Test-retest reliability is typically reported on a 0 to 1.0 scale. What does a score of 1.0 indicate?

Test-retest reliability is typically reported on a 0 to 1.0 scale. What does a score of 1.0 indicate?

• A. The person gets the same score every time
• B. The test is perfectly valid
• C. The test measures only one construct
• D. There is no measurement error

Answer: (C)

Test-retest reliability looks at whether scores stay the same across two administrations. A score of 1.0 means perfect stability: each person's score is identical on both occasions, showing no random fluctuation in observed scores over time. That’s why the best description is that the person gets the same score every time. Validity and unidimensionality are different concerns—validity asks if the test measures what it’s supposed to, and unidimensionality asks if it measures a single construct. The idea of "no measurement error" is related but the key point here is perfect stability of scores across time.

Test-retest reliability looks at whether scores stay the same across two administrations. A score of 1.0 means perfect stability: each person's score is identical on both occasions, showing no random fluctuation in observed scores over time. That’s why the best description is that the person gets the same score every time. Validity and unidimensionality are different concerns—validity asks if the test measures what it’s supposed to, and unidimensionality asks if it measures a single construct. The idea of "no measurement error" is related but the key point here is perfect stability of scores across time.