STR deck and railing movement: an evidence study

Methodology

1. Scope and question: What evidence supports a conservative handoff when a guest reports movement, looseness, cracking, separation, impact, or deterioration at a deck, balcony, stair, or guard? The unit is one occupied or turnover observation involving a walking surface, post, guard, handrail, stair, connection, visible crack, unusual movement, impact event, or access concern. Freeze a consecutive review window, property set, inclusion rules, and cutoff before examining outcomes. Exclude drills, duplicate notifications, training examples, marketing claims, and records without a traceable property event. This is a retrospective descriptive review, not an experiment, inspection, medical assessment, code review, or evaluation of STRVirtualAssistant performance.
2. Evidence dictionary: Capture original wording; exact structure and level; property-local time; occupancy and access; movement only as observed; visible damage and image provenance; weather or impact context; injury only if volunteered; keep-clear instruction; physical restriction; manager acknowledgement; qualified structural or building finding; repair evidence; and reopening authority. Each value needs a source role, original timestamp and timezone, retrieval time, correction history, and authority. Keep guest-reported, staff-observed, device-indicated, official-source-published, vendor-stated, manager-approved, contradicted, stale, and unknown as separate codes. A sent message is not acknowledgement; an appointment is not arrival; a completed task is not verification; silence is not resolution.
3. Topic-specific method 1: Location precision matters because 'the railing' may refer to an interior handrail, a porch guard, a rooftop barrier, or a stair segment. Record the level, approach, adjacent doorway, and feature without sending the guest back for measurements. A photograph should be optional evidence, never a reason to approach an edge or reproduce movement.
4. Topic-specific method 2: Movement descriptions belong to the observer. 'Wobbles when touched' can be preserved, but the coordinator should not translate it into loose bolts or structural failure. The qualified finding may later identify a connection, substrate, or nonstructural component. Keep both statements so the later diagnosis does not overwrite what triggered the restriction.
5. Topic-specific method 3: Access consequences can extend beyond an amenity. A deck may be the only route to a door, stair, grill, pool, or emergency exit. The handoff should map affected access and route any alternate-path question to the responsible authority. It must not improvise a path or assume every guest can use stairs.
6. Sampling and denominators: Build the register from the frozen channel and property list rather than memorable incidents. Count screened, included, excluded, duplicate, incomplete, corrected, conflicting, escalated, reopened, released, and unresolved records. Stratify by occupied versus pre-arrival state, after-hours status, channel, property, reporter, consequence, and response owner. Publish denominators beside every percentage and avoid pooling categories whose evidence or decision rights differ.
7. Source protocol: The closest public guidance in this review is Decks and Porches; Older Adult Falls; First Aid; Urban Flooding: Guidance for Homeowners and Renters. Information-control and communication context comes from Weather-Ready Nation; Plan and Practice Your Escape; ADA Checklist for Emergency Shelters; Privacy Framework; Federal Plain Language Guidelines; Records Management Guidance. All URLs were checked on 2026-09-28. Record title, publisher, URL, retrieval date, geographic scope, issue or revision date when shown, and whether the material is guidance, alert, regulation, or general information. None of these sources studied this site's customers or validates a local workflow.
8. Reliability and adjudication: Two reviewers independently code an ordinary-and-exception sample without seeing each other's labels. Compare disagreement on location, status, timestamp conversion, source class, restriction, acknowledgement, and release. Adjudicate against the frozen dictionary, retain both initial codes, and recode a stable sample after changing any rule. Agreement measures reproducibility, not truth, safety, completeness, or service quality.
9. Counterexample search: Actively seek records where A still photograph, lack of a visible crack, ordinary use by another person, or a tightened fastener is treated as proof that the assembly is safe. Also seek records that initially appear minor but later require restriction or correction. Retain negative cases and late evidence. A benign later outcome does not make an earlier unsupported assertion valid; an adverse result does not prove a prior message caused it.
10. Sensitivity checks: Repeat the tables after excluding the largest property group, automation-originated records, cases reconstructed later, and records without final acknowledgement. Vary the stale-information cutoff and complete-handoff definition. Report whether direction or magnitude changes. Missing records may cluster during urgent, overnight, disconnected, or cross-channel events, so missingness is a result rather than a safe state.
11. Authority boundary: The coordinator may relay approved keep-clear and emergency instructions, preserve observations, contact named property and qualified responders, restrict the feature under policy, and track a manager decision. The coordinator cannot ask a guest to load-test, shake, climb, measure deflection, lean over an edge, diagnose structure, choose a repair, or reopen access. Immediate-danger procedures outrank research collection. The review never delays emergency action to complete a field. Proposed workflow changes require a named owner, exception route, review date, and retirement condition instead of converting a correlation into an automatic rule.
12. Limitations: Remote artifacts cannot establish load capacity, connection integrity, concealed decay, foundation movement, corrosion, guard height, code compliance, repair adequacy, causation, or safe occupancy. Retained digital records overrepresent events with usable connectivity and disciplined documentation. Properties, local rules, equipment, vendors, guest behavior, weather, staffing, and severity differ. This convenience sample cannot estimate prevalence, benchmark companies, attribute causality, quantify avoided harm, interpret law or contract, or certify any property.