Validity: What it means for tests

What validity means

A test can measure cleanly and reliably while still missing the actual target. Validity asks whether it measures the right thing. According to the widely used Standards for Educational and Psychological Testing, validity does not refer to the questionnaire as an object. It refers to whether the conclusions drawn from the scores are appropriate for a specific purpose. In other words, the test itself is not valid. A particular interpretation of its results is supported as valid within a particular context.

This idea is not new. Cronbach and Meehl described construct validity as early as 1955, and Messick emphasized in 1995 that validity concerns evaluating the meaning and use of test scores. For self-tests, this leads to an important limitation. Evidence supporting an instrument for one purpose does not automatically support it for every other purpose.

Different perspectives on the same question

Several types of validity are often distinguished. Content validity asks whether the items adequately cover the subject, including whether important aspects are missing or unrelated features are being measured. Construct validity asks whether the intended characteristic is actually being measured rather than something else that merely looks similar. Criterion validity compares test scores with another standard, such as a careful professional assessment. In newer standards, these are not treated as separate types but as different sources of evidence that together support an argument for validity.

A simple example

Imagine a questionnaire that is intended to measure concentration problems but mainly reflects sleep deprivation. It could reliably produce similar scores again and again while still measuring the wrong characteristic. If you then read those scores as a measure of concentration, that interpretation would have weak validity. This example is fictional and is only meant to illustrate the principle. It shows why validity does not depend on a single number but on what a score means in its specific context.

Common confusion

Most often, "validated" is equated with "diagnostically certain." That is incorrect. The fact that an instrument has been studied does not make an individual result a diagnosis. Validity is also confused with reliability . Reliability describes how consistently something is measured. Validity asks whether the right thing is being measured. A test can be reliable and still have low validity. Reliability is a prerequisite, but it does not replace validity.

What this means for self-tests on medtests.net

When we say that a test has a scientific basis, we mean that the instrument has been studied, not that its result replaces a diagnosis. The Depression Test with the PHQ-9, the Anxiety Test with the GAD-7, and the ADHD Test with the ASRS are based on established questionnaires. Their test quality applies to specific populations and purposes, usually screening. With the Dissociation Test we also point out that its values need particular care when you read them. A high value remains a sign in every case, not proof.

What the term does not mean

Validity is not a binary quality seal and not proof that an individual result is correct. A validated instrument does not provide a diagnostically certain result, and "scientifically proven" is the wrong wording here. A self-test is intended as a guide and does not replace medical, psychotherapeutic, or diagnostic evaluation.