Explained clearly 6 min read

ASTM E29 vs. GUM: Choosing the Right Significant Figure Convention

ASTM E29 governs significant-digit rounding for test-data conformance to specifications, while the GUM governs significant digits for reporting measurement uncertainty. Choose based on whether you are making a pass/fail decision or expressing a measured result with uncertainty.

Short Answer

ASTM E29 governs significant-digit rounding for test-data conformance to specifications, while the GUM governs significant digits for reporting measurement uncertainty. Choose based on whether you are making a pass/fail decision or expressing a measured result with uncertainty.

ASTM E29 and the GUM are two authoritative standards that govern significant figures and rounding, but they answer different questions. ASTM E29 tells you how to round test data when deciding whether a material, product, or process conforms to a specification. The GUM tells you how to evaluate and report measurement uncertainty. Choosing the wrong convention can change a pass/fail decision or misstate a measurement result. This precision-and-rounding reference explains the rules, compares the conventions, and provides worked examples for students, engineers, scientists, and educators.

Rule Statement

ASTM E29 rule: For conformance testing, round the observed test data to the number of significant digits consistent with the specification limit, then compare the rounded value to the limit. ASTM E29-22a generally uses round-half-even for exact ties: if the digit to be dropped is less than 5, leave the retained digit; if greater than 5, increase it by one; if exactly 5 with no nonzero digits following, round to the nearest even retained digit. The standard also defines a rounding interval, which is the unit in the last retained place.

GUM rule: For reporting a measurement result, express the uncertainty to at most two significant digits, then round the measured value to the same decimal place as the uncertainty. The GUM (JCGM 100:2008, 7.2.2 and 7.2.6) recommends two significant digits for uncertainty, although one or three may be justified by the application. The last retained digit of the result and the last retained digit of the uncertainty should align.

Use ASTM E29 for pass/fail decisions against specifications. Use the GUM for measurement uncertainty and result reporting.

Standards Citation

The primary documents are:

  • ASTM E29-22a, Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications. See Sections 4 through 6 for scope, significance, rounding procedure, and comparison with specification limits.
  • JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement (GUM). See Clauses 7.2.2, 7.2.6, and 8.2 for reporting and rounding recommendations.
  • ISO 80000-1:2022, Quantities and units — Part 1: General. See Clause 7.3.4 and Annex B for general rounding and significant-digit conventions.
  • NIST TN 1297, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results. See Sections 7 and 8 for practical uncertainty reporting.
  • NIST SP 811, Guide for the Use of the International System of Units (SI). See Section 7.3 for rounding and significant figures in SI usage.

These documents are mutually supporting: ISO 80000-1 and NIST SP 811 provide general rounding rules, ASTM E29 applies them to conformance, and the GUM applies them to uncertainty. For more general rules, see our rounding rules and significant figures references.

Convention Comparison Table

Dimension ASTM E29 GUM
Primary purpose Determine conformance to specifications from test data Evaluate and report measurement uncertainty
Rounding target Test data rounded to specification significant digits Uncertainty rounded to two significant digits; result to same decimal place
Tie-breaking Round half even (exact 5 with no nonzero digits) Round half even, consistent with ISO 80000-1
Application point Before comparison to specification limit After uncertainty evaluation, before final reporting
Typical output Pass/fail or conformance statement Value ± expanded uncertainty with coverage factor k
Key clause ASTM E29-22a, Sections 4-6 JCGM 100:2008, 7.2.2, 7.2.6, 8.2

Worked Examples

Example 1: ASTM E29 conformance decision

A specification requires a maximum thickness of 10.00 mm. A test yields 10.0049 mm. The specification limit has four significant digits, so the rounding interval is 0.01 mm. Look at the first digit to be dropped: 4, which is less than 5. Retain 10.00 mm. Compare 10.00 mm to 10.00 mm: the item conforms. If you compared the unrounded 10.0049 mm, you would incorrectly fail it. This is the central purpose of ASTM E29.

Example 2: GUM uncertainty reporting

A measurement gives a mean of 12.34567 V with a standard uncertainty u = 0.02345 V. First, round the uncertainty to two significant digits: 0.02345 V → 0.023 V. The last retained digit is in the thousandths place. Next, round the mean to the same decimal place: 12.34567 V → 12.346 V. If the expanded uncertainty is U = k u with k = 2, then U = 0.04690 V → 0.047 V, and the result is reported as 12.346 V ± 0.047 V. The digits align at the thousandths place.

