The normalized error used to score calibration proficiency tests and comparisons: the lab's value minus the reference value, divided by the square root of the sum of the squares of the two expanded uncertainties. An absolute value of 1 or less is satisfactory; above 1 the difference cannot be explained by the stated uncertainties. A very small En on every round can mean the lab is overstating its uncertainty.
En number on the 10 kg was 0.4, fine, but on the 1 g it's 1.6, we've underestimated something.
An En of 1.6 means either the value is off or the uncertainty is too small; inflating the budget only fixes it if the budget really was missing something.
Also heard as
- En
- E sub n
- normalized error
Related terms
Proficiency testing
Judging a lab's performance by having it measure circulated items and scoring its results against a reference value.
Round robin
An interlaboratory comparison in which one artifact or sample travels from lab to lab to be measured by each.
Expanded uncertainty
Combined standard uncertainty multiplied by a coverage factor to give an interval with a stated coverage probability.