Measurement Uncertainty – Complete Technical Guide

Measurement uncertainty defines the doubt that exists about a measurement result. Under ISO 17025, laboratories must estimate and report uncertainty to ensure traceability and reliability of analytical results.

Why Measurement Uncertainty Matters

A measurement without uncertainty is incomplete.

Types of Measurement Uncertainty

Type A (Statistical)

  • Based on repeated measurements
  • Standard deviation method
  • Calculated using statistical analysis
  • Represents random error

Type B (Non-Statistical)

  • Calibration certificate data
  • Manufacturer specifications
  • Instrument resolution limits
  • Environmental effects

Uncertainty Calculation Steps

  1. Identify all uncertainty sources
  2. Quantify Type A components (standard deviation)
  3. Quantify Type B components
  4. Convert to standard uncertainty
  5. Calculate combined uncertainty (root sum square)
  6. Apply coverage factor (k=2 typical)
  7. Report expanded uncertainty

Combined Uncertainty Formula

Combined standard uncertainty is calculated using root sum square method:

u_c = √(u₁² + u₂² + u₃² + ...)

Expanded uncertainty: U = k × uc

Practical Example – GC Analysis

Each component must be converted to standard uncertainty before combination.

Common Mistakes in Uncertainty Estimation

Reporting Format Example

Example Result:

Benzene concentration = 1.25 % ± 0.03 % (k = 2, 95% confidence)

Always state coverage factor and confidence level.