What must accompany an expanded uncertainty for a floor resistance result?

The test report should state the result and its uncertainty together, then explain how that uncertainty was determined. JCGM 100:2008 (section 7.2.3) calls for: What was measured, the reported value and its units. The expanded uncertainty and coverage factor used. The approximate confidence level and how it was determined.

Calculating Floor Resistance Measurement Uncertainty

Q: Can Craftsman evaluate measurement uncertainty for flooring installed by another contractor?

Yes. Craftsman calculates measurement uncertainty for flooring systems installed by Craftsman and other contractors. Craftsman also provides the facility’s ESD program team with guidance on whether further testing or corrective actions may be required, based on the uncertainty evaluation, documented test conditions and applicable project requirements.

Q: Does an uncertainty evaluation establish that an ESD flooring system performs correctly?

No. It helps the facility’s ESD program team determine whether the reported resistance results are precise enough for the intended evaluation. It does not, by itself, confirm that the testing was performed correctly or that the complete footwear–flooring ESD system meets its requirements.

Q: Is a meter accuracy specification enough to calculate uncertainty for a floor resistance result?

No. A meter specification may contribute to the evaluation, but it does not account for every relevant influence. The calculation must also consider the defined result, test conditions and other applicable sources of variation, such as electrode contact and repeated-reading variation.

Q: What does the repeated-reading calculation tell the facility’s ESD program team?

The calculation described in JCGM 100:2008 (Guide to the Expression of Uncertainty in Measurement, sections 4.2.1–4.2.3) estimates the Type A standard uncertainty of an average based on at least two independent readings taken under the same measurement conditions. It describes the precision of that average, not an individual reading or variation across different floor locations.

An uncertainty evaluation helps a facility’s ESD program team judge how much confidence to place in the resistance results reported during ESD system testing. It considers whether repeated readings provide independent information, whether shared conditions such as humidity influence them, and whether the measurements are representative of the floor area being evaluated.

For Craftsman Concrete, the practical question is whether the test results adequately represent the flooring system under the conditions evaluated. Craftsman calculates measurement uncertainty for both flooring systems installed by Craftsman and those installed by other contractors, and provides the facility’s ESD program team with guidance on whether further testing or corrective actions may be required.

A large uncertainty relative to the project’s acceptance criteria, or evidence that shared conditions influence multiple readings, may warrant additional test locations, repeated measurements or checks of the equipment and procedure. Any follow-up testing should remain consistent with the current specified testing methods. A more precise resistance result does not, by itself, validate the existing test results or establish the performance of the complete footwear–flooring ESD system.

A floor resistance measurement’s uncertainty cannot be calculated from the meter’s accuracy specification alone. It also depends on what the reported result represents, how the test was performed and the relevant sources of variation. JCGM 100:2008 (Guide to the Expression of Uncertainty in Measurement) provides the framework for evaluating and combining those contributions. There is no universal uncertainty percentage for ESD floor tests.

What information is needed?

An uncertainty evaluation needs:

The defined result and procedure: whether the result represents one reading, a mean of repeated readings or measurements from several floor locations, and how those measurements were made.
Sources of variability: information about relevant influences, such as sampling, environmental conditions, instrument resolution and electrode contact.
Repeated observations: the individual readings and their number, where repeated readings are used.
Calibration and testing records: relevant calibration certificates, manufacturer specifications, installation closeout documentation and previous testing records. Use records that apply to the floor, equipment and test conditions being evaluated. Record any stated uncertainty and coverage factor, and document the basis for any upper and lower limits used.
The measurement model: how each input affects the result, including sensitivity coefficients and significant correlations.
Details to include in the report: if reporting expanded uncertainty, state the coverage factor used, why it was chosen and the approximate level of confidence associated with the reported interval.

Define the result the uncertainty describes

Under JCGM 100:2008 (sections 3.1.1 and 3.1.2), uncertainty describes a specific measured quantity, called the measurand. The person calculating the uncertainty must first establish what the reported result represents and identify the test method and procedure used to obtain it.

