Indicators of effectiveness derive from the approved test plans, test execution and reporting and the Stage Two findings; they are reported for each certified component and, where applicable, the integrated system; recorded on the certificate and schedule with boundaries, versions and contexts; and summarised publicly subject to legitimate confidentiality or legal limits. The levels are Basic (accuracy at least 80 percent, at least 30 subjects, low-risk use), Effective (at least 90 percent, at least 300 subjects in one demographic group, moderate risk), Highly Effective (at least 95 percent, at least 300 subjects per group across no fewer than three skin tone groups plus both genders, high risk) and Strict (at least 99 percent, at least 3,000 subjects spanning several demographics and environments, very high risk), applied to components and system results unless excluded under Clause 4.10. Acceptable statistical methods are fixed-sample designs meeting those minimums, sequential or adaptive designs with controlled error rates, non-inferiority or equivalence tests against the threshold using one-sided intervals, and hierarchical or Bayesian models with stratum-specific lower bounds; acceptance rules are pre-registered with confidence level, thresholds, stopping rules, priors and missing-data handling, and reduced samples are allowed where sequential designs or operational monitoring demonstrate conformity with controlled error rates.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.