Which statement about divergent validity is true?

• A. It requires high correlation with similar constructs.
• B. It is demonstrated by low correlations with dissimilar constructs.
• C. It is demonstrated by high correlations with dissimilar constructs.
• D. It is the same as predictive validity.

Answer: (B)

Divergent validity shows that a measure is capturing a distinct construct and not something unrelated. The best evidence is that the test has low correlations with measures of dissimilar constructs. This weak relationship indicates the instrument isn’t tapping into unrelated traits, supporting its specificity. For example, a test designed to measure social anxiety should not show a strong link to a measure of athletic ability. If a measure correlated highly with dissimilar constructs, that would undermine its discriminant validity, and if it correlated highly with similar constructs, that would reflect convergent validity rather than divergent validity. Predictive validity, by contrast, involves how well the measure forecasts future outcomes, not how it relates to other constructs.