Construct validity refers to:

Construct validity refers to:

• A. How well the test measures a theoretical construct and relates to related concepts.
• B. The stability of results over time.
• C. The consistency of ratings by different judges.
• D. The breadth of sample questions included in the test.

Answer: (A)

Construct validity asks whether a test truly measures the intended theoretical construct and whether its relationships with other measures align with theory. In practice, this means the test should show appropriate connections to related concepts (convergent validity) and weaker or no connections to unrelated ones (discriminant validity), all grounded in the underlying theory of the construct. It also involves examining how the test behaves in relation to other established measures and patterns, such as the expected factor structure and known groups that should differ on the construct. Stability over time is about reliability (test-retest), not whether the test captures the intended construct. Consistency of ratings by different judges speaks to inter-rater reliability, which is about scoring agreement rather than validity. The breadth of sample questions speaks more to content coverage or content validity, ensuring the test represents the domain, rather than to whether it measures the construct in question and relates to related concepts as theory predicts. So the best description is that construct validity concerns measuring the actual construct and the expected relationships with related concepts, supported by evidence from multiple sources like factor structure and related measures.

Construct validity asks whether a test truly measures the intended theoretical construct and whether its relationships with other measures align with theory. In practice, this means the test should show appropriate connections to related concepts (convergent validity) and weaker or no connections to unrelated ones (discriminant validity), all grounded in the underlying theory of the construct. It also involves examining how the test behaves in relation to other established measures and patterns, such as the expected factor structure and known groups that should differ on the construct.

Stability over time is about reliability (test-retest), not whether the test captures the intended construct. Consistency of ratings by different judges speaks to inter-rater reliability, which is about scoring agreement rather than validity. The breadth of sample questions speaks more to content coverage or content validity, ensuring the test represents the domain, rather than to whether it measures the construct in question and relates to related concepts as theory predicts.

So the best description is that construct validity concerns measuring the actual construct and the expected relationships with related concepts, supported by evidence from multiple sources like factor structure and related measures.