What does predictive validity refer to within criterion-related validity?

What does predictive validity refer to within criterion-related validity?

• A. It examines whether test scores predict future outcomes.
• B. It examines whether test scores correlate with others at the same time.
• C. It measures internal consistency.
• D. It assesses test content coverage.

Answer: (A)

Predictive validity is about the ability of a test to forecast an outcome that will occur in the future. When a test scores correlate with a later criterion, like how well college entrance scores forecast college GPA or how a job aptitude test predicts future job performance, that relationship supports predictive validity. The idea is that the test provides information that helps predict something that matters down the line, using data collected after the test to see how well predictions hold up. Why the other ideas don’t fit here: correlating test scores with other measures taken at the same time points to concurrent validity, not prediction of future outcomes. Internal consistency is about reliability—whether the test items hang together consistently—rather than whether the test can predict something later. Assessing content coverage speaks to content validity, ensuring the test adequately represents the domain, not its predictive power.

Predictive validity is about the ability of a test to forecast an outcome that will occur in the future. When a test scores correlate with a later criterion, like how well college entrance scores forecast college GPA or how a job aptitude test predicts future job performance, that relationship supports predictive validity. The idea is that the test provides information that helps predict something that matters down the line, using data collected after the test to see how well predictions hold up.

Why the other ideas don’t fit here: correlating test scores with other measures taken at the same time points to concurrent validity, not prediction of future outcomes. Internal consistency is about reliability—whether the test items hang together consistently—rather than whether the test can predict something later. Assessing content coverage speaks to content validity, ensuring the test adequately represents the domain, not its predictive power.