A community-clinic booking pack repeats the happy path eight times and never books the last slot of the day. Which quality metric is most clearly weak?
Select an answer to reveal the explanation.
Short Explanation
Eight copies of the same sunny path and zero last-slot edge cases means poor diversity—you need a wider menu of behaviors, not more clones.
Full Explanation
A community-clinic booking pack that repeats the happy path eight times and never books the last slot of the day shows weak diversity—the need for a wide range of inputs, user behaviors, and edge cases without harmful repetition. Civic clinic access testing must exercise scarce last-slot and edge behaviors that real residents hit. Defining accuracy only as formal grammar fails because grammatical clones can still lack behavioral range. Vendor uptime of the LLM tool overnight fails as an infrastructure metric unrelated to pack variety. Redefining recall as how fast testers click the UI fails by conflating manual speed with coverage variety of generated cases. Exam caveat: diversity penalizes repetitive happy paths and rewards edge variation grounded in the basis. Operational check: inventory unique booking behaviors in the pack, add last-slot and edge civic scenarios, cap near-duplicate happy paths, and re-evaluate diversity before the clinic’s evening booking window. Retain the municipal audit note so later testers can see why the decision protects public service quality and fairness. Retain the municipal audit note so later testers can see why the decision protects public service quality and fairness.