What is the term for the success with which a test predicts the behavior it is designed to predict, often assessed by the correlation with the criterion behavior?

• A. Face validity
• B. Construct validity
• C. Predictive validity
• D. Content validity

Answer: (C)

Predictive validity is about how well a test forecasts the behavior or outcome it is designed to predict. In practice, this means checking how strongly test scores relate to a criterion measure that represents the actual behavior of interest, usually using a correlation. A strong correlation indicates the test does a good job predicting future performance or outcomes, which is why this type of validity is often described as criterion-related validity. For example, a test intended to predict job performance is evaluated by correlating test scores with later job performance ratings; the higher the correlation, the stronger the predictive validity.

Face validity looks at whether the test appears to measure what it’s supposed to on the surface, not whether it actually predicts behavior. Construct validity asks whether the test truly measures the intended construct and behaves as that construct theory would predict. Content validity concerns whether the test adequately covers the domain it’s meant to assess. Among these considerations, predictive validity specifically addresses the test’s power to forecast future outcomes via its correlation with the criterion behavior.