Which statement best defines test-retest reliability?

Which statement best defines test-retest reliability?

• A. It measures inter-rater agreement
• B. It measures reproducibility of scores across different populations
• C. It measures the consistency of scores across items within a single administration
• D. It measures the reproducibility of a score when the measure is repeated under the same conditions at the same point in time

Answer: (D)

Test-retest reliability asks whether a measure yields consistent results when the same people take the same test again under the same conditions. The idea is to show stability of the instrument over time, so that any variation in scores reflects true change in the underlying construct rather than random error or momentary factors. When the scores are highly reproducible on the second administration, we have evidence that the test reliably measures the construct across repeated use under the same conditions. Inter-rater agreement would involve different people scoring or rating the same responses and achieving similar scores, which is a different type of reliability. Consistency of scores across items within a single administration reflects internal consistency, not test-retest reliability. Reproducibility across different populations speaks to generalizability or external validity rather than stability over repeated administrations.

Test-retest reliability asks whether a measure yields consistent results when the same people take the same test again under the same conditions. The idea is to show stability of the instrument over time, so that any variation in scores reflects true change in the underlying construct rather than random error or momentary factors. When the scores are highly reproducible on the second administration, we have evidence that the test reliably measures the construct across repeated use under the same conditions.

Inter-rater agreement would involve different people scoring or rating the same responses and achieving similar scores, which is a different type of reliability. Consistency of scores across items within a single administration reflects internal consistency, not test-retest reliability. Reproducibility across different populations speaks to generalizability or external validity rather than stability over repeated administrations.