If data are transformed during processing, what must be possible?

• A. If data are transformed during processing, it should always be possible to compare the original data and observations with the processed data.
• B. Transformed data should replace the original data, with no trace.
• C. Only processed data should be retained.
• D. Original data must be discarded after processing.

Answer: (A)

Maintaining data integrity and traceability is essential when data are transformed during processing. If you change or derive values, you must still be able to compare the original observations with the processed results to verify that the transformations were applied correctly and to support audits and regulatory review. This means keeping the original data accessible and recording exactly how, why, and under what rules the transformation was performed, with a clear link between the original observation and the transformed value.

In practice, you’d document the transformation in an audit trail or data management plan, show the mapping from original to transformed values, and keep the original measurements so that they can be re-checked or re-derived if needed. For example, converting units, calculating derived variables, or adjusting dates should allow you to trace back to the source values and confirm that the derived data match the original data when applied correctly.

If you replaced or discarded the original data, or retained only the processed data, you would lose the ability to verify the accuracy of the transformation and to perform audits or resolution of discrepancies.