Which term indicates how well an assessment measures what it is intended to measure; a test is not valid if it is not reliable?

Which term indicates how well an assessment measures what it is intended to measure; a test is not valid if it is not reliable?

• A. Validity
• B. Reliability
• C. Bias
• D. Fairness

Answer: (A)

Validity is the degree to which an assessment measures what it is intended to measure. In ESOL testing, that means the items and the overall score should truly reflect the language skill or construct you’re aiming to assess—such as speaking fluency, listening comprehension, or vocabulary use—and align with the standards you’re targeting. When a test targets the right construct, the results meaningfully inform decisions about a student’s language abilities. It’s useful to keep in mind how reliability fits in: a test can produce consistent, stable scores (reliability) without actually measuring the intended construct (validity). So reliability is about consistency, while validity is about accuracy to the intended measurement. Bias and fairness relate to whether the test treats all test-takers equitably, but they don’t by themselves determine whether the test measures the right thing.

Validity is the degree to which an assessment measures what it is intended to measure. In ESOL testing, that means the items and the overall score should truly reflect the language skill or construct you’re aiming to assess—such as speaking fluency, listening comprehension, or vocabulary use—and align with the standards you’re targeting. When a test targets the right construct, the results meaningfully inform decisions about a student’s language abilities.

It’s useful to keep in mind how reliability fits in: a test can produce consistent, stable scores (reliability) without actually measuring the intended construct (validity). So reliability is about consistency, while validity is about accuracy to the intended measurement. Bias and fairness relate to whether the test treats all test-takers equitably, but they don’t by themselves determine whether the test measures the right thing.