Make the inference visible before trusting it
A reference material can remain within its expiry date while repeated opening changes its headspace, moisture or contamination risk. This guide turns container history into a defensible decision about continued use.
A reference material can remain within its expiry date while repeated opening changes its headspace, moisture or contamination risk. This guide turns container history into a defensible decision about continued use. The practical aim is not to force one neat answer. It is to show what is known, what remains conditional and what action is justified now.
A gold-standard process protects the chain from evidence to inference. It uses original worked examples, changed-context transfer checks, explicit limits and a record that another person can test. Readers can enter through the closest route below or move chapter by chapter.
Choose the route closest to your decision
Chapter contents
Route 1: chapters 1–3
Route 2: chapters 4–6
Route 3: chapters 7–9
1. Start with the certificate and intended use
Read the certificate or information sheet before inventing a handling rule. Identify measurand, assigned value, uncertainty, minimum sample mass, storage, preparation, expiry and any single-use statement. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Does the planned use stay inside the conditions for which the value was established? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A soil reference material is certified for elemental mass fractions with a stated minimum test portion and storage instruction, but the local method proposes a smaller portion. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Repeat the check when using the material only for precision monitoring rather than value assignment. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. An expiry date does not excuse noncompliant handling, and a valid certificate does not certify every possible use. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Record certificate version, lot, unit number, intended use, required conditions and any clarification from the producer. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
2. Treat the closure as part of the measurement system
Bottle, vial, cap, liner, septum, ampoule and crimp form a barrier system. Their permeability, cleanliness and resealing behaviour influence the material presented to the method. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Can the closure still control the pathways that matter for this analyte and matrix? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A polymer liner that is adequate for a dry mineral may be unsuitable when trace organics can sorb to it or permeate through it. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the analysis to a reference gas, where valve dead volume and leakage replace powder moisture as the dominant concern. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. A lid that feels tight is not evidence of a known seal, and visual inspection cannot detect every leak. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Record closure type, manufacturer information, opening tool, observed damage and resealing method. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
3. Map headspace and exchange pathways
Opening replaces the original headspace with laboratory air and creates routes for water vapour, oxygen, carbon dioxide, volatile compounds, particles and microbes. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Which exchanged species could alter the measurand, matrix or sample preparation? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A hygroscopic powder gains moisture after repeated exposure; weighing the same nominal mass now introduces less dry material than before. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the map to volatile reference solutions where analyte loss into headspace and through a damaged septum may dominate. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Headspace volume alone does not determine risk; partitioning, exposure time, temperature and agitation matter. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Document opening duration, room conditions, remaining fill, agitation and the plausible exchange mechanisms. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
| Checkpoint | Question |
|---|---|
| Object | Which exchanged species could alter the measurand, matrix or sample preparation? |
| Worked comparison | A hygroscopic powder gains moisture after repeated exposure; weighing the same nominal mass now introduces less dry material than before. |
| Transfer | Transfer the map to volatile reference solutions where analyte loss into headspace and through a damaged septum may dominate. |
| Boundary | Headspace volume alone does not determine risk; partitioning, exposure time, temperature and agitation matter. |
Contents · Previous · Archaeology root · Next
4. Create an opening-history record
Count openings, but also capture duration, environment, amount withdrawn, tools, resealing, storage return and anomalies. One poorly controlled opening may matter more than ten brief controlled ones. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Could another analyst reconstruct the container’s exposure history? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A log shows five routine openings under dry conditions and one unplanned thirty-minute exposure during a balance failure; the exceptional event drives review. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Apply the log to shared facilities by using a barcode scan and mandatory exception field. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. A simple tally can create false equivalence between very different exposures. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Keep date, operator, purpose, withdrawn mass, exposure time, ambient controls, closure inspection and exception notes. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
5. Watch mass, moisture and container integrity
Use suitable controls such as gross-mass trends, moisture checks, blank exposure, leak checks or closure inspection, chosen for the material and required sensitivity. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Does the control detect the failure mode before it changes reported results? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A bottle’s gross mass rises beyond balance and handling variability, prompting a moisture assessment before further certified use. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the idea to ampoules by checking break integrity and transfer recovery rather than resealing. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Stable gross mass does not rule out composition change when gains and losses offset or the measurand is at trace level. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Record method capability, acceptance limits, baseline, trend, uncertainty and disposition of each alert. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
| Checkpoint | Question |
|---|---|
| Object | Does the control detect the failure mode before it changes reported results? |
| Worked comparison | A bottle’s gross mass rises beyond balance and handling variability, prompting a moisture assessment before further certified use. |
| Transfer | Transfer the idea to ampoules by checking break integrity and transfer recovery rather than resealing. |
| Boundary | Stable gross mass does not rule out composition change when gains and losses offset or the measurand is at trace level. |
Contents · Previous · Archaeology root · Next
