The Gold Standard · Archaeology
Make the decision path visible
An authentic-sample panel is the bridge between a reference material and the specimens a laboratory intends to measure. It can become unrepresentative without any vial degrading. New excavation regions, preservation states, preparation methods or analytical ranges may enter practice while the old panel remains frozen in an earlier case mix.
Renewal is not simply replacing exhausted aliquots. The laboratory must preserve enough overlap to distinguish panel change from method change, document what archived specimens still represent and decide when a new population needs its own commutability claim. The aim is continuity with honest boundaries, not permanent reuse of a convenient archive.
Choose the route closest to your decision
Chapter contents
Chapters 1–3
Chapters 4–6
Chapters 7–9
Chapters 10–12
CHAPTER 1 OF 12 · DEFINE THE TARGET POPULATION
1. Define the population the panel must stand for
Start with the intended measurand, procedures and authentic specimen population. Name material type, preservation state, expected concentration or ratio range, preparation route and any matrix features likely to affect measurement relationships.
The target is not every archaeological object. It is the population for which a particular commutability conclusion will be used. A panel for digested human enamel may not represent cremated bone, sediment extracts or metal corrosion even if the same instrument reports the same element.
Separate scientific interest from ethical and legal permission. Availability for analysis does not establish that destructive reuse, international transfer or a new research question is authorised.
CHAPTER 2 OF 12 · DEFINE THE TARGET POPULATION
2. Map coverage before choosing specimens
Build a coverage map across matrix class, site or geological context, preservation, analyte level, interference pattern and preparation history. Use ranges and clusters rather than assuming one sample per label is enough.
Include difficult but legitimate specimens near analytical limits or with common interferences. A panel of clean, abundant and well-preserved samples may produce reassuring relationships that do not travel to routine work.
Do not force balance when the operational population is genuinely uneven. Preserve the real distribution for population estimates while deliberately adding challenge specimens for diagnostic power; label the two purposes.
Practical check
Write the decision, evidence and limit in three separate sentences. If one sentence depends on an unstated assumption, name it before continuing.
CHAPTER 3 OF 12 · DEFINE THE TARGET POPULATION
3. Set renewal triggers in advance
Renew when aliquots are depleted, storage evidence weakens, a method or software version changes, new matrix populations become material, or coverage diagnostics show that current cases lie outside the panel.
Time alone can be a review trigger but is not proof that a specimen is unsuitable. Conversely, passing a storage date does not guarantee representativeness after collection practice changes.
Assign an owner and a decision date. Panels often persist because everyone agrees they should be reviewed but no one is accountable for declaring the next version.
| Element | Specify |
|---|---|
| Matrix | Material, preservation and treatment |
| Range | Measurand and interference coverage |
| Procedure | Preparation, instrument and software |
| Authority | Permission, custody and use limits |
CHAPTER 4 OF 12 · AUDIT THE ARCHIVE
4. Audit archived specimen identity and permission
Confirm chain of custody, catalogue identifiers, subsampling history, treatments, storage conditions and remaining quantity. Reconcile labels across databases before selecting aliquots.
Review consent, permits, community agreements, repository terms and destructive-analysis limits for the proposed reuse. A previous commutability study may not authorise a new method, export or repeated destruction.
Record unresolved provenance separately from analytical suitability. A chemically useful specimen with uncertain authority should not enter the panel until that issue is resolved.
CHAPTER 5 OF 12 · AUDIT THE ARCHIVE
5. Reassess stability and matrix integrity
Archived specimens can change through evaporation, adsorption, oxidation, microbial activity, repeated freeze-thaw cycles or container interaction. The relevant risk depends on material and measurand.
Use inspection, mass balance, replicate measurements, control charts or targeted stability tests appropriate to the specimen. Compare new aliquots with retained results only when preparation and reporting can be bridged.
Stability of the measured value is necessary but not sufficient. A sample may remain stable while no longer representing the matrices now arriving in the laboratory.
Practical check
Write the decision, evidence and limit in three separate sentences. If one sentence depends on an unstated assumption, name it before continuing.
CHAPTER 6 OF 12 · AUDIT THE ARCHIVE
6. Work a panel-renewal case
Suppose a 24-specimen panel contains mostly well-preserved enamel from two regions. Eight specimens have too little material for another full comparison, and current work increasingly includes cremated material and low-concentration samples.
The laboratory retains 12 old specimens with adequate volume, adds eight new enamel specimens spanning the lower range and creates a separate six-specimen cremated challenge set. Old and new procedures are run on the 12 overlaps and all new specimens. The renewed general panel supports enamel claims; the cremated set is reported as exploratory because its coverage and quantity are still limited.
