List an Audience's survey validations

Everything this Audience has been checked against. batches lists each validation request, newest first, with its status, progress while it runs, its scored surveys (score = 100 − MAE with a 95% range, and people: how far the Audience's composition could be checked against the survey's respondents) and the combined result across its surveys. overall is how close the Audience is to real surveys across all of them: each survey counts once, by its latest scored check, weighted by questions scored. combined and overall carry basis: kind: audience when the score rests on surveys of the Audience's own people, or kind: wider_population with confidence: lower when no check held enough of them and the score is a comparison with the wider population they belong to (comparedWith names who those surveys asked; audienceQuestions 0 means no survey of these people was found). The two are never mixed: while any check of the Audience's own people exists, overall rests on those checks alone. included is the plan's validations that cost no responses this month (plan identifies the tier; perMonth −1 = unlimited, 0 = none included; further validations are billed in responses). responses is the synthetic responses the caller has left this month, which validations beyond the included ones are billed from like any Study (remaining −1 = unlimited; null when the plan could not be read). minds is how many of the Audience's Minds are ready and the minimum a validation needs; below it a validation is refused with AUDIENCE_TOO_SMALL. Each survey of a batch carries relation: who it asked relative to the people who define the Audience (same, part, neighbouring, or wider for the broad population they belong to), and surveys are listed in that order. It also carries fieldPeriod, when the survey was in the field as its source states it, or null where the source states no date. outcome.search on an automatic validation says how many sources the web search read, how many surveys it found, and what each source came to. data lists the individual validations without per-question detail, and summary.verdict is the latest settled one: validated, approximate, not_comparable, or not_validated. identity on batches and overallIdentity report whether survey identity was verified. With needs_review, individual results remain visible, combined/overall scores and distinct-survey counts are null, and reason identifies the unresolved reading. Reading never starts or scores anything. Surveys tested on a subset carry respondentScope: population, selected and total Audience members, confirmed/supported matches, unknowns and excluded members. Their match percentages describe only that subset; they contribute to combined or overall only on a subgroups basis (too few questions were put to the whole Audience, so surveys of parts of it, each answered only by the matching Minds, carry the score), and each survey's inCombined says whether the combined score rests on it.