Operational whitepaper · publication standard

Methodology

The GII-US Operational Whitepaper and Publication Standard: data, scoring, publication, governance, and implementation rules.

Canonical version v1.1.1 · 2026-07-19 · immutable snapshot · changelog

The whitepaper is the operational contract of GII-US: the data model, scoring formulas, evidence rules, review gates, and publication controls that any released value must satisfy. It pairs with the research paper, which grounds the same framework in the evaluation, causal-inference, and composite-indicator literature.

Download DOCXDownload PDF
The one rule that governs everything else: no observed headline score without an authoritative decision record, verified implementation, measurable exposure, a documented counterfactual, adequate coverage, and reviewable evidence. Withholding a score is a successful control, not a failure.

Contents

#SectionCovers
1Product definitionMission, intended users, what GII measures and does not measure, design principles.
2Scope and classificationFederal jurisdictional boundary, nine decision origins (Congress through the Federal Reserve), scoreable event types, materiality screen, six impact domains, distributional overlays.
3The Decision Impact RecordStable GII identifiers, the minimum data contract, six clocks, status and maturity states.
4Policy lineage and attributionTyped relationship graph; legal provenance versus causal contribution; judicial controls; transfer reconciliation.
5GII scorecardGII-M, GII-D over the oriented set O, GII-C, GII-K as total variation on fractional shares, GII-V coverage with the withhold rule, GII-S maturity.
6Causal evidence standardRequired causal statement, the normative A–F evidence ladder, method assignment, forecast scoring, rights and qualitative effects, attribution language.
7Source architectureSource hierarchy, authoritative decision and outcome systems, retrieval preference, data quality, PROV-O/DCAT provenance, and the PEPI external source adapter.
8Production pipelineTen stage gates, automation limits, registry architecture, preregistration mechanism, reproducible release packages.
9Review, governance, and publicationSix review roles, eligibility, corrections, semantic versioning, independence and first-party source disclosure, appeals, privacy and ethics.
10Public products and API standardDecision dossiers, institutional views, minimum API resources, explainability, GII-US naming.
11Pilot and development roadmap24–36-lineage historical pilot, work packages, success criteria including named inter-rater statistics, phased roadmap.
12Worked exampleA fictional event carried end to end: theory of change, preregistered outcome map, publication decision, and the calculation worksheet behind every dimension.
13Interpretation guidanceReading an event scorecard, comparing institutions, comparing versions, and the common interpretation errors — starting with reading GII-M as approval.

Versioning and immutability#

GII uses semantic versioning: major versions break comparability, minor versions add backward-compatible modules, patch versions clarify and correct without changing any construct, weight, threshold, taxonomy, or lifecycle definition. Prior versions remain published, retrievable, and visibly superseded — corrections create new vintages, never silent rewrites. A machine-readable statement of the current version is served at /methodology/version.json.