Skip to content

R0052 — 2026-03-31

Mode: claim Claims: 14 Model: Claude Opus 4.6 (1M context)

Results

C001 — ICD 203 Defines Nine Tradecraft Standards

Verdict: The claim is factually correct. ICD 203 defines exactly nine analytic tradecraft standards.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The number is not nine — Eliminated - H3: Nine standards exist but govern something else — Eliminated

Sources: 4 | Searches: 4

Full analysis


C002 — No Prior Unified IC + Scientific Methodology

Verdict: No published work combining IC analytical standards with scientific methodology frameworks into a unified methodology was found.

Probability: Likely (55-80%)

Hypotheses: - H1: No prior unified methodology exists — Supported - H2: Such a methodology exists — Inconclusive - H3: Partial combinations exist but not full unification — Supported

Sources: 4 | Searches: 3

Full analysis


C003 — GRADE Separates Evidence Quality from Conclusion Strength

Verdict: GRADE explicitly separates evidence quality from recommendation strength as a "critical and defining feature."

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: GRADE does not separate them — Eliminated - H3: Separate but not truly independent — Eliminated

Sources: 3 | Searches: 2

Full analysis


C004 — IPCC Two-Axis Confidence Model

Verdict: The IPCC uses exactly the described two-axis model with the stated terminology.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Axes exist but with different terminology — Eliminated

Sources: 2 | Searches: 2

Full analysis


C005 — Mulrow 1987: None of 50 Reviews Met All Eight Criteria

Verdict: All specific details confirmed: 50 reviews, eight criteria, none meeting all eight.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Numbers or details slightly off — Eliminated

Sources: 3 | Searches: 2

Full analysis


C006 — CONSORT 2010 Was 25 Items; CONSORT 2025 Expanded to 30

Verdict: Both numbers confirmed by official CONSORT publications.

Probability: Almost certain (95-99%)

Hypotheses: - H1: 25 items in 2010, 30 in 2025 — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Numbers correct but "expanded" is misleading — Eliminated

Sources: 4 | Searches: 2

Full analysis


C007 — Chamberlin 1890/1897 and Platt 1964 Citation

Verdict: All dates and the explicit citation confirmed by primary and secondary sources.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Dates or citation slightly different — Eliminated

Sources: 3 | Searches: 1

Full analysis


C008 — Platt Numbered Final Step 1' (One-Prime) to Signal a Loop

Verdict: The 1' notation is confirmed. The "deliberate signal" interpretation is reasonable but not explicitly stated by Platt.

Probability: Very likely (80-95%)

Hypotheses: - H1: Platt used 1' to signal a loop — Supported - H2: Platt used 4, not 1' — Eliminated - H3: Notation correct but deliberate intent unverifiable — Supported

Sources: 3 | Searches: 4

Full analysis


C009 — ICD 203 Seven-Point Probability Scale

Verdict: Seven points, dual terminology, numeric ranges, and 95-99% cap all confirmed.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Scale exists but with different details — Eliminated

Sources: 3 | Searches: 1

Full analysis


C010 — NAS Published 21 Standards with 82 Elements Across Four Stages

Verdict: All numbers confirmed: 21 standards, 82 elements, four stages (8+6+4+3).

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Numbers correct but structure differs — Eliminated

Sources: 3 | Searches: 3

Full analysis


C011 — Wardle and Derakhshan Information Disorder Taxonomy

Verdict: Three categories confirmed. "Two dimensions" framing is widely used and reasonable, though original is more typological.

Probability: Very likely (80-95%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Taxonomy exists but dimensional framing is oversimplification — Supported

Sources: 3 | Searches: 2

Full analysis


C012 — Journalism Is Principles-Based, Not Methodology-Based

Verdict: No journalistic framework with the four stated features was found. Journalism operates on principles, not quantified methodology.

Probability: Likely (55-80%)

Hypotheses: - H1: Journalism lacks these methodological features — Supported - H2: Journalistic frameworks have these features — Eliminated - H3: Journalism has some structured methodology but not at this specificity — Supported

Sources: 4 | Searches: 3

Full analysis


C013 — Different Domains Use Different Terms, Creating Blind Spots

Verdict: Well-established principle in information science with strong empirical support.

Probability: Almost certain (95-99%)

Hypotheses: - H1: The claim is substantially correct — Supported - H2: The claim is substantially incorrect — Eliminated - H3: Correct but "systematic" is overstated — Eliminated

Sources: 3 | Searches: 1

Full analysis


C014 — ROBIS Catches Process Errors but Not Interpretation Errors

Verdict: ROBIS focuses on process compliance. The gap for source-level interpretation verification is genuine.

Probability: Very likely (80-95%)

Hypotheses: - H1: ROBIS catches process but not interpretation errors — Supported - H2: ROBIS catches interpretation errors — Eliminated - H3: Distinction valid but ROBIS provides partial coverage — Supported

Sources: 3 | Searches: 3

Full analysis


Collection Analysis

Cross-Cutting Patterns

The 14 claims divide into three categories by evidence strength:

Strong factual claims (Almost certain, 95-99%): C001, C003, C004, C005, C006, C007, C009, C010, C013. These are specific factual assertions about published frameworks with well-defined, publicly verifiable content. They represent the strongest subset — straightforward facts about documented standards, scales, and findings.

Interpretive claims (Very likely, 80-95%): C008, C011, C014. These combine confirmed factual components with interpretive characterizations. Platt's 1' notation is factual; the "deliberate signal" framing is interpretive. The Wardle-Derakhshan categories are factual; the "two dimensions" framing is a reasonable summary but not the original authors' exact language. The ROBIS gap is apparent from the tool's design but has not been formally documented as a limitation by its developers.

Negative existence claims (Likely, 55-80%): C002, C012. These assert the absence of something in published literature. Both are supported by targeted searches returning no contradictory evidence, but proving a universal negative is inherently limited.

A notable pattern: the claims that serve the researcher's methodology narrative most directly (C002 novelty, C012 journalism comparison, C014 ROBIS gap) all received lower probability ratings than the pure factual claims. This is appropriate — these are the claims where researcher bias risk is highest, and the evidence is inherently more ambiguous.

Collection Statistics

Metric Value
Claims investigated 14
Sources scored 44
Evidence extracts 58
Results dispositioned 74 selected + 52 rejected = 126 returned

Source Independence

Sources are genuinely independent for most claims. The strongest claims (C001, C003-C010, C013) draw from different publication ecosystems: government (DNI, IPCC, NAS), academic (BMJ, JEB, PLOS), and open-source references. The weakest independence is within individual claims where multiple sources cite the same primary document (e.g., all ICD 203 sources ultimately trace to the same directive).

For the collection as a whole, the claim set covers nine distinct frameworks across intelligence, science, healthcare, journalism, and information science — providing strong cross-domain diversity.

Collection Gaps

Gap Impact
ICD 203 PDF inaccessible (403 error) Compensated by multiple authoritative secondary sources
Platt 1964 PDF unreadable (encoded) Compensated by JEB retrospective quotation
Full text of Wardle-Derakhshan 2017 not directly parsed Compensated by multiple summaries and analyses
No access to classified IC publications for C002 Cannot rule out internal IC methodology unification
Non-English literature not searched May contain relevant frameworks for C002, C012
Academic database access limited to web search Niche publications may be missed

Resources

Metric Value
Duration ~45 minutes
Searches 32
Sources scored 44
Files produced 57