U.S. Democracy Framework - Conceptual Research Planning Doc

A Nonpartisan Democracy Strengthening Framework

NextDemocracy.us · Version 0.44 · March 2026 · Draft — Not for Distribution

What This Is — and Why It Matters

Every major democracy framework — the EIU Democracy Index, Freedom House, V-Dem, the WJP Rule of Law Index — measures democratic health at the country level. None of them tells a voter whether a specific candidate on their ballot, company, or initiative is a steward or a threat to the institutions they depend on. The U.S. Democracy Index fills that gap.

It scores candidates, elected officials, policies, corporations, and media organizations against the democratic standards in the U.S. Constitution — using publicly available evidence, quantitative benchmarks where available, and an evidence-grading system aligned with the Collapse to Flourishing Index (CFI).

The framework is structured in the simple progression for how we “live democracy”:

Pillar 1 - The conditions citizens need to participate in democracy: Civic participation, civil society freedom, and economic conditions that prevent corruption and private capture of the political process.

Pillar 2 - What citizens need to make informed decisions: Press freedom, protection from disinformation and foreign interference, and privacy from government surveillance.

Pillar 3 - How citizens exercise power: Whether elections are free and fair, outcomes are respected, and money doesn't corrupt the equal weight of each vote.

Pillar 4 - How elected officials are held accountable once in office: Independent courts, executive accountability, and the public's right to know what its government is doing.

The 5 Pillars

Nonpartisan by Design

Every dimension rests on documented support from across the political spectrum. Questions apply identical criteria regardless of party, and a cross-partisan expert panel review will be done before any public release. The framework explicitly covers issues where the right has historically led: IRS targeting of conservative nonprofits, executive power overreach, FISA Section 702 reform, and antitrust enforcement. It covers issues where the left has historically led: voting access, media independence, and worker ownership. The goal is to push both parties toward stronger democratic practice — not to advantage one over the other.

Scoring

Every question uses a −5 to +5 scale. +5 always means independently verified, measurably positive outcomes; −5 always means documented violations or the use of official power to undermine democratic institutions. Every score requires specific, cited evidence, no score can be assigned without a public record to point to. The full scoring rubric, differentiated by subject type (candidate/politician, policy, organization (corporate/union/etc), media outlet), is in the Appendix.

How Assessments Are Produced

Assessments are produced through a structured expert-led process. An AI-assisted draft is prepared from publicly available records — FEC filings, voting records, court documents, and quantitative data anchors. Domain experts review that draft and adjust scores where context changes the picture. Reviewed assessments are published with full evidence citations. After publication, citizens and subject-matter experts are invited to submit additional evidence, and scores are updated whenever the record changes. If different reviewers have divergent scores, the reasons will be published alongside them.

Get Involved

We're seeking academic reviewers from democracy studies, political science, and law; cross-partisan practitioners across the ideological spectrum; data partners; and funding partners committed to nonpartisan civic infrastructure.

To contribute or respond to this draft:

Appendix A: All 36 Questions

Questions are worded to apply across subject types — candidates, policies, organizations, and media. Full rubrics, evidence checklists, constitutional anchors, and data source links are in the Assessment spreadsheet.

Appendix B: Scoring Rubric by Subject Type

The −5 to +5 scale is consistent across all questions. What counts as sufficient evidence differs by subject type. Any dimension score ≤ −3 or individual question score ≤ −4 appears in the 🚨 RED FLAGS summary. Q24 (E-B) auto-flags at ≤ −3 regardless of all other scores.

Framework Architecture — 4 Pillars · 12 Dimensions · 36 Questions