Which scenario illustrates observer drift?

Which scenario illustrates observer drift?

• A. The observer uses fixed criteria consistently across sessions
• B. The observer uses new measurement tools but maintains thresholds
• C. The observer ignores data tables but still records observations
• D. A psychologist unintentionally changes the standards being used to code a behavior over time

Answer: (D)

Observer drift happens when the standards used to code or measure a behavior shift over time, so the data reflect the coder’s changing criteria rather than real changes in the behavior. In the scenario described, the psychologist unintentionally alters the standards used to code a behavior across sessions. That means what counts as a target behavior becomes looser or stricter over time, producing systematic changes in the recorded frequency or duration that aren’t due to the person’s behavior but to the coder’s evolving criteria. This is precisely what drift looks like: a gradual, unconscious recalibration of how observations are scored. Sticking with fixed criteria yields stable data, because the measurement boundaries stay the same across sessions. Using new measurement tools but keeping thresholds keeps the defined behavior the same, so comparability is largely preserved. Ignoring data tables but still recording observations introduces issues with data management or completeness, not a change in the coding standards themselves.

Observer drift happens when the standards used to code or measure a behavior shift over time, so the data reflect the coder’s changing criteria rather than real changes in the behavior. In the scenario described, the psychologist unintentionally alters the standards used to code a behavior across sessions. That means what counts as a target behavior becomes looser or stricter over time, producing systematic changes in the recorded frequency or duration that aren’t due to the person’s behavior but to the coder’s evolving criteria. This is precisely what drift looks like: a gradual, unconscious recalibration of how observations are scored.

Sticking with fixed criteria yields stable data, because the measurement boundaries stay the same across sessions. Using new measurement tools but keeping thresholds keeps the defined behavior the same, so comparability is largely preserved. Ignoring data tables but still recording observations introduces issues with data management or completeness, not a change in the coding standards themselves.