Which type of reliability assesses the stability of scores across time?

• A. Inter-rater reliability
• B. Parallel-forms reliability
• C. Test-retest reliability
• D. Internal consistency

Answer: (C)

Stability of scores across time is what test-retest reliability measures. When the same group takes the same test on two separate occasions, the scores should be similar if the trait being measured is stable and the test is reliable. The connection between the two administrations is expressed as a correlation: a high correlation means the scores are consistent over time, indicating good stability. The time interval matters—too short can inflate consistency due to memory or practice, while too long can allow real changes in the trait to occur, reducing stability.

Inter-rater reliability is about consistency between different evaluators, not stability over time. Parallel-forms reliability assesses consistency between two equivalent versions of a test, often to control for memory effects, but it’s about equivalence of forms rather than stability of a single form over time. Internal consistency looks at how well items on one test measure the same construct at one point in time, not across multiple administrations.