Two laboratories receive material identified as the same research compound. Both have working analytical instruments, trained personnel and established testing procedures. Yet their reports do not agree. One laboratory records a higher purity result, while the other finds a different concentration or a small impurity peak that was absent from the first report.
The immediate temptation is to decide that one laboratory made a mistake. Sometimes an error is responsible. But two different results do not, by themselves, establish which result is correct or whether either laboratory performed its work improperly.
Measurements depend on the question being asked, the analytical method used, the condition of the sample and the uncertainty associated with the result. Differences that appear alarming on a certificate can sometimes be explained by how laboratories define and measure the same property.
Understanding these differences matters to research organizations that depend on consistent analytical evidence. A disagreement can delay a project, trigger unnecessary retesting or cause researchers to question materials that have not actually changed.
Why two laboratories can measure the same material and reach different conclusions
A laboratory result is not a description of every characteristic of a material. It is the outcome of a particular measurement performed under specified conditions.
Consider a research peptide examined by high-performance liquid chromatography (HPLC). One laboratory might use a chromatographic method with a particular column, mobile-phase composition, gradient and detector setting. Another might use a different analytical configuration. Those differences can change how components separate, which peaks are detected and how the resulting chromatogram is interpreted.
One method may resolve two closely related components into separate peaks. Another may not distinguish them under its operating conditions. The apparent difference in purity could therefore reflect the selectivity of the methods rather than a change in the underlying sample.
The same problem arises when laboratories are answering different questions. A chromatographic area percentage, a quantitative assay result and an identity determination are not interchangeable.
The distinction is explained in the educational resource understanding laboratory purity, identity, concentration and mass. A high relative chromatographic purity result does not independently establish molecular identity, total material quantity or concentration.
If one laboratory reports relative peak area and another reports a mass-based assay result, comparing the two percentages as though they describe the same property can create a disagreement that is largely semantic.
Before comparing results, researchers need to establish whether the laboratories measured the same property, reported it in equivalent units and used procedures suitable for the same analytical purpose.
Measurement uncertainty is not an admission that testing failed
Every quantitative measurement has limitations. Instruments have finite resolution, calibration processes introduce uncertainty, sample preparation can vary and statistical estimates depend on the observations available.
The National Institute of Standards and Technology (NIST) treats measurement uncertainty as an essential part of expressing what a numerical result means. A measured value without adequate information about its uncertainty can communicate more confidence than the evidence supports.
For illustration, suppose one laboratory reports a result of 97.8% and another reports 98.2%. These fictional values are not evidence of any actual testing dispute. Their difference is 0.4 percentage points, but that difference cannot be evaluated properly without knowing the uncertainty, methods and definitions behind both results.
A numerical difference alone cannot establish a statistically or scientifically meaningful disagreement. Researchers need to understand the uncertainty statements, the coverage factors where applicable and whether relevant sources of variation have been included.
It is also necessary to distinguish uncertainty from avoidable error. Uncertainty describes incomplete knowledge about a measurement. An incorrectly calibrated instrument, contaminated sample or transcription mistake may be an error requiring corrective investigation.
Neither concept should be used to excuse an inadequate method. A test that consistently measures the wrong property does not become appropriate simply because its repeated results are closely grouped.
This distinction becomes particularly important when analytical results determine whether a research material is suitable for its intended experimental purpose.
Sample preparation can influence the measurement before an instrument begins
Analytical testing begins well before a sample reaches a detector. The material must be identified, transferred, prepared and preserved in a condition appropriate for the planned analysis.
Small differences in the sample’s condition or preparation can influence later results. For example, exposure to moisture may matter for a hygroscopic material. Chemical degradation may become relevant when a compound is sensitive to environmental conditions. Incomplete dissolution, contamination or losses during sample handling can also affect certain analyses.
These are method-dependent risks, not proof that any particular research material is unstable. Some compounds tolerate ordinary handling conditions, while others require more restrictive controls supported by stability data.