Example 3: Using both standards in one report

Suppose a lab tests a tensile specimen. For the conformance section, it applies ASTM E29 and rounds the measured strength to the specification digits, e.g., 350.7 MPa → 351 MPa. For the measurement uncertainty section, it applies the GUM and reports 350.7 MPa ± 2.4 MPa. The two rounded values serve different purposes and should not be mixed. The ASTM value answers pass/fail; the GUM value communicates measurement quality.

Counter-Examples

  • Wrong convention for the question: Using the GUM two-significant-digit rule to round a test result before a pass/fail decision can move a value across a specification limit. ASTM E29 controls that decision.
  • Over-rounding uncertainty: Reporting u = 0.02345 V as 0.02 V loses information and is not GUM-compliant when two significant digits are warranted.
  • Double rounding: Rounding 2.445 to 2.45 and then to 2.5 introduces error. Round once, at the final reporting stage, unless a standard explicitly requires intermediate rounding.
  • Misaligned decimal places: Reporting 12.346 V ± 0.05 V is inconsistent because the uncertainty is in hundredths and the value in thousandths. Align the last digit.
  • Ignoring the spec limit digits: Rounding test data to more digits than the specification can support creates false precision and may not follow ASTM E29.

Common Mistakes

  1. Treating ASTM E29 and GUM as interchangeable. They address different metrological tasks.
  2. Rounding intermediate results. Keep extra digits in calculations; round only at the end.
  3. Using round-half-up by habit. ASTM E29 and ISO 80000-1 generally use round-half-even for exact ties.
  4. Reporting too many significant digits. The GUM recommends two for uncertainty; ASTM E29 limits digits to the specification.
  5. Confusing significant digits with uncertainty. Significant digits are a shorthand; an uncertainty budget is the rigorous basis.
  6. Forgetting the coverage factor. In GUM reporting, expanded uncertainty U requires a stated k, often k = 2 for approximately 95% coverage.
  7. Not documenting the convention. State whether ASTM E29 or GUM was used, and cite the clause or section.

Quick Reference Table

Scenario Use Rounding action Example
Pass/fail against a specification ASTM E29 Round test data to specification significant digits, then compare 10.0049 mm → 10.00 mm, pass
Report measurement uncertainty GUM Round u to two significant digits; round result to same decimal place 12.34567 V, u = 0.02345 V → 12.346 V ± 0.023 V
General scientific reporting ISO 80000-1 / GUM Use significant digits that reflect uncertainty; round half even 0.02345 → 0.023
Intermediate calculation Neither for rounding Keep extra digits; do not round Store 12.34567 V until final report
US legal metrology or test method ASTM E29 if specified Follow the test method and ASTM E29 Spec limit 10.00 mm

Continue with our precision-and-rounding reference:

Sources & Further Reading

  • ASTM E29-22a, Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications, ASTM International, West Conshohocken, PA.
  • JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement, BIPM, Sèvres.
  • ISO 80000-1:2022, Quantities and units — Part 1: General, ISO, Geneva.
  • NIST TN 1297, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results, NIST, Gaithersburg, MD.
  • NIST SP 811, Guide for the Use of the International System of Units (SI), NIST, Gaithersburg, MD.

FAQ

What is the main difference between ASTM E29 and the GUM?

ASTM E29 governs significant-digit rounding for determining conformance to specifications. The GUM governs evaluation and reporting of measurement uncertainty. They can both be used in the same report, but for different sections.

Do ASTM E29 and the GUM use different tie-breaking rules?

Current ASTM E29 and ISO 80000-1, which the GUM references, generally use round-half-even for exact ties. The more important difference is where and why rounding is applied.

How many significant digits should an uncertainty have?

The GUM recommends two significant digits for uncertainty. One significant digit may be acceptable when the uncertainty is large or the application is conservative; three may be used when needed to avoid loss of information, but two is the default.

Can I use ASTM E29 to report uncertainty?

No. ASTM E29 is not an uncertainty-reporting standard. Use the GUM and related NIST guidance for uncertainty statements.

Is ASTM E29 still current?

Yes. ASTM E29-22a is the current standard practice. Always check the latest revision and the specific test method, which may impose additional rounding requirements.

Verified sources

References

  1. ASTM E29-22a, Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications, ASTM International, West Conshohocken, PA.
  2. JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement, BIPM, Sèvres.
  3. ISO 80000-1:2022, Quantities and units — Part 1: General, ISO, Geneva.
  4. NIST TN 1297, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results, NIST, Gaithersburg, MD.
  5. NIST SP 811, Guide for the Use of the International System of Units (SI), NIST, Gaithersburg, MD.

Leave a Reply

Your email address will not be published. Required fields are marked *