Discriminant Validity

What is Discriminant Validity?

Lead consultant at Test Partnership, Ben Schwencke, explains discriminant validity.

Discriminant validity is a form of construct validity that involves highlighting that scores on an assessment are not associated with scores on different assessments that measure different psychological constructs. The objective of discriminant validation is to show that a specific assessment is measuring what it is intended to measure and is not measuring something else by showing a lack of association with other assessments that measure different construct.

For example, if candidates that score highly on a verbal reasoning test do not score highly on a measure of extraversion, this is evidence of discriminant validity, suggesting that the two assessments measure different psychological construct. However, if the assessment shows an association with extraversion scale (but you did not expect it to), this suggests the original assessment lacks discriminant validity and may be measuring something other than verbal reasoning. Discriminant validation is an essential element of psychometric test development and a significant proportion of any research and development process will involve discriminant validation.