VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

The Gold Standard of Intelligence | RDS Protocol-Control Library Versioning, Retired Guidance, Contradiction Checks, Translation Propagation, Training Acknowledgement, Field Adoption Evidence and Estimator Compatibility

Three secondary students review open books and notes together in a classroom.

THE GOLD STANDARD OF INTELLIGENCE · SPECIALIST STUDY

Keep every RDS field team on one explainable protocol while preserving the evidence needed to compare data across versions

A deviation review may produce new instructions, translated prompts, training memos, device rules and estimator notes. If those controls are released independently, an interviewer can follow one document while a supervisor audits another and an analyst assumes a third. A protocol-control library makes the active rules visible, resolves contradictions, propagates translations, records acknowledgement and tests actual field adoption. It also protects comparability: every case, recruitment link and degree measure should be traceable to the version that shaped it.

This guide belongs to the Intelligence subject root, the Gold Standard of Education, and the How X Works hub. It begins with a concrete decision, then tests whether the method still holds when the context changes.

The whole problem at a glance

Library questionRequired evidenceUnsafe shortcut
What is authoritative?Version, owner and effective scopeNewest file
Which rules conflict?Requirement-level comparisonKeyword search
Who received the change?Role and language propagationEmail sent
Did practice change?Observed adoption and case lineageAcknowledgement clicked
A protocol library governs meaning, not just files

Authority, propagation, adoption and analysis require separate evidence.

CHAPTER 1 OF 13 · Define the protocol-control system

1. Define a control as a governed instruction

A protocol control is more than a document: it is an authoritative rule, prompt, validation, training step or supervisory check that changes field behaviour or analytical meaning. Give it an owner, scope, version, effective time and relationship to the approved protocol.

Evidence record: Keep control ID, source authority, requirement text, rationale, affected roles, languages, sites, systems, effective period, dependencies and superseded versions. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Reject unofficial notes as active controls until the authorised process incorporates or explicitly quarantines them. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A supervisor memo clarifies rather than changes a rule; record the interpretation and its authority without inventing a new protocol requirement. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 1 check · what should survive an audit
  • State the decision unit for “Define a control as a governed instruction”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Next chapter

CHAPTER 2 OF 13 · Define the protocol-control system

2. Build one source-of-truth register

Files can remain distributed, but the register should resolve which version applies to each role, language, site, mode and date. It must show pending, active, suspended and retired states plus exceptions. A single global “current version” is insufficient during staged rollout.

Evidence record: Record artefact location, checksum, status, scope expression, activation, exceptions, owners, acknowledgements and machine or device deployment state. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Block field use when the applicable active version cannot be resolved deterministically for a case. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A remote site activates later; represent the planned split instead of backdating one global version. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 2 check · what should survive an audit
  • State the decision unit for “Build one source-of-truth register”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 3 OF 13 · Define the protocol-control system

3. Link every change to evidence and authority

A deviation trend, ethics amendment, software defect or participant remedy may motivate a change, but the control release needs documented authority. Preserve the issue, alternatives, consultation, approval and intended outcome so later teams understand why wording or procedure changed.

Evidence record: Keep change request, evidence, affected participants, ethical and legal review, decision, dissent, approval, expected effect, validation and review date. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Refuse an emergency change without a named authority, bounded duration and later ratification path. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A safety clarification must deploy immediately; use a temporary controlled release and retrospective full review. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 3 check · what should survive an audit
  • State the decision unit for “Link every change to evidence and authority”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 4 OF 13 · Resolve versions and contradictions

4. Detect contradictions at requirement level

Two documents can use different words consistently, while identical words can impose conflicting actions in different contexts. Compare actors, triggers, actions, prohibitions, timing, exceptions and precedence. Include device specifications and training examples, not only narrative manuals.

Evidence record: Store statement IDs, parsed conditions, conflict type, affected scopes, evidence, resolver, precedence decision, repair and regression fixture. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Escalate any contradiction that can change consent, eligibility, payment, recruitment lineage or analytic interpretation before release. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A local accessibility exception seems to conflict with a standard timing rule; preserve the authorised exception rather than deleting nuance. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 4 check · what should survive an audit
  • State the decision unit for “Detect contradictions at requirement level”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 5 OF 13 · Resolve versions and contradictions

5. Resolve precedence without hiding superseded meaning

Precedence rules should identify the controlling authority and effective scope, not merely choose the newest timestamp. Preserve superseded language and decisions because cases collected under it remain part of the study record.

Evidence record: Keep authority rank, amendment relation, effective interval, sites and languages, emergency overrides, supersession graph and unresolved conflicts. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Prevent a lower-authority artefact from silently overriding an approved protocol even when it was published later. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A device hotfix predates formal protocol approval; separate technical containment from lasting procedural authority. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 5 check · what should survive an audit
  • State the decision unit for “Resolve precedence without hiding superseded meaning”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

Worked version register

ControlActive versionAffected casesCompatibility
Consent confirmation3.2All after 14 MayBreak
Degree probe2.4Language B onlyBridge
Coupon reconciliation5.1Sites North/CentralCompatible
Incentive script4.0Digital paymentsConditional
Fictional entries expose scope and analytic consequences

A real register records exact effective times and exceptions.

