THE GOLD STANDARD OF INTELLIGENCE · SPECIALIST STUDY
Treat respondent-driven recruitment as a timed process: record when coupons become usable, expire or stall, distinguish recruitment delay from failure and show how the closing date censors unfinished chains
Respondent-driven sampling unfolds in calendar time. A coupon may be issued late, offered after several days, accepted but redeemed just after expiry, or remain active when fieldwork closes. Fast recruiters and fast recruits therefore have more opportunity to appear in the final sample, while slow branches can be cut off even when they would eventually continue. The gold standard records coupon clocks and policy changes, distinguishes expiry from refusal and ineligibility, analyses time-to-recruit with censoring, tests seed and subgroup timing, simulates stopping rules, protects voluntary participation and reports how the fieldwork calendar changes coverage, weighting and uncertainty.
This guide belongs to the Intelligence subject guide, the Gold Standard of Education, and the How X Works learning library. Choose a reading route, work the examples, then use the transfer checks to test whether the method survives a changed context.
Choose a reading route
Build the foundations · chapters 1–2
Apply the core method · chapters 3–5
Test evidence and alternatives · chapters 6–7
Handle limits and disagreement · chapters 8–9
Transfer and govern · chapters 10
The whole problem at a glance
| Clock event | Observed evidence | Do not infer automatically |
|---|---|---|
| Coupon activated | Issue or activation timestamp | Immediate offer |
| Coupon offered | Recruiter report if lawful | Acceptance |
| Coupon redeemed | Visit timestamp | No access barrier |
| Study closes | Administrative censoring date | Recruitment failure |
The interval between recruiter interview and recruit enrolment mixes distribution, decision, travel, scheduling and eligibility processes.
CHAPTER 1 OF 13 · Build the foundations
1. Define the coupon clock before recruitment begins
Core idea. Time-to-recruit is not well defined until activation, validity, interview availability, expiry and closure rules are fixed. Calendar days, operating days and appointment windows can produce different exposure times.
Method. Publish the clock origin and time unit, timestamp coupon issue and activation, record site closures and policy changes, and distinguish coupon expiry from study closure.
Worked example. A coupon printed on Friday becomes usable when the site reopens Monday. Measuring from printing adds two days that were never available for enrolment.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Electronic and paper coupons may have different activation evidence, and precise timestamps can increase disclosure risk.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 1 is complete only when the claimed result survives that review.
Misconception to avoid. Recruitment time always begins at the recruiter interview. It begins at the first defensible opportunity under the design.
Practical evidence check
- Recalculate ten lags using calendar, operating and active-coupon time and explain the differences.
- Return to chapter contents
CHAPTER 2 OF 13 · Build the foundations
2. Keep coupon states mutually interpretable
Core idea. Unused, unoffered, refused, lost, expired, duplicate, ineligible and still active coupons correspond to different mechanisms. A single nonresponse code hides the pathway that shaped recruitment.
Method. Use a minimal state machine, permit unknown states, retain the last observed state and timestamp, and collect reasons only when safe, voluntary and necessary.
Worked example. A coupon accepted by an eligible contact but redeemed after expiry differs from one never distributed by the recruiter, even though neither creates an enrolled recruit.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Field teams cannot observe every offer, and asking recruiters to report contacts can breach privacy or alter behaviour.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 2 is complete only when the claimed result survives that review.
Misconception to avoid. Every unredeemed coupon is a refusal. Some never reach an eligible person or remain at risk when observation ends.
Practical evidence check
- Ask two coders to classify synthetic histories and resolve every ambiguous transition.
- Return to chapter contents
CHAPTER 3 OF 13 · Apply the core method
3. Treat observed lag as a composite process
Core idea. The time between recruiter and recruit interviews can include delay in choosing a contact, delivering the coupon, deciding, arranging transport, waiting for an appointment and completing eligibility screening.
Method. Draw the process, retain timestamps for stages the study can observe proportionately, analyse the total lag and avoid naming a causal stage without corresponding evidence.
Worked example. A 20-day lag may reflect a coupon held for two weeks or an immediate offer followed by limited clinic hours; the same total carries different remedies.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Detailed stage timing is often unavailable, and self-reported dates are rounded or recalled imperfectly.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 3 is complete only when the claimed result survives that review.
Misconception to avoid. Long recruitment lag proves low motivation. Logistics, safety and fieldwork design can dominate.
Practical evidence check
- Construct two histories with the same enrolment lag but different observable stages and show which claims remain valid.
