If data quality is suspect due to administration issues, what is the appropriate step?

If data quality is suspect due to administration issues, what is the appropriate step?

• A. Proceed with interpretation even if there were administration errors.
• B. Ignore suspicious data patterns.
• C. Reassess data quality, review administration, and gather additional information.
• D. Discard data that doesn't fit expectations.

Answer: (C)

When data quality is suspect because of how data were collected, the first priority is to verify and improve the data before drawing conclusions. This means rechecking how administration occurred, looking for signs of misadministration (such as inconsistent procedures, timing errors, missing responses, or scorer drift), and then gathering additional information or data as needed. By doing this, you determine whether the existing data can be trusted, whether corrections or retraining are enough, or if re-administration is necessary to obtain valid results. Jumping straight to interpretation, ignoring suspicious patterns, or discarding data without understanding the impact can lead to biased or invalid decisions.

When data quality is suspect because of how data were collected, the first priority is to verify and improve the data before drawing conclusions. This means rechecking how administration occurred, looking for signs of misadministration (such as inconsistent procedures, timing errors, missing responses, or scorer drift), and then gathering additional information or data as needed. By doing this, you determine whether the existing data can be trusted, whether corrections or retraining are enough, or if re-administration is necessary to obtain valid results. Jumping straight to interpretation, ignoring suspicious patterns, or discarding data without understanding the impact can lead to biased or invalid decisions.