What Is Measurement Uncertainty in Calibration?

Measurement uncertainty in calibration is a quantified range that describes how much a measurement result could reasonably differ from the true value. Every calibrated instrument carries some degree of doubt in its readings — and measurement uncertainty captures exactly how much. It is expressed as a ± value, such as ±0.05°C or ±0.2%, and appears on a calibration certificate alongside the measurement results. Without it, you cannot determine whether your instrument’s readings are reliable enough for your process, product specifications, or regulatory requirements.

This is not about instruments being broken or inaccurate. It is about honesty in measurement science. A well-calibrated instrument with a clearly stated uncertainty value is far more useful than one that only says “passed.”

Why Measurement Uncertainty Matters in Quality Systems

Quality standards don’t just ask whether an instrument passed calibration — they ask how reliably it passed. ISO 9001:2015 requires that measurement equipment be calibrated in a way that provides confidence in the results. ISO/IEC 17025:2017, which governs accredited calibration laboratories, goes further: it mandates that uncertainty be evaluated, documented, and reported on every calibration certificate issued.

In practical terms, this matters when you are making decisions based on measurements. If you are testing a product to a tolerance of ±1.0 mm and your measurement instrument has an uncertainty of ±0.8 mm, you are operating with almost no margin. That context would be invisible without an uncertainty value.

Calibration laboratories that meet ISO/IEC 17025 accreditation requirements are independently audited to confirm their uncertainty calculations are valid. If you are working with a lab that doesn’t report uncertainty — or only reports it as a footnote — that is a sign worth investigating before your next audit. You can learn more about what separates accredited from non-accredited certificates in our guide to calibration certificates and their role in quality programs.

The Two Types of Measurement Uncertainty: Type A and Type B

Calibration labs evaluate uncertainty from two directions, and both contribute to the final number reported on your certificate.

Type A Uncertainty is calculated statistically from repeated measurements. If a technician measures the same reference point ten times, the spread in those results becomes a statistical input into the uncertainty budget. The more consistent the readings, the smaller this contribution.

Type B Uncertainty comes from everything else — manufacturer specifications for the reference standard, temperature effects, resolution of the measuring equipment, drift since the last calibration, and any other known influence that isn’t captured by repeated measurement. These contributions are evaluated using engineering judgment, published data, and calibration history.

Both types are combined using a mathematical process called quadrature summation — the square root of the sum of squares — which produces a combined standard uncertainty. From there, a coverage factor is applied to give the final reported number.

How Measurement Uncertainty Is Expressed on a Calibration Certificate

When you review a calibration certificate, the uncertainty value is typically found near the measurement results section. It is expressed as an expanded uncertainty — the range within which the true value is expected to lie with a stated level of confidence, most commonly 95%.

You will typically see it written in one of these forms:

• U = ±0.05°C (k=2, 95% confidence)
• Expanded Uncertainty: ±0.2% of reading (k=2)

The “k” value is the coverage factor. A coverage factor of 2 corresponds to approximately 95% confidence for a normal distribution. Some labs use k=3 for 99.7% confidence in critical applications.

Here is a quick reference for interpreting what you see on a certificate:

Uncertainty vs. Accuracy vs. Error: Clearing Up the Confusion

These three terms are related but not interchangeable. Getting them mixed up can lead to poor decisions about whether equipment is fit for its intended purpose.

Accuracy describes how close an instrument’s reading is to the true value under ideal conditions. It is usually given as a specification by the manufacturer — for example, “±0.1% full scale.”

Error is the difference between a specific measurement result and the accepted reference value. If your pressure gauge reads 101.3 kPa when the true pressure is 100.0 kPa, the error is +1.3 kPa.

Uncertainty is different from both. It is not a statement about one specific reading — it is a statement about the reliability of the entire measurement process. A well-characterized measurement process with low uncertainty means you can trust your results consistently, not just on a single day.

Think of it this way: error tells you where one reading fell. Uncertainty tells you how much you should trust any reading made by that system.

What a Good Uncertainty Statement Looks Like — and What to Watch For

Not all calibration certificates handle uncertainty the same way. Here is what to look for when reviewing documentation from your calibration provider.

A well-prepared certificate will include a clearly stated expanded uncertainty value for each parameter measured, the coverage factor (k) used, the confidence level (typically 95%), and a reference to the uncertainty evaluation method — usually the GUM (Guide to the Expression of Uncertainty in Measurement, published by BIPM).

Watch out for certificates that report only “within specification” or “passed” without any numerical uncertainty. These documents may satisfy a checkbox in your records system, but they provide no usable confidence information for decision-making. In regulated industries — aerospace, medical devices, pharmaceuticals, defense — auditors routinely flag these as non-conformances.

If your current calibration provider doesn’t report uncertainty, it is worth asking whether they are ISO/IEC 17025 accredited. Accredited labs are held to this standard as a condition of their accreditation. You can find more about what accreditation actually requires — and how it protects your measurement data — in our overview of quality program requirements.

How Calibration Laboratories Calculate and Control Their Measurement Uncertainty

Accredited calibration laboratories maintain what is called an uncertainty budget — a documented breakdown of every input that contributes to the final uncertainty value for a specific measurement type. This budget is reviewed by the accreditation body as part of the ISO/IEC 17025 assessment process.

Keeping uncertainty values low requires ongoing investment: maintaining reference standards that are traceable to national standards (such as NIST in the United States), controlling the laboratory environment, using equipment with inherently low uncertainty, and conducting regular inter-laboratory comparisons to validate results.

The same principles apply across all calibrated instrument types. Whether you are looking at flow meter calibration or dimensional measurement, the uncertainty value on your certificate is only as reliable as the processes that produced it. When you receive a certificate from an accredited lab, you are receiving the output of a system that has been independently verified — not just a document generated on request.

If you are building or reviewing your organization’s measurement assurance program, treating measurement uncertainty as a core input — not an afterthought — is what separates facilities that pass audits from those that scramble to explain gaps. Our instrument calibration services include full uncertainty reporting as standard on every certificate, in line with ISO/IEC 17025 requirements.