Skip to content
VTValiTracAI
Metrology

Type A vs Type B uncertainty evaluation

Type A uses statistics on repeated observations; Type B uses any other information. Both produce standard uncertainties that combine the same way.

ValiTrac AI editorialUpdated 2026-09-132 min read

The GUM classifies uncertainty components by the method used to evaluate them. A Type A evaluation uses the statistical analysis of a series of observations — typically the standard deviation of the mean of repeated readings. A Type B evaluation uses other means: calibration certificates, manufacturer specifications, published data, experience, and an assumed probability distribution.

Worked examples

  • Repeatability: ten readings with standard deviation 0.02 °C give a Type A standard uncertainty of the mean of 0.02/√10 ≈ 0.006 °C.
  • Reference certificate: expanded uncertainty 0.04 °C at k = 2 gives a Type B standard uncertainty of 0.02 °C.
  • Resolution 0.1 °C: a rectangular distribution of half-width 0.05 °C gives 0.05/√3 ≈ 0.029 °C.

The distinction stops mattering once each component is a standard uncertainty; they are then combined by the same root-sum-of-squares rule.

Frequently asked questions

Is Type A always smaller than Type B?
No. Either can dominate. Repeatability of a poorly stabilised measurement can easily exceed the reference's certificate uncertainty.
Does Type B mean 'systematic'?
No. The GUM deliberately avoids equating A/B with random/systematic; the classification is about the evaluation method only.

References

  1. [1]JCGM 100:2008 — Guide to the expression of uncertainty in measurement (GUM)

General technical guidance written against the cited sources. It is not regulatory or legal advice and does not replace the applicable standard, guideline or a qualified reviewer's judgement.

Related articles