Which term refers to the consistency of scores across multiple administrations or items in an assessment?

Which term refers to the consistency of scores across multiple administrations or items in an assessment?

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

Answer: (C)

Reliability is about consistency in measurement—how stable and repeatable an assessment’s scores are across different occasions or across the items that make up the test. When a test is reliable, you’d expect a student to receive similar scores if they take the same or a very similar version again, assuming no real change in ability and that conditions are roughly the same. This consistency can come from different sources, such as test-retest reliability (similar scores on separate administrations) or internal consistency (the items within the test all measuring the same underlying skill in a coherent way). If we contrast this with validity, that’s about whether the test actually measures the intended skill or trait. Bias refers to systematic error that can skew results for some groups, not the overall consistency of scores. Relevance concerns how appropriate or meaningful the items are for the construct or context, not the stability of scores over time or across items. So the term that best fits the idea of consistent scores across different times or items is reliability.

Reliability is about consistency in measurement—how stable and repeatable an assessment’s scores are across different occasions or across the items that make up the test. When a test is reliable, you’d expect a student to receive similar scores if they take the same or a very similar version again, assuming no real change in ability and that conditions are roughly the same. This consistency can come from different sources, such as test-retest reliability (similar scores on separate administrations) or internal consistency (the items within the test all measuring the same underlying skill in a coherent way).

If we contrast this with validity, that’s about whether the test actually measures the intended skill or trait. Bias refers to systematic error that can skew results for some groups, not the overall consistency of scores. Relevance concerns how appropriate or meaningful the items are for the construct or context, not the stability of scores over time or across items. So the term that best fits the idea of consistent scores across different times or items is reliability.