Dictionary · Statistics
Normalized error (En)
The interlaboratory comparison statistic En = (x − X) / √(U²(x) + U²(X)). |En| ≤ 1 indicates agreement within expanded uncertainties; it is the standard scoring for calibration proficiency tests.
Related terms
More from Statistics
Measurement trueness
Closeness of agreement between the average of infinitely many replicate measured values and a reference value. Trueness relates inversely to systematic error, not random error.
VIM (JCGM 200:2012) 2.14
Repeatability condition of measurement
A condition set that includes the same procedure, operators, measuring system, operating conditions, and location, with replicates over a short period.
VIM (JCGM 200:2012) 2.20
Measurement repeatability
Measurement precision under repeatability conditions — the tightest, most flattering precision an instrument and operator can show.
VIM (JCGM 200:2012) 2.21
Z-score
A proficiency-testing statistic z = (x − X)/σ_pt comparing a participant result with the assigned value in units of the standard deviation for proficiency assessment; |z| ≤ 2 is conventionally satisfactory.
ISO 13528:2022
Metrological compatibility of measurement results
The property of a set of results for a stated measurand such that the absolute difference of any pair is smaller than a chosen multiple of the standard uncertainty of that difference — the VIM's formal test for "do these results agree".
VIM (JCGM 200:2012) 2.47
Measurement precision
Closeness of agreement between indications or measured values from replicate measurements under specified conditions — usually expressed as a standard deviation, variance, or coefficient of variation.
VIM (JCGM 200:2012) 2.15