Measurement Uncertainty (MU)
Definition
A quantified parameter associated with a result that characterises the dispersion of values reasonably attributable to the measurand. Estimated by the GUM (Guide to the Expression of Uncertainty in Measurement) approach, combining Type A (statistical, from repeated measurement) and Type B (instrument calibration, environmental, reference-material) components, and reported as an expanded uncertainty U with coverage factor k = 2 for roughly 95 percent confidence.
- Standard calculation method
- GUM (JCGM 100:2008), used to combine all sources of uncertainty into a single expanded uncertainty U
- Coverage factor
- k = 2 for roughly 95 percent confidence interval
- Types of sources
- Type A (statistical, from repeated measurement) and Type B (instrument calibration, environmental, reference-material)
- Regulatory requirement
- Mandatory on all quantitative results in accredited forensic laboratories under ISO 17025:2017
Common questions
What does measurement uncertainty tell you about a test result?+
It is a number attached to your result that says how much the true value might differ from what you measured. The standard practice is to report expanded uncertainty at a coverage factor k equal to 2, corresponding to roughly a 95 percent confidence interval, so the true value most likely falls within the reported range around your measured value.
Why do forensic labs have to calculate and report measurement uncertainty?+
ISO 17025:2017 requires all accredited forensic labs to include it on quantitative results. It shows how trustworthy your measurement is. In drug purity cases, the uncertainty directly affects whether a result crosses legal quantity thresholds that trigger enhanced penalties, so getting it right is not optional.
How do you calculate measurement uncertainty in practice?+
You use the GUM standard (Guide to the Expression of Uncertainty in Measurement). You combine sources of doubt: statistical variation from repeated tests (Type A), plus effects from instrument calibration, the environment, reference materials, and method recovery (Type B). The result is expressed as expanded uncertainty U at coverage factor k equal to 2, meaning roughly 95 percent confidence.
Related terms
- Certified Reference Material (CRM)
- A sample with one or more analyte concentrations established by an accredited body and traceable to a national metrology institute. NIST SRM...
- Internal Standard (IS)
- A compound chemically near-identical to the analyte, often a deuterated isotopologue (d3-morphine, d5-EDDP, d5-alprazolam, pyrene-d10), spiked into every sample at a known...
- ISO/IEC 17025:2017
- The international standard specifying general requirements for the competence of testing and calibration laboratories. Covers management (document control, non-conforming work, audits) and...
- Accuracy and Precision
- Accuracy is closeness to the true value, measured as recovery from a spiked CRM with the 80 to 120 percent window. Precision...
- Category a Method
- An analytical method that provides the highest level of discriminating power in drug identification, generating structural or molecular fingerprint information. Category A...
- Category B Method
- An analytical method with intermediate discriminating power, measuring a characteristic but less structurally specific property. Includes UV-Vis spectroscopy, TLC, GC without MS,...
- Control Chart
- Time-series plot of a QC measurement with warning limits at ±2σ and action limits at ±3σ from validation data. Patterns such as...
- ENFSI Drugs Working Group (DWG)
- European Network of Forensic Science Institutes Drugs Working Group. Publishes best-practice manuals for drug analysis aligned with SWGDRUG tiering principles, operates an...
- ISO 17034
- The international standard specifying general requirements for the competence of reference material producers; an ISO 17034-accredited producer has demonstrated the technical competence...
- Limit of Detection (LOD)
- The lowest concentration of analyte that produces a signal reliably distinguishable from the instrument noise, conventionally three times the standard deviation of...
- Limit of Quantification (LOQ)
- The smallest concentration that can be quantified with acceptable precision and accuracy, typically corresponding to a signal-to-noise ratio of 10:1 or an...
- Linearity
- The range of analyte concentrations over which an analytical method produces instrument responses that are proportional to concentration (the calibration curve is...
Explained in these topics
- ISO 17025 and Quality Management in Forensic LaboratoriesA quantified parameter associated with a result that characterises the dispersion of values reasonably attributable to the measurand. Estimated by the GUM (Gui...
- Method Validation, Measurement Uncertainty and Proficiency TestingThe dispersion of values reasonably attributable to the measurand. Calculated through GUM (JCGM 100:2008) and reported as an expanded uncertainty U at a covera...
- Reference Standards, Certified Reference Materials and the Quantification BaselineA parameter characterising the dispersion of values that could reasonably be attributed to a measurement result, calculated by combining all sources of uncerta...
- SWGDRUG Identification Tiers and the UV / TLC / GC-MS / LC-MS/MS WorkflowA quantitative statement of the doubt associated with a measurement result. ISO 17025:2017 requires expanded MU (k=2, 95% coverage interval) for all quantitati...