CHAPTER 6 OF 13 · Propagate language and training

6. Propagate changes across translations as governed versions

Translation is not a file-copy task. A source amendment can alter politeness, voluntariness, time reference or network terminology differently across languages. Use forward translation, review and targeted cognitive or field testing proportionate to risk; retain translation decisions.

Evidence record: Record source segment ID, language, translator and reviewer, terminology, unresolved choices, test evidence, approval, release and linkage to source version. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Do not activate a translated control solely because the source-language release is urgent; use an authorised interim path and document coverage. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A term has no direct equivalent; preserve the functional explanation and test respondent understanding. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 6 check · what should survive an audit
  • State the decision unit for “Propagate changes across translations as governed versions”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 7 OF 13 · Propagate language and training

7. Test source-to-device propagation

A correct library entry can fail to reach interviewer devices, printed aids or call-centre scripts. Validate package generation, distribution, installation, cache invalidation and displayed content in each operating mode.

Evidence record: Keep build ID, device or kit cohort, distribution time, install result, checksum, displayed fixture, offline state, failed deployments and remediation. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Quarantine cases from a known stale device when the changed control could affect rights or data meaning. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: An offline tablet reconnects after activation; verify version before the next interview. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 7 check · what should survive an audit
  • State the decision unit for “Test source-to-device propagation”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 8 OF 13 · Propagate language and training

8. Make training acknowledgement evidential but limited

Acknowledgement proves that a person received or opened material; it does not prove comprehension or practice. Pair role-specific notice with teach-back, scenario work or supervised observation when the change is material. Accommodate leave, language and accessibility needs.

Evidence record: Store assigned version, role, delivery, acknowledgement, assessment, observed practice, support request, exemption, make-up route and release to duty. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Do not equate a clicked acknowledgement with competence or use it as automatic blame evidence. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A returning interviewer missed the launch window; complete a bounded re-entry check before active cases. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 8 check · what should survive an audit
  • State the decision unit for “Make training acknowledgement evidential but limited”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 9 OF 13 · Measure field and analytical adoption

9. Measure field adoption through observable practice

Compare expected version use with case metadata, paradata, device build, observation and reinterview evidence. Adoption metrics need eligible opportunities and detection coverage. Avoid intrusive surveillance and separate learning from employment decisions.

Evidence record: Keep case version, device version, applicable control, opportunity, observed action, review channel, uncertainty, site context and corrective follow-up. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Escalate mixed-version practice that changes participant protection, recruitment or variable meaning; do not average it away. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: Two versions produce the same visible action but different documentation; test the evidence pathway as well as behaviour. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 9 check · what should survive an audit
  • State the decision unit for “Measure field adoption through observable practice”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

Retirement gate

GateRetire evidenceKeep available when
AuthorityReplacement approvedAppeal or open case
AdoptionField use verifiedMixed practice persists
AnalysisVersion bridge documentedEstimator assumption differs
RecordsRetention decision appliedHold or audit remains
Retired guidance can remain evidentially necessary

Unavailable for current use does not mean destroyed.

CHAPTER 10 OF 13 · Measure field and analytical adoption

10. Route exceptions without creating shadow protocols

Accessibility needs, safety constraints or local laws may require authorised deviations. Register the condition, authority, alternative procedure, duration and analytical implication. Repeated exceptions can signal that the core control needs redesign.

Evidence record: Record request, protected details minimised, approval, scope, alternative, participant impact, affected variables, review, expiry and recurrence count. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Reject informal workarounds that cannot be distinguished from noncompliance or reproduced by another authorised team. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: An exception recurs across sites; open a systemic design review instead of renewing it silently. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 10 check · what should survive an audit
  • State the decision unit for “Route exceptions without creating shadow protocols”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 11 OF 13 · Measure field and analytical adoption

11. Map versions to data and recruitment lineage

Every interview, coupon, recruitment edge, degree value and remedy should be traceable to the control versions that shaped it. Mixed versions can alter eligibility, reporting and network structure. Build analytical flags without exposing unnecessary staff or participant identities.

Evidence record: Keep case ID, protocol and language versions, device build, site, interviewer pseudonym, recruitment links, deviations, repairs, exclusions and audit access. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Do not overwrite case lineage when a control is corrected later; add the new interpretation and decision. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A questionnaire version changes mid-chain; test whether derived recruitment measures remain comparable. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 11 check · what should survive an audit
  • State the decision unit for “Map versions to data and recruitment lineage”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 12 OF 13 · Retire safely and preserve comparability

12. Assess estimator compatibility before combining versions

Compatibility depends on whether versions preserve the construct, recruitment process and missingness mechanism—not on file-format similarity. Use bridge cases, measurement checks, network diagnostics and scenario sensitivity. State when pooling is conditional or unsupported.

