If a measure yields similar results when administered again at a later time, this demonstrates which type of reliability?

If a measure yields similar results when administered again at a later time, this demonstrates which type of reliability?

• A. Test-retest reliability
• B. Inter-observer reliability
• C. Intra-observer reliability
• D. Construct validity

Answer: (A)

Reliability over time is being tested here. When the same measure is given to the same people on a different occasion and the results are similar, it shows the instrument yields stable scores across time. This is called test-retest reliability. It indicates the measure is consistent in what it records when the underlying trait hasn’t truly changed and the testing conditions are similar. Inter-observer reliability would involve different raters agreeing on scores, not how stable a single score is over time. Intra-observer reliability looks at how consistently the same observer scores across repeats, which can be about consistency within a single session or across occasions but focuses on the observer, not the test’s stability over time. Construct validity is about whether the test actually measures the intended construct, not about repeating results over time.

Reliability over time is being tested here. When the same measure is given to the same people on a different occasion and the results are similar, it shows the instrument yields stable scores across time. This is called test-retest reliability. It indicates the measure is consistent in what it records when the underlying trait hasn’t truly changed and the testing conditions are similar.

Inter-observer reliability would involve different raters agreeing on scores, not how stable a single score is over time. Intra-observer reliability looks at how consistently the same observer scores across repeats, which can be about consistency within a single session or across occasions but focuses on the observer, not the test’s stability over time. Construct validity is about whether the test actually measures the intended construct, not about repeating results over time.