Changed-context transfer check: if the new collection consists of marine shells with different preparation chemistry, adding a few shells to the enamel panel would not establish representativeness. The same renewal logic leads to a distinct target population and a new evidence programme.
| Risk | Evidence |
|---|---|
| Misidentification | Catalogue and custody reconciliation |
| Degradation | Storage and stability checks |
| Selection bias | Coverage map against current cases |
| Unauthorised reuse | Permit and agreement review |
CHAPTER 7 OF 12 · BRIDGE OLD AND NEW PANELS
7. Design an overlap bridge
Retain enough specimens from the previous panel to connect versions across the analytical range and important matrix effects. Choose overlap for information, not merely because those aliquots are plentiful.
Run old and new panel members in a balanced order across batches, operators and instruments where relevant. Include reference materials and routine controls so panel effects are not confused with run drift.
Pre-specify the models, exclusion rules and criterion. A bridge designed after seeing results can select the most convenient overlap and hide discontinuity.
CHAPTER 8 OF 12 · BRIDGE OLD AND NEW PANELS
8. Test relationships, not just mean agreement
Commutability concerns the relationship among measurement procedures for authentic samples and the candidate material. Examine residual patterns, slopes, nonlinearities and matrix-dependent effects, with uncertainty.
A similar panel mean can coexist with changed behaviour at low concentrations or in one preservation class. Plot individual specimens and meaningful groups rather than relying on a single summary.
If an archived sample becomes an outlier, investigate identity, degradation, preparation and measurement before excluding it. Document any exclusion and repeat conclusions with and without borderline cases.
Practical check
Write the decision, evidence and limit in three separate sentences. If one sentence depends on an unstated assumption, name it before continuing.
CHAPTER 9 OF 12 · BRIDGE OLD AND NEW PANELS
9. Manage depletion and archived reuse
Create a quantity budget that reserves material for bridging, dispute resolution and future checks. Using every remaining aliquot in one study can make the panel impossible to verify later.
Where destructive reuse is constrained, consider fewer carefully chosen replicates, non-destructive screening or a tiered design. Reduced quantity must be reflected in uncertainty and claim breadth.
Do not manufacture continuity by renaming replacement specimens as the old panel. New membership creates a new version even when the intended population is unchanged.
| Component | Purpose |
|---|---|
| Retained members | Connect panel versions |
| New members | Fill current coverage gaps |
| Challenge set | Probe emerging matrices |
| Balanced runs | Separate panel from batch effects |
CHAPTER 10 OF 12 · GOVERN CONTINUING RENEWAL
10. Version panels and certificates
Give each panel version a membership list, specimen provenance, preparation state, storage record, target population, procedures tested, criterion and effective period.
Report whether a result concerns the full panel, a bridge subset or a challenge set. Users should be able to tell which evidence supports routine calibration and which is exploratory.
Retain superseded documentation. Historical results may need to be interpreted under the panel and procedures that were valid when they were produced.
CHAPTER 11 OF 12 · GOVERN CONTINUING RENEWAL
11. Monitor representativeness after renewal
Compare incoming routine samples with the panel using matrix descriptors, analytical range and residual behaviour. New cases outside the mapped support should trigger review rather than automatic inclusion in the claim.
Track which strata contribute most uncertainty or failures. Target future collection and permission work toward those gaps instead of repeatedly adding easy specimens.
Representativeness is a continuing relationship between panel and practice. It should be reviewed when the collection, laboratory or research question changes.
Practical check
Write the decision, evidence and limit in three separate sentences. If one sentence depends on an unstated assumption, name it before continuing.
CHAPTER 12 OF 12 · GOVERN CONTINUING RENEWAL
12. Know when reuse is no longer defensible
Retire or narrow a panel when identity or permission cannot be verified, stability evidence is inadequate, material is depleted below a defensible design, or new matrices dominate routine work.
A small bridge cannot rescue a fundamental population change. When overlap does not cover the relevant matrix effects, begin a new panel and state that continuity is limited.
The final record should explain what the renewed panel represents, what it does not, how versions connect and which evidence would trigger the next renewal.
| Outcome | Action |
|---|---|
| Comparable and covered | Adopt new version |
| Comparable but narrow | Limit claim |
| New matrix effect | Create distinct panel |
| Uncertain identity or permission | Exclude pending resolution |
Frequently asked questions
Can a stable archived specimen be reused automatically?
No. Confirm identity, permission, quantity, preparation history and relevance to the current target population.
How much overlap is enough?
Enough to cover the analytical range and important matrix effects with useful uncertainty. A fixed percentage is not a substitute for design.
Should new matrix types be added to the old panel?
Only when the target population and evidence support a shared claim. A materially different matrix may need a distinct panel.
Does renewal invalidate historical results?
Not automatically. Preserve versioned evidence so historical results can be interpreted under the panel and procedures then in force.
Continue through verified routes
Core routes
Archaeology subject root · Gold Standard of Education · Gold Standard directory · How X Works hub
Specialist routes
- Reference-material commutability drift
- Commutability matrix mismatch
- Reference-material lot changes
- Stability studies and aliquot planning