What is the validity of a selection test
Validity is the degree to which a test actually measures what it claims to measure and lets you draw useful conclusions for the decision it supports.
Validity is the degree to which a test actually measures what it claims to measure and allows you to draw useful conclusions for the decision it’s going to support. It’s not a “yes or no” property, but a matter of degree: a test is more or less valid for a specific use. Together with reliability, it’s one of the two properties that distinguish a serious assessment from just any questionnaire.
Validity is not the same as reliability
It’s the most common confusion, so it’s worth settling up front. Reliability is consistency: that the test gives stable results. Validity is relevance: that it measures the right thing and that its results serve what they’ll be used for. A test can be perfectly consistent and still measure something irrelevant to the role.
The three types usually cited
| Type of validity | What it assesses |
|---|---|
| Content | Whether the items adequately cover what you want to measure |
| Criterion | Whether the results relate to a relevant external behavior or performance |
| Construct | Whether the test measures the abstract concept it claims to measure, and not a different one |
In selection, the important nuance is that validity depends on use: the same test can be relevant for one role and of little use for another. That’s why choosing well what to measure based on the position is part of using valid tests. We cover it in how to choose competencies by role.
An honest note on validity figures
There’s academic debate about how much validity different types of tests add, and the estimates have been corrected over time. That’s why it’s worth handling carefully any validity number presented as definitive: a coefficient with no source, no context, and no role it applies to says little. Validity is argued with traceable evidence, not proclaimed.
See how we describe what each test measures, without inventing figures.
See the science behind itIn short
Validity is the degree to which a test measures what it claims to measure and serves the decision it’s going to support. It’s distinct from reliability —relevance versus consistency— and it isn’t a fixed label: it depends on the role and the use. It’s assessed through content, criterion, and construct, and it’s backed by evidence, not loose claims. At Kokoro we describe what each test in the library measures honestly, without attributing validity coefficients with no source; you can review the approach in the science behind it.