For floor testing, a reading at one location, an average of repeated readings at that location and a result intended to represent several floor locations require different considerations. Defining that scope determines which records, sources of variation and calculations belong in the evaluation. The uncertainty must be reported alongside the existing measurement results so the facility’s ESD program team can evaluate those results in context.

Information on what can vary in the measurement

Sources of uncertainty to consider

Common sources of measurement uncertainty identified in JCGM 100:2008 (section 3.3.2) include:

Nonrepresentative sampling.
Uncertainty about environmental conditions and how they affect the measurement.
Instrument resolution is finite. The smallest change an instrument can distinguish limits the detail available in its readings.
Differences between repeated readings even when test conditions appear unchanged.

These sources are not necessarily independent. The evaluation needs information about the possible variability of each input relevant to the defined result.

Influences reported for floor resistance tests

Influence	Source	What the source states
Ambient relative humidity	ASTM F150-25, section 8.3.1	Areas with lower ambient relative humidity could have resistance readings that vary from readings at higher ambient relative humidity.
Test voltage for materials near 1.0 × 10⁶ Ω	ANSI/ESD STM7.1-2020, foreword	A flooring material can measure ≥ 1.0 × 10⁶ Ω at 10 V but < 1.0 × 10⁶ Ω at 100 V. The concern applies to materials near that boundary.
Warping of electrode contact surfaces	ANSI/ESD STM97.1-2025, section 5.3 note	Conductive rubber electrode contacts can warp over time and change measurements. ANSI/ESD STM97.1-2025 identifies no standardized method for checking whether this warping has occurred.

Follow the test method’s voltage requirements and resolve equipment problems before evaluating uncertainty. These standards do not assign a numerical uncertainty to the effects listed above, but identify factors to consider when evaluating the measurement.

Type A: standard uncertainty from repeated readings

When an input is estimated as the arithmetic mean of at least two independent repeated observations under the same measurement conditions, JCGM 100:2008 (sections 4.2.1–4.2.3) gives the Type A standard uncertainty of that estimate as:

u = \frac{s}{\sqrt{n}}

Here, $s$ is the experimental standard deviation of the observations. Its square, $s^{2}$ , is the sum of the squared deviations from their mean, divided by $n - 1$ .

The required data are the individual observations and their number. This calculation describes the uncertainty of the mean, not an individual reading. It also does not, by itself, account for variation between different floor locations. It helps the facility’s ESD program team assess the precision of the average resistance reported for the tested location and conditions.

Type B: standard uncertainty from other information

For an input not estimated from repeated observations, JCGM 100:2008 (section 4.3.1) evaluates standard uncertainty through scientific judgment based on available information about that input’s possible variability.

The evaluation draws on the records and information identified above, including:

Previous measurement data.
Experience with or knowledge of the relevant materials and instruments.
Manufacturer specifications.
Calibration and other certificates.
Uncertainties assigned to reference data in handbooks.

These are possible information sources, not a requirement to use every category for every floor result.

A quoted uncertainty with a stated multiplier

When a specification, calibration certificate or other source expresses uncertainty as a stated multiple of a standard deviation, JCGM 100:2008 (section 4.3.3) obtains the standard uncertainty by dividing the quoted value by that multiplier. The estimated variance is the square of that quotient.

This calculation requires both the quoted uncertainty and its stated multiplier. The quoted uncertainty must apply to the input being evaluated.

When only upper and lower bounds are known

When justified bounds $a_{-}$ and $a_{+}$ contain an input for all practical purposes, and nothing favors one value within them, JCGM 100:2008 (section 4.3.7) models it with a uniform, or rectangular, distribution.

The input estimate is the midpoint of the interval. Its estimated variance is:

u^{2} = \frac{{(a_{+} - a_{-})}^{2}}{12}

If $a$ is the half-width of the interval, so that $a_{+} - a_{-} = 2 a$ , the standard uncertainty is:

u = \frac{a}{\sqrt{3}}

The evaluation needs the bounds and a basis for treating them this way. A meter tolerance or verification limit is not automatically a justified bound or evidence of a rectangular distribution.