When laboratories investigate inconsistent findings, they should review whether the samples were genuinely comparable. The fact that two containers have the same product name does not establish that they contain equivalent portions from the same lot or that both were maintained under matching conditions.
General considerations are described in research-material storage and handling considerations, including the importance of appropriate containers, environmental requirements and traceable sample records.
A useful investigation separates the condition of the original material from changes that may have occurred during transportation, storage or analytical preparation.
For a research organization, that distinction affects how a discrepancy should be resolved. Repeating an analysis without investigating the sample history may produce another number without clarifying why the original results differed.
What analytical validation tells researchers about conflicting results
In March 2024, the US Food and Drug Administration issued the final ICH Q2(R2) guidance on validation of analytical procedures. The guidance describes the principles used to demonstrate that an analytical procedure is suitable for its intended purpose.
Its concepts include accuracy, precision, specificity or selectivity, and other performance characteristics that may be relevant depending on the type of measurement.
Precision concerns the agreement among repeated measurements under defined conditions. Accuracy concerns agreement with an accepted reference value. Selectivity addresses whether the procedure can distinguish the property of interest from potential interferences.
These characteristics answer different questions. A method can give very similar results repeatedly without necessarily providing an accurate measurement. Another method may offer strong selectivity but require additional work to establish reliable quantitative performance.
The guidance also distinguishes repeatability, intermediate precision and reproducibility. Repeatability concerns measurements under closely controlled conditions. Intermediate precision examines variation within a laboratory, such as differences involving days, analysts or equipment. Reproducibility addresses variation across laboratories.
This distinction helps explain why an internal quality-control record cannot automatically predict agreement between independent laboratories.
A laboratory may show excellent repeatability for its own procedure while another laboratory, using different instruments or operating conditions, reports results that require further comparison.
ICH Q2(R2) was developed primarily for analytical procedures supporting pharmaceutical quality and regulatory submissions. Its principles are useful for understanding measurement performance more broadly, but the existence of this guidance does not mean every research-reagent testing method has been validated under it.
Researchers should ask which performance characteristics were actually evaluated for the procedure in question rather than assuming that the presence of an analytical instrument establishes method suitability.
Why independent testing helps without automatically settling every disagreement
Independent analysis can provide an additional source of evidence about a material. When a laboratory operates separately from a supplier, the resulting report may help researchers assess a claim without relying exclusively on the supplier’s own description.
Independence, however, is not a substitute for analytical detail. A report is most useful when the tested sample, analytical procedure and reported findings can be understood and connected to the material being evaluated.
The educational explanation this explanation of independent laboratory testing describes how submitted samples are identified, analyzed and associated with laboratory documentation.
A report may establish that the submitted sample met a particular analytical criterion under the conditions used. It cannot necessarily prove that every portion of a larger batch has identical characteristics, especially if sampling procedures and batch homogeneity have not been established.
Nor does an independent report automatically validate the supplier’s full manufacturing process or every possible characteristic of the material.
When two independent laboratories disagree, the appropriate response is to examine the analytical evidence rather than rank the reports solely by the laboratories’ names or reputations.
The investigation may benefit from reviewing the original chromatograms or spectra, sample identifiers, preparation records, analytical controls and documented method limitations. Which records are necessary depends on the measurement and the nature of the disagreement.
If the same material was not actually tested, the comparison should not be presented as a direct interlaboratory disagreement.
Reference materials can reveal whether the disagreement comes from the method
Reference materials provide an important basis for evaluating measurement processes. They allow laboratories to compare their results against materials with characterized properties or to assess how well a particular procedure performs.
NIST supplies several categories of reference materials, including Standard Reference Materials, Reference Materials and Research Grade Test Materials. These categories do not carry identical levels of characterization, intended use or metrological traceability.
For example, a NIST Standard Reference Material may have certified property values with associated uncertainties. A Research Grade Test Material can instead be intended for research or interlaboratory comparison without necessarily carrying an assigned measurement value.
The distinction matters because a reference material is not simply any sample described as highly pure or scientifically useful. Its value depends on whether its properties are adequately characterized for the measurement being evaluated.
