Measurement Bias: What can happen when we use an accuracy specification and assume all the measurements are centered to the specification limits?
Not correcting for Measurement Bias is a typical problem in the metrology community, where many papers assume a centered process or Measurement and that any known Measurement Bias has been corrected.
The Morehouse technical paper on Measurement Bias examines what happens if we fail to correct Measurement Bias at the bench level and gives examples based using Specific Risk. Specific Risk is the probability that an accepted item is non-conforming or that a rejected item does conform. This risk is based on measurements of a single item.
In general terms, Specific Risk is dependent on a single probability function and can be referred to as Probability of Conformance from the customer’s point of view.
Not correcting for Measurement Bias matters because when a known bias is ignored, meaning not corrected or not included in the Statement of Measurement Uncertainty on the Calibration Certificate, measurement traceability may not be fully achieved, and all subsequent measurements are suspect.
If you enjoyed this content, check out our LinkedIn and YouTube channel for more helpful posts and videos.
