What is construct validity and what are three common sources of evidence for it?

• A. Face validity, convergent validity, and reliability
• B. Content validity, test-retest reliability, and interrater reliability
• C. Content validity, convergent/divergent validity (criterion-related), and factor structure
• D. Convergent validity, discriminant validity, and reliability

Answer: (C)

Construct validity is about whether a test actually measures the intended theoretical construct, not just something it superficially seems to capture. To support this, you look at several kinds of evidence.

Content validity checks that the test items cover the full domain of the construct, typically through expert review and a careful mapping of the construct’s facets to the items. Convergent and discriminant validity examine how the measure relates to other variables: it should correlate with related constructs (convergent) and not correlate too strongly with unrelated ones (discriminant). Criterion-related validity can also be considered here, showing that the measure relates to external criteria in expected ways, whether assessed concurrently or predictively. Finally, the factor structure revealed by factor analysis tests whether the items cluster into the expected dimensions, supporting the measure’s internal organization and alignment with the theoretical construct.

So, the three common sources of evidence described are content validity, convergent/divergent (including criterion-related) validity evidence, and the factor structure.