Define reliability and provide an example related to a survey instrument.

• A. It is the extent to which a measure captures the intended construct.
• B. Reliability is the degree to which scores vary by participant preferences.
• C. It is the ability to generalize across contexts.
• D. Reliability is the consistency of a measure across items or time; e.g., a survey scale yields similar scores across two administrations (test-retest) or has high internal consistency.

Answer: (D)

Reliability is about consistency in measurement. A survey instrument is reliable if it produces stable scores under consistent conditions. For example, test-retest reliability looks at whether administering the same survey to the same people at different times yields similar results, indicating the measure is stable over time. Another facet is internal consistency, where the items on a scale that are intended to measure the same construct should yield related responses, often evaluated with a statistic like Cronbach’s alpha. It’s important to remember that a measure can be reliable yet not valid: it can be consistently wrong about what it’s trying to assess. Statements describing validity or generalizability refer to other properties, not reliability.