Every measurement has limitations. Even when an instrument is properly calibrated and an experiment is carefully performed, the measured result cannot generally be known with unlimited certainty. Measurement uncertainty provides a scientifically meaningful way to express those limitations. It is not simply a mistake or an indication that an experiment was performed incorrectly. Rather, it is an essential characteristic of a measurement result. The measurand is the quantity intended to be measured, such as the length of a metal rod, the mass of a sample or the temperature of a liquid. A measurement result is an estimate of that quantity, normally accompanied by an uncertainty. Uncertainty may arise from instrument calibration, resolution, environmental variation, measurement repeatability, reference standards, mathematical models and other influences. The uncertainty evaluation should reflect the actual measurement process rather than rely on an arbitrary percentage. Uncertainty is commonly expressed as standard uncertainty, which has the form of a standard deviation, and expanded uncertainty, which provides a wider interval based on a coverage factor. A measurement result should be interpreted together with its uncertainty, measurement unit, method and relevant experimental conditions. For example, a rod measured as (25.40 ± 0.05) cm communicates more information than simply reporting 25.40 cm. The uncertainty indicates the estimated spread associated with the result under the stated evaluation model. It does not guarantee that the unknown true value lies inside the interval.
Introduction to Measurement Uncertainty
Measurement uncertainty is a quantitative indication of the doubt associated with a measurement result. It describes the range of values that can reasonably be attributed to the measurand, considering the available measurement information, instrument characteristics, experimental conditions and statistical evidence.
For a measurement result x with standard uncertainty u(x), the result may be represented as x ± u(x). For expanded uncertainty, the representation is x ± U, where U = k u_c(x).
Explanation
Variables
| Symbol | Meaning | Si Unit |
|---|---|---|
x | Measurement result or estimate of the measurand | Unit of the measured quantity |
u(x) | Standard uncertainty associated with the measurement result | Same unit as x |
U | Expanded measurement uncertainty | Same unit as x |
k | Coverage factor used to obtain expanded uncertainty | Dimensionless |
Example or application
A worked example has not yet been added.
Definition
Measurement uncertainty is a non-negative parameter associated with a measurement result that characterizes the dispersion of values that could reasonably be attributed to the measurand, based on the information used in the evaluation.
Purpose
To communicate the reliability and limitations of measurement results, support comparisons between measurements, and provide a quantitative foundation for scientific and engineering decisions.
Importance
Measurement uncertainty is essential for evaluating experimental quality, comparing results from different laboratories, validating scientific models, calibrating instruments, establishing tolerances and making decisions in engineering, manufacturing, medicine and research.
Measurand And Measurement Model
Measurand
The quantity intended to be measured must be clearly defined, including relevant conditions. For example, the length of a metal rod may depend on its temperature.
Measurement Model
A mathematical relationship connecting the measurand to measured input quantities, correction factors and other relevant variables.
Measurement Result
An estimate obtained by applying the measurement model to the available measurement information.
Uncertainty Evaluation
An assessment of the uncertainty contributions associated with the input quantities and the measurement model.
Uncertainty And Error
Measurement Error
The difference between a measured value and a reference value.
Measurement Uncertainty
A parameter describing the dispersion of values reasonably attributable to the measurand.
Key Difference
Error is a deviation from a reference, whereas uncertainty describes the doubt associated with a measurement result. They are related but are not interchangeable.
Important Note
A measurement may have a small estimated uncertainty and still contain an unknown systematic bias if an important influence has not been identified.
Main Characteristics
Uncertainty is non-negative.
It is normally expressed in the same unit as the measurand.
It depends on the measurement method, model, conditions and available information.
It can be evaluated using statistical or non-statistical information.
It does not necessarily represent the actual error.
It should be evaluated and reported with sufficient detail to support interpretation.
Standard And Expanded Uncertainty
Standard Uncertainty
Uncertainty expressed as a standard deviation. It is commonly represented by u or u_c.
Expanded Uncertainty
An uncertainty obtained by multiplying combined standard uncertainty by a coverage factor. It is commonly represented by U.
Interpretation
Expanded uncertainty provides an interval intended to cover a substantial portion of the distribution of values attributable to the measurand. Its interpretation depends on the probability distribution, degrees of freedom and selected coverage factor.
Worked Example
Problem
A metal rod is measured as 25.40 cm. The estimated standard uncertainty is 0.05 cm. Express the result.
Calculation
x = 25.40 cm; u(x) = 0.05 cm
Result
25.40 cm ± 0.05 cm
Interpretation
The measurement estimate is 25.40 cm, with a standard uncertainty of 0.05 cm. The uncertainty describes the estimated dispersion, not a guaranteed maximum error.
Practical Evaluation Steps
Define the measurand and measurement conditions.
Develop an appropriate measurement model.
Identify all significant uncertainty sources.
Evaluate each uncertainty contribution using available information.
Convert contributions into standard uncertainty where necessary.
Combine relevant contributions using an appropriate mathematical model.
Report the result, uncertainty, unit and evaluation information.
Common Misconceptions
Uncertainty is not the same as measurement error.
A larger number of decimal places does not automatically mean lower uncertainty.
A calibrated instrument does not have zero uncertainty.
Repeated measurements cannot automatically reveal every systematic influence.
Expanded uncertainty is not necessarily a guaranteed error limit.
Applications
Scientific laboratory measurements.
Instrument calibration and testing.
Quality control and manufacturing.
Engineering design and safety margins.
Environmental monitoring.
Medical and pharmaceutical measurements.
Comparisons between research results.
Reference
JCGM 100:2008, Evaluation of Measurement Data — Guide to the Expression of Uncertainty in Measurement (GUM).
JCGM 200, International Vocabulary of Metrology (VIM).
NIST Technical Note 1297, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results.
Related Topic
SCI-REF-003-073
SCI-REF-003-082
SCI-REF-003-086
SCI-REF-003-088
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.
Suggest a correction or source · More in Units & Measurement