- Return to chapter contents
CHAPTER 4 OF 13 · Apply the core method
4. Use survival analysis with the correct risk set
Core idea. Time-to-event methods compare recruitment timing while accounting for coupons or recruiters that have not yet produced an event, but the unit and risk set must match the question.
Method. Define whether the unit is coupon, recruiter or eligible referral; state entry and exit times, cluster dependence, competing outcomes and repeated coupons before estimating survival or hazard summaries.
Worked example. At day seven, only coupons activated and not yet redeemed, expired or otherwise closed remain in the coupon risk set.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Coupons from one recruiter are dependent, and unobserved distribution times make coupon-level entry uncertain.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 4 is complete only when the claimed result survives that review.
Misconception to avoid. A Kaplan–Meier curve automatically solves RDS timing bias. It summarises a declared process; it does not restore missing offer data or random assignment.
Practical evidence check
- Hand-calculate the risk set and event count at three times for a small coupon table.
- Return to chapter contents
CHAPTER 5 OF 13 · Apply the core method
5. Handle expiry as policy and possible competing risk
Core idea. Expiry is an administrative rule that can prevent later enrolment and may affect subgroups differently when travel, work, stigma or site access delay redemption. It is not simply a natural recruitment failure.
Method. State the expiry rule, measure time available, record extensions consistently, compare expiries with other terminal states and model cause-specific or sensitivity summaries when appropriate.
Worked example. Two identical contacts receive coupons with fourteen days remaining and three days remaining because fieldwork is nearing closure; their redemption opportunity is not comparable.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Extending coupons can also change behaviour and create protocol inconsistency, while reasons for delay may remain unknown.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 5 is complete only when the claimed result survives that review.
Misconception to avoid. Expired means the person would never participate. Expiry censors or competes with the enrolment event under an administrative design.
Practical evidence check
- Extend every late coupon in a simulation and compare final subgroup composition without claiming the extension would change preferences.
- Return to chapter contents
Time-to-recruit example
| Coupon | Observed path | Analysis state |
|---|---|---|
| A | Redeemed on day 3 | Event at 3 days |
| B | Expires unused on day 14 | Competing administrative outcome |
| C | Still active at closure day 10 | Right-censored at 10 days |
| D | Recruit found ineligible day 6 | Separate screening outcome |
Coding all four as zero recruitment destroys timing and mechanism evidence.
CHAPTER 6 OF 13 · Apply the core method
6. Account for left truncation and late entry
Core idea. Some timing analyses observe coupons only after activation or recruiters only after becoming eligible to recruit. Including pre-entry time or excluding late entrants incorrectly can distort hazards.
Method. Enter each unit into the risk set when it becomes observable and able to produce the event, retain seed and replacement-seed timing and distinguish calendar phase from time since entry.
Worked example. A new seed added in week six has no chance to recruit during weeks one to five; its chain should not be labelled slow for that absent exposure.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Exact activation can be unknown for informal electronic invitations or redistributed coupons.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 6 is complete only when the claimed result survives that review.
Misconception to avoid. All chains are at risk from day one of fieldwork. Opportunity begins when the chain and coupon exist.
Practical evidence check
- Plot calendar time and time-since-entry for early and late seeds and compare interpretations.
- Return to chapter contents
CHAPTER 7 OF 13 · Test evidence and alternatives
7. Measure subgroup and branch timing without overclaiming cause
Core idea. Recruitment lags can vary by subgroup, seed, branch, geography, coupon mode or fieldwork phase, affecting which people arrive before closure. These associations mix network, choice, response and logistics.
Method. Report stratified survival and lag distributions with uncertainty, cluster by tree or recruiter, test interactions cautiously and pair quantitative patterns with ethical field observations.
Worked example. One subgroup has similar coupon acceptance but longer travel-to-site delays, so its active chains are disproportionately censored at closure.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Nonparticipants are partly unobserved, subgroup labels can be sensitive and small branches yield unstable curves.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 7 is complete only when the claimed result survives that review.
Misconception to avoid. A slower branch is less connected. Timing alone cannot identify network degree or willingness.
Practical evidence check
- Hold coupon mode constant and examine whether the timing difference remains, then list alternative explanations.
- Return to chapter contents
CHAPTER 8 OF 13 · Test evidence and alternatives
8. Diagnose chain ageing and survival bias
Core idea. Chains that recruit quickly produce more generations within a fixed fieldwork window and become overrepresented among observed descendants. Surviving branches are selected by both network reach and elapsed-time dynamics.
