstrate before validity can even be asked about.

Why it matters when the plan changes

The word is applied to everything from a rigorously validated instrument to a ten-question quiz, and buyers cannot tell them apart from the front page. The difference sits in evidence that most suppliers never show: whether scores are free from measurement error, what they predict for whom, and whether they treat groups equally. Reliability is that precondition, and predictive validity only holds for the specific use it was established against, not for the instrument in general. Without that evidence the label is a claim, not a property.

The tension is between what an instrument measures and what a decision needs. A well-built assessment describes a stable attribute of a person. A decision about a role needs to know whether that attribute fits what the role will demand under pressure, and the standard that governs high-stakes use also requires fairness testing across groups before that judgement is made. The instrument is necessary and not sufficient, and treating its score as the answer is the commonest misuse.

In practice

A leadership team is assessed with an instrument the supplier calls scientifically validated. Asked for the reliability figures and the population the validity study used, the supplier sends a brochure. The team's results are then used to inform a restructuring. Whether the numbers meant anything was never established, and now cannot be.

Evidence

What it cannot tell you

A psychometric assessment can confirm that its scores are consistent, but consistency alone says nothing about whether the attribute measured predicts performance in a given role. Reliability is a precondition, not proof of validity for a specific use, and a reliable score can still be applied to a decision it was never built to inform.

Questions

Evidence. A psychometric instrument demonstrates that its scores are free from measurement error, the definition of reliability given by Reliability (statistics), 2026, and that they relate to what they claim to measure, and behave fairly across groups, for a stated use. A questionnaire only asks; the label is earned by evidence, not format.

Reliability figures per scale, the validity evidence with the criterion and the population it was gathered on, fairness analysis by group, the norm group scores are compared against, and the intended use the instrument was validated for. A supplier who cannot produce these has an unvalidated instrument.

No. Validity is a property of a use, not of an instrument. Predictive validity, as defined in Predictive validity, Wikipedia (2026), is the extent to which a score predicts a later criterion, so evidence for graduate performance says nothing about executive performance, and evidence from one population does not transfer to another.

As one input. An instrument describes a relatively stable attribute; a role decision needs to know whether that attribute fits what the role will demand, especially under pressure. Reading the score against the demand is the step that turns a measurement into evidence for a decision.

Treating a score as a verdict on a person, using an instrument for a purpose it was never validated for, comparing people against an irrelevant norm group, and skipping fairness testing. Each produces a confident number with nothing behind it, and the confidence is what does the damage.