R0051/2026-03-31/Q002/SRC02/E01¶
W3C tech report explicitly acknowledges no standard scales exist for credibility scoring.
URL: https://www.w3.org/2018/10/credibility-tech/
Extract¶
The report organizes credibility assessment around four strategies: Inspection (content feature analysis), Corroboration (cross-referencing against fact-checks), Reputation (provider trustworthiness), and Transparency (provider disclosures).
On the four queried features:
-
Evidence quality scale: "At present, there are no standard scales, so these scores cannot be meaningfully compared." The report proposes future development based on "statistical probability that the given information will not measurably mislead each member of the population."
-
Calibrated confidence language: The report distinguishes "believable" from "credible" but provides no numerical confidence framework.
-
Bias assessment: Mentions "partisan leanings" and "bias" as reputation standards but defers methodology to "future development."
-
Source reliability tiering: Discusses granular assessment across levels (individuals, organizations, claims, time periods) without defining tiers.
The report was published as "Final" in February 2020 without textual changes from the 2018 draft — indicating the work stalled rather than completing.
Relevance to Hypotheses¶
| Hypothesis | Relationship | Strength |
|---|---|---|
| H1 | Strongly contradicts | Report itself says standard scales do not exist |
| H2 | Strongly supports | Substantial conceptual framework exists but acknowledges formal features are absent |
| H3 | Contradicts | Completed report exists with real analysis |
Context¶
The report's own acknowledgment that "there are no standard scales" is the most direct evidence for the absence of formal evidence evaluation features. This is a primary source admission, not an external observation.