Criterion-Related Validity

Criterion-related validity is evidence that scores from a measure are meaningfully related to an external criterion that represents the outcome or attribute the measure is intended to reflect or predict. It is one part of evaluating whether a scale or test produces interpretations that are useful for its intended purpose .¹

Two common forms are concurrent and predictive criterion validity. Concurrent validity examines the relationship between the measure and a relevant criterion assessed at roughly the same time. Predictive validity examines whether the measure is associated with a criterion assessed later, such as whether a screening score predicts a subsequent outcome. In both cases, the usefulness of the evidence depends heavily on the quality and relevance of the chosen criterion .¹

Criterion-related validity is different from reliability. Reliability concerns the consistency or precision of measurement, whereas validity concerns whether the evidence supports the intended interpretation of scores. A measure can be highly reliable yet have weak criterion-related validity if it consistently measures something that is not closely related to the relevant external criterion. It is also distinct from content validity and construct validity, which address different forms of evidence about what a measure represents. For this reason, no single correlation with an external variable is usually sufficient to establish the overall validity of a measure .¹