Follow-Up Visit or Survey Protocol¶
Data-collection protocol — instantiates Longitudinal Follow-Up Validation
Recontacts the very people a validation claim was made about — patients, trainees, participants — on a defined schedule to measure directly whether the intended outcome still holds.
Some validation claims are about people, and the only way to know whether they still hold is to go back and ask. Follow-Up Visit or Survey Protocol is the scripted procedure for reaching the original human subjects again — by return visit, phone call, or structured questionnaire — at defined points after the initial result, and recording a fixed set of outcome measures each time. Its defining character is that it depends on direct contact with named individuals who can move, decline, or disappear: it does not read a sensor or mine a database, it re-establishes a relationship with a person and collects what only that person can report. That dependence is both its power — it captures delayed outcomes nothing else can see — and its central vulnerability.
Example¶
A state-funded welding certification program wants to know whether its six-week course actually produces durable employment, not just a passing exam on the final day. Immediate pass rates say nothing about whether graduates are still welding a year later. So the program writes a follow-up protocol: every graduate is contacted at 6 and 12 months by a structured phone survey, and a subsample gets a short workplace visit. The recontact procedure fixes how many attempts to make, across which channels (phone, text, the emergency contact each trainee gave at enrollment), and in what window around each anniversary. The survey asks the same battery every time — currently employed, currently welding, hourly wage, and whether the specific certified techniques are used on the job. Twelve months out, the protocol surfaces something the exam could not: a cluster of graduates who passed cleanly but left welding within months because local shops wanted a different certification. That finding is delayed-outcome evidence the original validation had no way to produce, and it flows straight back to the program's curriculum committee.
How it works¶
- Fix the recontact procedure first. Contact windows, channels, number of attempts, and the fallback contacts collected at baseline — decided before follow-up starts, so effort is uniform across subjects rather than concentrated on the easy-to-reach.
- Run on a schedule tied to the outcome, not the calendar. The 6- and 12-month points are chosen because that is when skill transfer and job durability become visible, not because the fiscal year ends.
- Ask the same indicators every wave. A stable battery that represents the original claim, so waves are comparable and drift in wording does not masquerade as drift in outcome.
- Choose the mode to fit the outcome. A survey for self-reported status; an in-person visit where the outcome must be observed rather than reported.
Tuning parameters¶
- Contact cadence — more waves catch faster-moving outcomes but raise cost and respondent fatigue; too few and a reversal between waves is invisible.
- Contact effort ceiling — how many attempts and channels before a subject is marked unreachable; more effort lowers non-response but spends heavily on the hardest cases.
- Mode (survey vs. visit) — visits observe directly and resist self-report bias but cost far more per subject; surveys scale but trust the respondent's account.
- Indicator battery length — a longer battery captures more but depresses completion; each added question trades coverage for response rate.
- Incentive — a token for responding lifts participation but can skew who bothers to answer.
When it helps, and when it misleads¶
Its strength is unique: for outcomes that live inside a person — retention of a skill, persistence of a health improvement, whether a behavior stuck — direct recontact is the only instrument that reads them at all. It captures Kirkpatrick's Level 3, on-the-job behavior[1], which a completion metric never touches.
Its failure mode is non-response bias, and it is quiet and severe. The people who answer a follow-up survey are systematically different from those who do not — often better off — so a protocol that reports only on responders can manufacture a reassuring result out of the disappearance of the affected. Self-report adds recall and social-desirability distortion on top. The classic misuse is to declare the program a success from a 40% response rate without asking who the missing 60% were. The guarding discipline is to at least describe who did not respond and how they differ at baseline — a rough characterization, not the formal missingness correction that Longitudinal Cohort Study performs — and to hold response rate itself up as a headline number, not bury it.
How it implements the components¶
This protocol owns the human-recontact side of the archetype — the machinery for getting delayed evidence out of people:
stakeholder_recontact_protocol— it is the recontact procedure: the channels, windows, attempt limits, and baseline fallback contacts that define how each subject is reached again.longitudinal_observation_schedule— the fixed wave structure (6 and 12 months) that keeps follow-up from collapsing into one late, convenient snapshot.sustained_outcome_indicator_set— the stable survey battery that represents the original validation claim across every wave.
It does not statistically correct for who was lost, attrition_and_missingness_control, nor stand the followed group against a matched control_or_comparison_panel — both belong to Longitudinal Cohort Study, which turns raw recontact data into an unbiased estimate.
Related¶
- Instantiates: Longitudinal Follow-Up Validation — it is the archetype's hands-on instrument for reaching human subjects again.
- Sibling mechanisms: Longitudinal Cohort Study · Incident and Adverse-Event Reporting · Post-Market Surveillance Registry · Scheduled Revalidation Review · Telemetry Drift Dashboard · Security Patch Effectiveness Monitor · Warranty and Failure-Return Analysis
Editorial Notes¶
Form Classification¶
Form family: Monitoring, Sensing & Alerting
Rationale: Follow-Up Visit or Survey Protocol operates as an ongoing sensing arrangement that repeatedly observes actual state and surfaces changes or alerts because it recontacts the very people a validation claim was made about — patients, trainees, participants — on a defined schedule to measure directly whether the intended outcome still holds.
Independent corroboration: The frozen evidence defines Follow-Up Visit or Survey Protocol as 'Recontacts the very people a validation claim was made about — patients, trainees, participants — on a defined schedule to measure directly whether the intended outcome still holds', so its operative form is Monitoring, Sensing & Alerting.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Statistics & Experimental Design
Origin pattern: Convergent development
Present-day reach: Multi-domain
Rationale: Longitudinal study design is primary because a follow-up protocol standardizes when and how later observations are collected. Clinical visits, administrative surveys, ethnographic revisits, and educational follow-ups supply independent professional lineages; this is an established multi-domain protocol.
Related originating lineages:
- Education & Pedagogy — Program evaluation applies follow-up to durable training and employment outcomes.
- Ethnography & Qualitative Methods — Direct participant recontact materially shapes contextual outcome collection.
- Medicine & Healthcare — Scheduled follow-up of the same patients or participants is canonical clinical and epidemiological practice.
- Public Administration & Policy — Program evaluation adapted follow-up surveys to trainees and service recipients.
Review resolution: Longitudinal study design is primary because a follow-up protocol standardizes when and how later observations are collected. Clinical visits, administrative surveys, ethnographic revisits, and educational follow-ups supply independent professional lineages; this is an established multi-domain protocol.
Review outcome: Researched adjudication after independent review; high confidence.
Sources consulted:
Notes¶
The protocol collects; it does not interpret. A stream of survey waves only becomes a validation judgment once someone corrects it for who dropped out and compares it against a baseline expectation — which is why this mechanism is almost always paired with an analytic sibling rather than run alone. A follow-up protocol reporting its own raw numbers is the single most common way a program flatters itself.
References¶
[1] Kirkpatrick, D. L., and Kirkpatrick, J. D. Evaluating Training Programs: The Four Levels. 3rd ed. Berrett-Koehler Publishers (2006). Defines Level 3 as post-training change in job behavior—the transfer of learning back to work. registry ↩