6. Work a repeated-opening case
Combine certificate requirements, opening log, control results and method sensitivity instead of relying on one indicator. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Would the material pass if the most adverse plausible exposure were used in the calculation? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. After eight openings, a powder has a small mass gain and a control result near its warning limit. The lab restricts it to qualitative checks while a sealed unit confirms the bias. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Change the context to a robust major-element method and reconsider whether the same change is decision-relevant. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. A statistically detectable change may be practically negligible, while a small undetected change may be critical near a decision threshold. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Save the calculation, comparison unit, replicate data, control charts, reviewer and temporary restrictions. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
7. Separate contamination from composition drift
Contamination adds or redistributes constituents; composition drift can occur through moisture, oxidation, volatilisation or segregation. The diagnostic plan must distinguish them. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Does the observed pattern match a blank contaminant, an exposure pathway or heterogeneous withdrawal? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. Iron rises only in portions withdrawn with one stainless spatula, while moisture changes across the whole bottle; two mechanisms require two controls. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the separation to isotope work by considering fractionation and carry-over independently. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. A single confirmatory result cannot identify every mechanism, especially when contamination is patchy. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Retain blank results, tool history, spatial sampling notes, replicate design and alternative explanations. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
8. Decide between continued use, restriction and quarantine
Define graded dispositions: continue as certified, continue for a narrower purpose, use only as a check material, quarantine pending study, or retire. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Is the proposed use less demanding than the evidence supports, not merely more convenient? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A material no longer supports trace-level value assignment but remains suitable for training digestion recovery at percent levels. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the rule across instruments by recognising that one method may be sensitive to moisture while another uses dry-mass correction. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Downgrading must not be used to disguise failed controls or evade the certificate. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Record disposition, permitted methods, prohibited claims, label change, system status and authoriser. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
| Checkpoint | Question |
|---|---|
| Object | Is the proposed use less demanding than the evidence supports, not merely more convenient? |
| Worked comparison | A material no longer supports trace-level value assignment but remains suitable for training digestion recovery at percent levels. |
| Transfer | Transfer the rule across instruments by recognising that one method may be sensitive to moisture while another uses dry-mass correction. |
| Boundary | Downgrading must not be used to disguise failed controls or evade the certificate. |
Contents · Previous · Archaeology root · Next
9. Prefer sub-aliquots when they reduce total exposure
Plan validated sub-aliquots or single-use units when homogeneity and minimum mass permit. The gain is fewer openings of the master container. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Does sub-aliquoting reduce exposure without introducing segregation, contamination or insufficient mass? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A well-mixed powder is divided under controlled humidity into units above the certified minimum sample mass and verified against the parent. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the plan to solutions only if container compatibility, evaporation and adsorption are addressed. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Convenient small portions can be invalid when the material is heterogeneous or the producer requires preparation of the whole unit. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Keep the aliquoting protocol, homogeneity evidence, container lot, labels, fill order and verification results. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
10. Set review triggers before the next opening
Use material-specific triggers such as maximum exposure time, control drift, mass change, damaged liner, unlogged access or producer notice. Avoid a universal number of openings. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Would the trigger fire early enough to prevent a questionable result? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. The lab requires review after any humidity excursion and after cumulative open time exceeds the validated study range, not after an arbitrary tenth opening. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the trigger to a freezer-stored material by including thaw cycles and condensation risk. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. A fixed opening count may be too strict for a robust material and too loose for a volatile one. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Record trigger rationale, monitoring owner, alert mechanism, response time and reset rule. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
| Checkpoint | Question |
|---|---|
| Object | Would the trigger fire early enough to prevent a questionable result? |
| Worked comparison | The lab requires review after any humidity excursion and after cumulative open time exceeds the validated study range, not after an arbitrary tenth opening. |
| Transfer | Transfer the trigger to a freezer-stored material by including thaw cycles and condensation risk. |
| Boundary | A fixed opening count may be too strict for a robust material and too loose for a volatile one. |
Contents · Previous · Archaeology root · Next
11. Answer practical laboratory questions
Analysts need clear answers about wiping threads, replacing caps, purging headspace, returning material, and using results produced before a failure was discovered. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Can the answer be traced to a certificate, producer instruction, validation study or risk assessment? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A new cap is not automatically equivalent because liner material and torque differ; the laboratory first establishes compatibility and documents the change. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the answer format to contractors so that off-site handling is controlled consistently. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Do not generalise one SRM’s handling rule to another material with a different matrix or container. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Publish concise questions, approved answers, evidence basis, escalation path and revision date. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Can one rule settle every case?
