Methodology

A transparent framework for comparing very different online information systems.

What the score means

The overall score is a weighted average of six 0–10 assessments. It estimates the institutional and technical safeguards that reduce exposure to false or misleading claims. It does not certify that every item is accurate, nor does it measure political agreement.

Editorial review

Pre-publication human review, professional standards, and separation of reporting from commentary.

Corrections

Visibility, timeliness, permanence, and accessibility of corrections or revision histories.

Evidence & citations

Use of named sources, primary records, links, citations, and evidence readers can inspect.

Transparency

Clarity about ownership, funding, authorship, policies, methodology, and conflicts of interest.

Moderation / review

Processes that identify, contextualize, label, demote, remove, or correct misleading content.

Accuracy record

Historical performance, including major failures, repeated patterns, and responsiveness to error.

Important comparison rule

Publishers create most of their own material; social platforms distribute mostly user-generated material; Wikipedia reviews material collaboratively after publication; and Snopes evaluates selected claims. The common scale is useful only when those structural differences remain visible.

Formula

Overall = Σ(metric score × chosen weight) ÷ Σ(weights)

Recommended scoring process

  1. Review the source's published policies.
  2. Consult independent academic, legal, and media-accountability evidence.
  3. Record both strengths and limitations.
  4. Have at least two reviewers score independently.
  5. Publish the evidence and revision date for every change.