Using Calibration Corrections in Practice: When and How Should a Correction Be Applied

A calibration certificate may report an instrument error, a correction, or both. These values are related, but their signs are opposite. Confusing them can increase the measurement error instead of reducing it. This working paper explains how calibration corrections should be understood, applied, recorded, and included in measurement uncertainty. It also discusses when a correction may be interpolated, when it should not be used, and how correction tables are managed in software and spreadsheets. Examples from medical metrology show the practical effect of sign, range, mode, and measurement conditions. The main conclusion is simple: a correction is useful only when its definition is clear, its conditions match the measurement, and its uncertainty is included in the final result.