Method. Track branch age, active coupons, generation intervals and descendants over calendar time; compare early and late cohorts and run branch-level sensitivity rather than rewarding prolific chains automatically.
Worked example. A fast seed produces six waves before closure while an equally productive but slower seed reaches only three; final counts make the first look structurally superior.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. True network differences and timing differences are entangled, and extending fieldwork may introduce new secular change.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 8 is complete only when the claimed result survives that review.
Misconception to avoid. Observed large trees prove better population coverage. They may partly reflect earlier entry and shorter generation intervals.
Practical evidence check
- Simulate two identical branching processes with different generation times and compare their observed size at one closing date.
- Return to chapter contents
Stopping-rule certificate
| Question | Evidence | Boundary |
|---|---|---|
| Why stop? | Predeclared calendar, budget or precision rule | Operational discretion |
| Which chains were active? | Coupon risk set at closure | Late-issued coupons |
| Who lost opportunity? | Timing by subgroup and branch | Unobserved nonparticipants |
| What changes estimates? | Sensitivity to delayed closure | Model dependence |
A target sample size does not remove the need to report unfinished recruitment opportunity.
CHAPTER 9 OF 13 · Handle limits and disagreement
9. Separate recruitment stopping from target attainment
Core idea. Stopping at a sample-size target, calendar date, budget, apparent equilibrium or precision threshold creates different selection and uncertainty. Post hoc stopping can favour fast chains or desirable observed estimates.
Method. Predeclare primary and safety stopping rules, record every operational pause, blind discretionary extensions from outcome preferences where possible and report active recruitment opportunity at stop.
Worked example. Fieldwork closes when 500 interviews are reached, leaving 86 live coupons concentrated in slower branches; the receipt reports those unfinished opportunities.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Ethical or safety events may require immediate unplanned stopping, and budgets can impose unavoidable limits.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 9 is complete only when the claimed result survives that review.
Misconception to avoid. Reaching the target sample size completes the design neutrally. The route to the target and unfinished chains still matter.
Practical evidence check
- Apply date-, size- and precision-based stops to the same simulated recruitment history.
- Return to chapter contents
CHAPTER 10 OF 13 · Handle limits and disagreement
10. Run closure and extension sensitivity
Core idea. Because unfinished chains are unobserved counterfactuals, no analysis can know exactly who would arrive later. Sensitivity can show how plausible delays or extensions might affect composition and estimates.
Method. Use observed lag distributions within defensible strata, simulate delayed closure and coupon extensions, bound missing recruits, re-estimate outcomes and disclose assumptions rather than impute certainty.
Worked example. A two-week extension under observed lag patterns modestly changes the total but materially increases one underrepresented branch’s contribution.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Future recruitment may not follow earlier lags because saturation, news or seasonal movement changes behaviour.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 10 is complete only when the claimed result survives that review.
Misconception to avoid. Sensitivity creates the missing sample. It maps dependence on assumptions; it does not recover facts.
Practical evidence check
- Compare best-, central- and worst-case continuation scenarios and identify which decision changes.
- Return to chapter contents
CHAPTER 11 OF 13 · Handle limits and disagreement
11. Protect participants while collecting timing evidence
Core idea. Timing data can reveal routines, location, relationships or participation in a sensitive study. Recruitment monitoring must not become contact surveillance or pressure to act before expiry.
Method. Use coarse or separated timestamps where possible, minimise contact data, prohibit recruiter tracking, allow refusal, communicate expiry without urgency manipulation and restrict linkage keys.
Worked example. A system records coupon activation and redemption dates but never the identity or location of the unredeemed contact.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Some appointment systems require contact information, and very coarse times can weaken analysis.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 11 is complete only when the claimed result survives that review.
Misconception to avoid. Better timing data always justifies more tracking. Proportionality and safety set hard limits on measurement.
Practical evidence check
- Review every timestamp and identifier for analytical purpose, retention and harm if disclosed.
- Return to chapter contents
CHAPTER 12 OF 13 · Transfer and govern
12. Monitor fieldwork drift in real time
Core idea. Site hours, incentives, coupon counts, seed additions, staff practice, eligibility interpretation and public events can change recruitment speed. A pooled lag distribution can hide these design shifts.
Method. Annotate the fieldwork calendar, version protocol rules, dashboard active coupons and lag bands without punitive targets, and investigate abrupt changes with participant safety first.
Worked example. Median lag doubles after one site shortens opening hours; the team repairs access rather than labelling recruiters ineffective.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Real-time monitoring can encourage metric gaming if staff are judged by speed.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 12 is complete only when the claimed result survives that review.
