Reading the Record

Begin with the complete document

A screenshot of a pass mark strips away the context needed to judge it. Request the complete report, including appendices, amendments and any referenced limitations. Check the report identifier, issuing laboratory, page sequence and issue information. If an amendment exists, establish which version supersedes which. A genuine extract can still be misleading when it omits an essential qualification.

Confirm authenticity where the decision warrants it. Contact the laboratory using details obtained independently, not only the phone number printed on a suspicious document. Laboratories may need their customer’s permission before discussing confidential work. That restriction is not itself evidence of a problem, but it may require the supplier to authorise verification or provide a suitable confirmation route.

Match three descriptions

Compare the purchase specification, the laboratory’s sample description and the equipment being supplied. Look for a shared material reference and a clear link between batch, model and revision. Colour, thickness or a supplier code may distinguish variants with different behaviour. A report for a predecessor material should not be silently reused for a changed construction.

Also distinguish information supplied by the customer from information established by the laboratory. A customer-provided composition label is not necessarily a chemical identification result. Likewise, a sample’s stated manufacturing date may be a declaration rather than an independently verified fact. Read the report’s attribution carefully before turning a description into a claim of laboratory verification.

Methods and conditions frame the number

Locate the method designation and edition, specimen preparation, conditioning and relevant departures. A modified method can be appropriate for a particular engineering question, but the modification needs to remain visible. Ask whether it affects comparability or the claimed conformity. Two results with the same property name can represent substantially different procedures.

Read units and reported statistics with equal care. A force is not the same as a stress; an individual specimen value is not a batch average. Check the specimen count and any reported spread. If the report contains only a mean, ask whether individual results are available and whether the decision criterion applies to the mean, each specimen or another statistic.

Uncertainty is not a manufacturing tolerance

Measurement uncertainty describes the dispersion of values that could reasonably be attributed to the measured quantity under the evaluation used. It is not simply operator error, and it is not the acceptable variation specified for a product. Those concepts can interact in a decision, but they should not be collapsed into one margin.

Not every report presents uncertainty in the same way. Its relevance depends on the method, result and reporting requirements. Where a result is close to a specification limit, ask how uncertainty was considered in the statement of conformity. Avoid doing an improvised subtraction or addition and presenting it as the laboratory’s conclusion.

Find the decision rule

A decision rule describes how measurement uncertainty is accounted for when stating conformity with a specified requirement. Depending on the applicable specification and agreement, the rule may use a guard band or another defined approach. The important reading task is to identify the actual rule used and the requirement to which it was applied.

A numerical result and a conformity statement therefore carry different information. If the report says “pass,” locate the acceptance limit and its source. If it reports only values, do not add a pass label without a justified criterion. Internal purchasing requirements can be legitimate, but they are not automatically a recognised product certification scheme.

Keep the exclusions attached

Read statements limiting results to the items tested, notes about sampling responsibility and identification of work performed externally. Check whether all reported activities were within the claimed accreditation scope. Opinions, interpretations or customer-supplied information may need separate attention. None of these features automatically invalidates a report; each defines how to use it responsibly.

Finally, record your own conclusion in bounded language. “This report supports the stated compression requirement for the identified specimens” is more defensible than “this equipment is certified safe.” List unresolved questions and assign an owner to each. A clear evidence gap is actionable; an inflated claim can hide the very issue the report was meant to clarify.

Source notes

1. ISO: ISO/IEC 17025 addresses laboratory reporting and statements of conformity within its requirements.
2. ILAC guidance series: ILAC-G8 discusses decision rules and statements of conformity; ILAC-G17 addresses measurement uncertainty in testing. Check the currently published guidance.
3. BIPM / JCGM publications: authoritative resources on measurement uncertainty and metrology vocabulary.

Prepare an evidence question →