How comparisons are rated

The fields answer different questions

Public label

A short reader-facing description of whether a comparison is fair and usable. It is not a confidence rating, review badge, or endorsement.

Technical verdict

A controlled classification describing the methodological condition of a comparison. Generic technical-verdict definitions are part of the methodology; applying one to a specific project requires a governed finding.

Confidence

Confidence describes evidence strength for the exact bounded finding—not confidence in a topic, institution, or reviewer. BasisMark uses controlled levels from insufficient through very high and states the scope of the confidence.

Source quality

Source quality concerns authority, provenance, method, transparency, and relevance. A strong source can still be misused if its metric or boundary does not match the claim.

Review status

Review status records whether internal and scoped external checks have occurred. It does not change the underlying metric and does not certify a result.

Alignment checks

• Metric alignment: do the values describe the same kind of water flow?
• Year alignment: do the time periods support the intended comparison?
• Geographic alignment: do national, regional, watershed, and local scopes match?
• Direct versus indirect accounting: are operational and supply-chain effects kept separate?
• System boundary: are included and excluded activities explicit?
• Value origin: are measured, reported, modeled, estimated, and extrapolated values labeled?

Check states

• Pass: the check supports the bounded comparison.
• Partial: usable only with a visible limitation or narrower scope.
• Fail: the issue materially undermines the proposed comparison.
• Not Applicable: the check does not apply and the reason is recorded.

No badge substitutes for reasoning

BasisMark does not use review status, confidence, or source quality as a seal. Readers should be able to see the relevant basis, limitations, and source trail.