Misconception to avoid. A lag trend reflects the population alone. Fieldwork is part of the observation system.
Practical evidence check
- Insert a known closure into a synthetic calendar and verify the dashboard shows a policy effect.
- Return to chapter contents
CHAPTER 13 OF 13 · Transfer and govern
13. Publish a reproducible stopping receipt
Core idea. A credible RDS report should let readers see when recruitment opportunity began and ended, how coupons were censored or expired, which branches remained active and how conclusions change under plausible timing alternatives.
Method. Release de-identified state definitions, timelines, lag summaries, risk-set rules, stopping criteria, sensitivity results and correction history; protect small cells and sensitive tree structure.
Worked example. An independent analyst can reproduce the active-coupon count at closure and the estimate under a two-week sensitivity extension without seeing participant identities.
Changed-context transfer. Changed-context transfer: apply the same reasoning structure to a school or public-service decision. Preserve the formal relationship, then replace the population, authority, privacy duties, evidence and remedy; never copy the domain conclusion.
Limits. Open tree and timing data can re-identify communities, so controlled access or aggregates may be necessary.
Evidence and review. For this chapter, keep a dated evidence record: the input, rule or assumption, result, accountable owner and a condition that would change the conclusion. Work an ordinary case and a boundary case, preserve rejected alternatives, and require an authorised independent reader to reproduce where judgement entered. Chapter 13 is complete only when the claimed result survives that review.
Misconception to avoid. A final sample size and wave count are enough. Timing and closure are part of the sampling mechanism.
Practical evidence check
- Reconstruct one stopping decision from the receipt and challenge one censoring classification.
- Return to chapter contents
Build the specialist decision record
For “The Gold Standard of Intelligence | RDS Coupon Expiry, Time-to-Recruit, Chain Censoring, Survival Bias and Fieldwork Stopping Rules”, begin with one bounded decision rather than an abstract claim. Name the affected people, data or institutions; the time window; the authority for acting; and the conclusion the method is allowed to support.
Preserve source material before interpretation. Record definitions, exclusions, transformations and the person responsible for each judgement. A reviewer should be able to separate an observed fact from an inferred result, a policy choice and a presentation shortcut.
Work one ordinary case in full. Show the input, every material rule or calculation, the output and the action that follows. Then write what the same evidence would mean under the simplest credible alternative method.
Keep an explicit unresolved state. Missing, protected, conflicting or unevaluated information must not become zero, false, consent or absence merely because a form expects one value.
Finish the record with a claim ledger. For every sentence that could drive action, identify its supporting source, inferential step, confidence or legal status, intended reader and expiry condition. This keeps descriptive evidence, formal consequence, professional judgement and institutional authority from collapsing into one impressive-looking conclusion.
Run the boundary-case audit
Select a case near the method’s domain edge: a missing input, a disputed definition, an endpoint, a sparse group, a resource limit or a legal deadline. Boundary cases reveal the contract that routine examples can conceal.
Change one assumption at a time and retain every output. Mark what stays stable, what reverses and what becomes undefined. If several assumptions move together, report that dependence instead of attributing the change to one convenient cause.
Search deliberately for a counterexample to the strongest claim. A method that works on the published illustration can still fail under a permitted input, an alternative grouping, a changed unit or an inaccessible evidence pathway.
Translate the counterexample into plain language: who would receive a different decision, what evidence would be hidden, and which safeguard or narrower claim repairs the problem.
Include an adversarial but plausible case. Let a missing group, ambiguous identifier, extreme weight, conflicting rule or urgent deadline interact with normal system incentives. The aim is not theatrical failure; it is to learn whether ordinary pressure makes the method conceal uncertainty, shift burden or reward a misleading shortcut.
Test independent reproduction
Give a reviewer the bounded inputs, public method contract and permitted access, but not the desired answer. Ask them to reproduce the result and identify every place where judgement, local knowledge or unavailable evidence was required.
Compare the reviewer’s intermediate states, not only the final number or label. Matching outputs can hide different denominators, inference paths, missing-data treatments or authorities that will diverge on the next case.
Where reproduction fails, classify the cause: ambiguous definition, undocumented transformation, incompatible implementation, inaccessible source, calculation error or legitimate alternative interpretation. Each class needs a different remedy.
Retain the disagreement and resolution. A corrected record should show what changed, why it changed and which prior outputs or decisions must be revisited.
