What is measurement invariance and what does it enable in cross-group comparisons?

• A. The measure yields identical scores across all groups.
• B. The measure's validity is independent of group membership.
• C. The measure is never affected by differential item functioning.
• D. Cross-group comparisons are valid only if measurement invariance is established.

Answer: (D)

Measurement invariance means the measurement tool works the same way in different groups—the construct has the same meaning, the same unit of measurement, and the same origin across groups. This is essential for cross-group comparisons because observed differences can then be attributed to true differences in the underlying construct rather than to the instrument itself or to how different groups respond to items.

When invariance holds, you can compare latent constructs meaningfully across groups. If you only establish the basic structure (configural invariance) but not metric or scalar invariance, you can compare relationships between constructs, but comparing means becomes problematic. If scalar invariance is established, you can compare latent means confidently. Without establishing invariance, observed score differences may reflect measurement bias or differential item functioning rather than real differences.

So, cross-group comparisons are valid only if measurement invariance is established.