In a controlled investigation, appropriate reference materials may help laboratories assess calibration, method performance or possible systematic differences.
However, the suitability of a reference material must be established for the specific measurement. A material designed to evaluate one analytical characteristic may not answer questions about another.
Researchers should also avoid treating commercial research reagents as certified reference materials unless documentation supports that designation. Similar appearance, chemical naming or stated purity does not establish equivalence to a formally characterized reference standard.
The objective is not to force two instruments to produce identical numbers. It is to determine whether their results are compatible with the properties being measured and the known performance of their methods.
Documentation gaps can turn a minor analytical difference into a larger research problem
Even a technically sound test can be difficult to interpret when its documentation is incomplete.
Imagine receiving two reports that appear to describe the same compound. One includes a sample identifier, lot reference, method description and analytical date. The other contains only a product name and a final percentage.
There may be no reason to distrust the second laboratory’s competence. Nevertheless, the missing information makes a meaningful comparison harder because researchers cannot readily establish whether the reported results describe equivalent samples and methods.
The National Academies of Sciences, Engineering, and Medicine addressed a broader version of this concern in its 2019 report, Reproducibility and Replicability in Science. The report emphasized clear descriptions of research methods, instruments, materials, measurements and uncertainty as part of making scientific findings understandable and assessable.
In analytical research, useful documentation should make it possible to reconstruct what question was asked and how the reported result was obtained.
This does not require every report to have the same format. It requires enough information to distinguish a genuine scientific discrepancy from an administrative mismatch.
When essential records are absent, additional testing may be necessary. But even a fresh analysis will not necessarily reconstruct the condition or identity of an earlier sample.
Preserving the original records is therefore a scientific requirement as much as an administrative convenience.
A practical framework for investigating conflicting laboratory reports
A disagreement is easier to evaluate when the investigation follows the evidence in a defined order. Before deciding that material has failed a specification, research teams can examine several questions:
- What was measured? Confirm that both reported results concern the same analytical property, rather than different forms of purity, identity or quantity.
- What sample was tested? Review sample identifiers, lot references and available records establishing whether the material was comparable.
- Which methods were used? Identify differences in analytical procedures and whether each was suitable for its stated purpose.
- How was uncertainty addressed? Determine whether the numerical difference is meaningful in relation to the performance and uncertainty of the measurements.
- Could sample condition matter? Review applicable storage, transport and handling records without assuming that degradation occurred.
- Can the findings be examined? Determine whether supporting analytical information and relevant quality-control records are available for scientific review.
The answers may show that the laboratories used different reporting conventions. They may reveal an analytical limitation, a sample-handling issue or a genuine difference requiring further investigation.
They may also establish that the available evidence is insufficient to identify the cause. That conclusion is more defensible than claiming certainty that the records cannot support.
Research organizations should define relevant acceptance criteria before reviewing the disputed results whenever feasible. Changing the criteria after examining the numbers risks replacing scientific judgment with a preferred outcome.
If additional analysis is warranted, its purpose should be defined clearly. Repeated testing should address an identified uncertainty or suspected source of variation rather than continue until a convenient result appears.
Better comparisons produce stronger scientific conclusions
Different laboratory results do not necessarily indicate failed testing, defective materials or conflicting scientific competence. They may reflect differences in the properties measured, the methods used or the condition of the samples examined.
At the same time, measurement uncertainty cannot explain away every disagreement. Significant discrepancies deserve investigation, and poorly documented results should not receive more confidence than their supporting evidence permits.
The useful question is not simply which laboratory reported the higher number. It is whether the two measurements are scientifically comparable and whether each provides reliable information for its intended purpose.
That approach demands more than reading the final percentage. It requires attention to method performance, sample identity, appropriate reference materials and documented uncertainty.
For laboratory research, the strongest analytical record is one that can be questioned, examined and understood by someone other than the person who produced it. When that standard is met, differing results become evidence to investigate rather than an argument that must be won.