Combining the input uncertainties

The combined standard uncertainty depends on the measurement model, the function that relates the result $y$ to its input estimates:

y = f (x_{1}, x_{2}, \dots)

For uncorrelated inputs, JCGM 100:2008 (section 5.1.2, equation 10) gives the combined variance as:

{u_{c}}^{2} (y) = \sum_{i} {(\frac{\partial f}{\partial x_{i}})}^{2} u^{2} (x_{i})

The combined standard uncertainty $u_{c} (y)$ is the positive square root of that variance.

Each $u (x_{i})$ is a standard uncertainty obtained from a Type A or Type B evaluation. Each partial derivative is a sensitivity coefficient describing how the result responds to that input.

The calculation therefore requires the measurement model, its sensitivity coefficients and the input standard uncertainties. Equation 10 uses a first-order Taylor series approximation. When the model’s nonlinearity is significant, higher-order terms must be included.

Correlated inputs

Significant correlations between input quantities cannot be ignored (JCGM 100:2008, section 5.2.5).

Where feasible, associated covariances should be evaluated experimentally by varying the correlated inputs. They can also be evaluated from available information about correlated variability through a Type B evaluation.

Common sources of correlation between inputs include shared environmental influences such as temperature, pressure and humidity. If their interdependence is negligible, the affected inputs can be treated as uncorrelated. Otherwise, the model must account for the relationship, potentially by introducing common influences as additional independent inputs.

Reporting an expanded uncertainty

The combined standard uncertainty $u_{c} (y)$ is an estimated standard deviation characterizing the dispersion of values that could reasonably be attributed to the measurand. Expanded uncertainty $U$ defines an interval around the result.

JCGM 100:2008 (section 6.2.1, equation 18) gives:

U = k u_{c} (y)

Here, $k$ is the coverage factor. The result is expressed as:

Y = y \pm U

The value $y$ is the best estimate of the measurand. The interval from $y - U$ to $y + U$ is expected to encompass a large fraction of the distribution of values that could reasonably be attributed to it.

When reporting expanded uncertainty, JCGM 100:2008 (section 7.2.3) calls for:

A full description of the measurand.
The result expressed as $Y = y \pm U$ , including units.
Relative expanded uncertainty $\frac{U}{| y |}$ , for nonzero $y$ , when appropriate.
The coverage factor $k$ , or both $k$ and $u_{c} (y)$ .
The approximate level of confidence associated with the interval and how it was determined.
The supporting information specified in section 7.2.7, or a reference to a published document containing it.

Frequently Asked Questions

Yes. Craftsman calculates measurement uncertainty for flooring systems installed by Craftsman and other contractors. Craftsman also provides the facility’s ESD program team with guidance on whether further testing or corrective actions may be required, based on the uncertainty evaluation, documented test conditions and applicable project requirements.

No. It helps the facility’s ESD program team determine whether the reported resistance results are precise enough for the intended evaluation. It does not, by itself, confirm that the testing was performed correctly or that the complete footwear–flooring ESD system meets its requirements.

No. A meter specification may contribute to the evaluation, but it does not account for every relevant influence. The calculation must also consider the defined result, test conditions and other applicable sources of variation, such as electrode contact and repeated-reading variation.

The calculation described in JCGM 100:2008 (Guide to the Expression of Uncertainty in Measurement, sections 4.2.1–4.2.3) estimates the Type A standard uncertainty of an average based on at least two independent readings taken under the same measurement conditions. It describes the precision of that average, not an individual reading or variation across different floor locations.

The test report should state the result and its uncertainty together, then explain how that uncertainty was determined. JCGM 100:2008 (section 7.2.3) calls for:

What was measured, the reported value and its units.
The expanded uncertainty and coverage factor used.
The approximate confidence level and how it was determined.

Calculating the Uncertainty of a Floor Resistance Measurement

What information is needed?

Define the result the uncertainty describes

Information on what can vary in the measurement

Sources of uncertainty to consider

Influences reported for floor resistance tests

Type A: standard uncertainty from repeated readings

Type B: standard uncertainty from other information

A quoted uncertainty with a stated multiplier

When only upper and lower bounds are known

Combining the input uncertainties

Correlated inputs

Reporting an expanded uncertainty

Frequently Asked Questions

Get a Free Estimate