Run the reproduction on a changed but structurally similar case after the first discrepancy is repaired. A method is not transferable merely because its authors can explain the original example. The new case tests whether the public contract carries the reasoning or whether hidden expertise is still doing essential work.
Design correction, challenge and remedy
Assign a visible route for correcting inputs, challenging assumptions and reviewing the application of the method. Technical accuracy without a usable remedy leaves the affected person carrying an error they cannot contest.
Separate urgent live-case repair from later system learning. A current decision may need pause, reassignment, fresh evidence or a human review before the organisation finishes its longer investigation.
Protect confidential and privileged material during review, but do not use confidentiality as a reason for unrecorded authority. Layer public explanation, protected oversight and individual reasons according to lawful access.
Verify that correction reaches cached results, derived records, dashboards and future decisions. Closing a complaint is not enough if the old conclusion continues to travel.
Measure remedy from the affected person’s position. A revised database row is useful, but it is incomplete when a hearing, service allocation, published claim or downstream model still relies on the old result. Record notification, restored opportunity, compensation or procedural repair where the governing framework provides it.
Set review and retirement triggers
Review after a change in definitions, population, law, source access, measurement scale, software, vocabulary or incentives. A stable formula can become invalid when the world or the meaning of its inputs changes.
Monitor leading signals as well as final outcomes: coverage gaps, timeouts, overrides, complaints, disagreement, missingness and boundary frequency. These often reveal failure before a headline metric moves.
Predefine thresholds for recalibration, suspension and retirement. Do not force a method to answer after its assumptions fail merely because replacing it is inconvenient.
At retirement, preserve the versioned contract, evidence, known limitations and migration decision. State which historic results remain interpretable and which require warning, restatement or withdrawal.
Schedule one prospective review before confidence fades or controversy forces it. Assign named owners for evidence, implementation and rights impact, and invite at least one reviewer outside the design team. The review should be able to narrow the claim, change the method or stop use without requiring proof of catastrophic failure.
Questions readers ask
What is time-to-recruit in RDS?
- A declared interval from recruitment opportunity to an observed enrolment event, with its stages and clock specified.
Is an expired coupon a refusal?
- Not necessarily; expiry is an administrative outcome that can censor or compete with enrolment.
What is chain censoring?
- Recruitment branches are cut off while still capable of continuing because observation or fieldwork ends.
Why can fast chains dominate?
- They fit more recruitment generations into the same calendar window.
Does survival analysis remove all bias?
- No. It needs valid risk sets and cannot recover unobserved offers or participation reasons.
What should a stopping receipt show?
- The rule, date, active coupons, unfinished chains, policy changes and sensitivity to plausible continuation.
Practise, verify and continue
Choose one consequential case. Reproduce the complete method, work one ordinary and one boundary case, test the strongest rival explanation, and state the evidence that would reverse or suspend the result. Build an evidence pack containing original inputs, definitions, calculations or rule steps, alternatives rejected, uncertainty, reviewer comments and the final action. Give the pack to a reader who did not design the method. Ask that reader to reproduce the result, identify one missing assumption and construct a case that would make the method return unresolved. Revise the public explanation only after the evidence pack and the operational process agree.
Move the method into an education or public-service setting. Keep the reasoning structure while replacing domain authority, population, privacy, evidence and correction routes. Start with a low-stakes historical example, then compare the specialist method with a transparent baseline. Check whether the same people, records and decisions remain inside scope; whether the scale or vocabulary still means the same thing; and whether a person affected by error can understand and challenge it. Pilot under human review, record reversals and burden, and stop the transfer if the formal resemblance hides a different ethical or legal problem.
Related eduKate reading
RDS recruiter choice and differential recruitment
RDS recruitment effectiveness and coupon nonresponse
RDS degree-question design and repeatability
Authoritative sources and scope
Gile et al., Diagnostics for Respondent-Driven Sampling
Lee, Ong and Elliott, Recruitment Mechanisms and Cooperation in RDS
McCreesh et al., Balancing Theory and Practice in RDS Recruitment
Lessons From RDS Recruitment in Nairobi
RDS Implementation and Expired Coupons in Nairobi
This article is sampling-method and fieldwork-design education, not a protocol for identifying or pressuring hidden populations. Examples are original teaching cases. Coupon and timing records can be sensitive. Participation must remain voluntary, incentives and expiry rules require ethical review, and unobserved people or unfinished chains cannot be reconstructed with certainty. Report censoring, policy changes and sensitivity without claiming causal mechanisms that the data do not identify.