Evidence record: Record construct mapping, eligible populations, overlap, bridge sample, response distributions, degree heaping, homophily, weights, estimates and uncertainty by version. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Block pooled claims when a version change materially alters the target or required estimator assumptions. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: A new probe reduces degree heaping but changes the distribution; model or stratify rather than silently recoding history. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 12 check · what should survive an audit
  • State the decision unit for “Assess estimator compatibility before combining versions”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Next chapter

CHAPTER 13 OF 13 · Retire safely and preserve comparability

13. Retire guidance while preserving evidence and reopen rules

Remove retired material from current field routes, revoke obsolete device packages and verify no new cases use it. Preserve records needed for audit, participant remedy, appeals and historical analysis under declared access and retention. Define triggers for restoring review when contradictions or stale use reappear.

Evidence record: Keep retirement authority, replacement, last-use date, dependency scan, device cleanup, archived versions, holds, retention, late-use alerts and reopen owner. Preserve timestamps, versions, denominators, excluded cases, failures and rejected alternatives so another reviewer can reconstruct the decision rather than inherit a conclusion.

Decision rule: Close retirement only when current practice is singular or explicitly governed and analytical compatibility is documented. Name the authorised owner, trigger, stop condition, remedy and verification window. A control is incomplete until a changed-context case tests the mechanism it is meant to protect.

Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.

Changed-context transfer: Transfer the method to multi-language school surveys while replacing RDS recruitment-specific controls. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.

Chapter 13 check · what should survive an audit
  • State the decision unit for “Retire guidance while preserving evidence and reopen rules”.
  • Bind the claim to versions, denominators, missing cases and time boundaries.
  • Connect evidence to an owner, bounded action and later verification.

Contents · Previous chapter · Return to the Intelligence guide

Verification lab: work the decision twice

Build a fictional protocol-control register across three sites, two languages and one staged device release. Resolve which version applies to six cases.

Compare two contradictory guidance statements at actor, trigger, action, exception and authority levels. Repair the conflict and create a regression fixture.

Pool or separate estimates across a degree-probe change. Use case lineage, bridge evidence, network diagnostics and uncertainty to justify the bounded decision.

Limits and misconceptions

The newest file is not automatically authoritative

Authority, effective scope and precedence decide applicability.

Acknowledgement is not adoption

Reception, understanding and observable practice are different evidence.

Translation release is not semantic equivalence

Material changes need reviewed terminology and appropriate testing.

A retired control may remain a record

Audit, remedy and historical analysis can require preserved versions.

Questions readers ask

What belongs in a protocol-control library?
  • Authoritative procedures, prompts, validations, training and supervisory checks with versions, scopes and owners.
How are contradictions found?
  • By comparing actors, conditions, actions, prohibitions, exceptions and authority—not only words.
Does a training click prove compliance?
  • No. It shows receipt at most; comprehension and practice need separate evidence.
How should translations be versioned?
  • Against source segments with reviewers, terminology decisions, testing, approval and effective scope.
Can data from two versions be pooled?
  • Only when construct, recruitment and estimator compatibility are supported by evidence.
When can retired guidance be deleted?
  • Only under retention, hold, audit, remedy and historical-analysis rules after current-use paths are removed.

Practise, verify and continue

Govern an RDS control amendment across languages, sites, devices and training cohorts. Include contradiction checks, adoption evidence, exceptions, case lineage, estimator compatibility and retirement.

Move the method to a multi-site education survey. Preserve authority, translation lineage, field adoption and version compatibility while replacing RDS coupon and degree fields.

Practical evidence check

Before implementation, write a one-page evidence map with the decision, unit of analysis, version or time boundary, inputs, outcomes, uncertainty, authority and stop rule. Attach one worked case that passes and one that fails. A reviewer should be able to reproduce the reasoning without relying on the memory of the person who built the system.

Run one changed-context rehearsal before approval. Alter a mechanism rather than a cosmetic detail, then identify which calculation, safeguard or governance gate must change. Record any conclusion that does not transfer. This short rehearsal is often the fastest way to expose a hidden assumption before it becomes an operational dependency.

Related eduKate reading

RDS protocol-deviation trend review
RDS supervisor intervention and deviation triage
RDS survey-language version changes
RDS adaptive fieldwork governance

Authoritative sources and scope

CDC NHANES 2025–2027 procedure manuals
CDC NHANES interviewer procedures and quality-control manual
US Census Bureau Statistical Quality Standard B1 for data collection methods
CDC Global Tobacco Surveillance System fieldwork and data-management documents

CDC and Census sources support comprehensive procedures, training, monitoring, corrective action and replicable documentation. They do not prescribe this RDS protocol-control library or employment decisions. Versions, examples and compatibility classifications are fictional; approved protocols, ethics, consent, privacy, labour processes and analytical methods govern real studies.

Back to contents · Return to the Intelligence guide