Entry
Entries A-Z
Discriminant Validity
The degree to which a measure of a construct is observed not to be similar to (diverges from) the measure of another construct to which it theoretically should not be related. In other words, discriminant validity is a measure of the degree to which the measures of two constructs that theoretically should not be related are in fact not related. A correlation coefficient is typically used to estimate the degree to which any two measures are related or unrelated to each other. For example, a successful demonstration of discriminant validity would show that a measure or test of depression is not highly correlated with a measure or test of intelligence (which it theoretically should not be related to).