docs/EVIDENCE_SCHEMA.md) is designed so this view could later be populated from real per-student evidence records or an LMS integration. Every number below is demonstration data, not observed production evidence.
Class capability snapshot — Contract Law, Section A Seeded Demo Cohort
n=42, illustrative
What the evidence suggests Illustrative Insight
Students often identify the correct doctrine but struggle to distinguish superficially similar authorities — on the ABC Components authority table, most students who missed the "most applicable" pick chose Fateh Chand (jumping to remedy before formation is settled) or Carlill (keyword overlap on "offer and acceptance") rather than assessing factual fit.
Rule Knowledge sits at 84% while Issue Spotting sits at 49% — a gap consistent with students who know s.7 exists but don't reliably spot the counter-offer buried in the supplier's "confirmation" email. 9 students in this seeded cohort have been flagged for targeted issue-spotting remediation.
12 of 42 seeded students accepted the AI Critique exercise's "automatically arose" framing as correct before revealing the model discussion — accepting an AI conclusion without checking the underlying authority.
Students scoring "Demonstrated" or higher on both Case Application and Issue Spotting are a reasonable cohort for a follow-up file with no scaffolded fact-to-issue table at all.
A short authority-hierarchy drill — distinguishing "relevant keyword" from "relevant fact pattern" — before the next Decision Lab, based on the authority-selection gap above.
Individual learner drill-down Seeded Demo Cohort
A few seeded profiles showing how individual capability patterns differ even within one cohort — this is what makes a class-level average misleading on its own. Not connected to the real evidence engine.