Definition

Construct validity

The degree to which a measure actually captures the concept it claims to measure.

Most interesting constructs — anxiety, engagement, trust — cannot be observed directly; we measure proxies. Construct validity is the argument that the proxy works: it correlates with related measures (convergent), diverges from unrelated ones (discriminant), and behaves as theory predicts.

Validation is an accumulation of evidence, not a checkbox — which is why using established, validated instruments beats inventing items, and why translated or modified scales need re-validation and a fresh reliability estimate in your sample.