No. Use a declared rule for the present purpose, then test whether a changed purpose, scale or decision threshold requires a different answer.
Does more data always improve the decision?
No. More data can duplicate the same blind spot, add poorly controlled observations or create false confidence if provenance and uncertainty are weak.
When should work pause?
Pause an irreversible decision when a plausible unresolved condition could change it and when waiting or gathering targeted evidence is proportionate to the risk.
What makes an answer auditable?
A reviewer must be able to find the inputs, transformations, assumptions, alternatives, authority, decision and residual uncertainty without relying on memory.
Contents · Previous · Archaeology root · Next
12. Audit the container lifecycle
Link receipt, storage, opening history, control data, dispositions and results that used the material. Review whether a late failure affects earlier measurements. In reference materials used to anchor archaeological laboratory measurements, the central claim is whether the contents of an opened container remain fit for the certified or assigned purpose. This chapter turns that claim into a bounded question so that evidence is not collected without knowing the decision it must support.
First-principles check. Can the laboratory identify every result that depends on the questioned container? Start with the object, pathway or restriction that could make the answer wrong. State the comparison, unit, time window and authority before calculating a score or accepting a familiar label. A result that cannot be traced to those choices is not yet a defensible result.
Worked practice. A closure failure discovered in June triggers a bounded look-back to the last acceptable control, with samples re-evaluated by decision risk. The useful move is to expose the intermediate reasoning: what was observed, what was inferred, which alternative explanation remained possible and which action followed. That sequence makes the example reviewable and prevents the final answer from carrying more certainty than the evidence.
Changed-context transfer. Transfer the audit to multiple units by distinguishing lot-level from unit-level evidence. Keep the reasoning structure but replace the local quantities, controls and decision threshold. If the conclusion survives only because the old context supplied a hidden assumption, it has not transferred; it has merely been copied.
Limit and misconception. Retiring the bottle does not resolve potentially affected historical results. A common mistake is to treat the most legible number, newest file or neatest optimisation as decisive. The gold standard is to identify what the evidence cannot establish, especially where the certificate is treated as a guarantee despite handling conditions that no longer match the material’s validated state.
Evidence record. Keep traceable usage records, impact assessment, corrective action, learning and verified routes to related guidance. Add the source, date, operator or analyst, method version, exclusions and uncertainty. Link raw observations to the derived judgement without overwriting earlier states, so a later reviewer can reproduce the chain and challenge a step without losing the record.
Decision-quality drill. Before closing this chapter, write one sentence naming the supported decision and one sentence naming the strongest unsupported claim. Then change one plausible condition and ask whether the action reverses. If it does, label the decision sensitive and set a review trigger rather than hiding the boundary.
Contents · Previous · Archaeology root · Next
Continue through verified routes
Use the core routes to return to the wider Gold Standard framework, then follow the specialist articles that extend this decision without duplicating it. The external sources below define or test the authoritative concepts used here.
Authoritative sources
- NIST: Standard Reference Material FAQs
- NIST: SRM definitions
- NIST: metrological traceability FAQs and policy
- NIST: stability testing for SRM 2492
Evidence and guidance can change. Recheck the linked source, the local data and the current decision context before carrying the conclusion into a new case.