How We Score Sources
Every source review on IAAH ends in a Source Credibility Score (SCS) from 0 to 100. This page explains exactly how that number is built, so you can check our work.
In an age of information overload and AI-generated content, distinguishing historical fact from fiction is critical. Our Source Credibility Assessment Framework provides a standardized way to evaluate the reliability of historical information sources.
The six criteria
Each criterion is scored from 0 to 5. Higher is always better, including Bias, where 5 means balanced.
Evidence Quality
Are claims supported by primary sources and peer-reviewed research?
Cited sources, use of data and statistics, accuracy of dates and documents, and independent verification.
- 5
- Verified primary documents, data with links, or peer-reviewed research
- 3
- “Official sources say” without direct links
- Below 2
- No citations, or unnamed sources only
Trustworthiness / Integrity
Is there a commitment to correcting errors and avoiding fabrication?
Corrections history, accuracy over time, no proven disinformation, and non-manipulative language. Sensational or clickbait headlines lower the score.
Expertise
Does the author or organization possess verified historical qualifications?
Training and credentials, professional publications, and experience in the field.
- 5
- A recognized academic or historian
- 3
- A general reporter
Transparency
Are methods, funding, and authorship clearly disclosed?
Author identity, funding, conflict-of-interest disclosures, and organizational policies.
Consistency
Does the narrative align with established historical consensus?
Alignment with research, internal coherence, and consistent terminology.
Bias
Are ideological leanings acknowledged and managed?
One-sided narratives, demonizing language, agenda-driven framing, and a lack of balanced sources. Scored inverted: the more balanced the source, the higher the score.
- 5
- Clinical neutrality, balanced sources
- 4
- A mainstream editorial stance
- Below 3
- Strong partisan framing or demonization
Ground rules
- A 5 is the gold standard, reserved for peer-reviewed journals, verified primary documents and investigative reporting that releases its raw data.
- Even high-quality news outlets usually score 3.5 to 4.2 per criterion, because they summarize evidence rather than present it.
- Live blogs and breaking news are held to about 3 to 3.5 on Evidence and Consistency, because the facts are still emerging.
- Unless the author is a historian or holds a PhD in the field, Expertise is capped at 3.5.
- Scores are based only on content we can access and verify. A source we can't check isn't scored.
From six scores to one number
Each criterion score is multiplied by its weight. The weighted total, still on the 0 to 5 scale, is then converted to a score out of 100 and rounded to a whole number.
The formula
SCS = (0.20 × Evidence + 0.20 × Integrity + 0.15 × Expertise + 0.15 × Transparency + 0.10 × Consistency + 0.20 × Bias) ÷ 5 × 100
Worked example
| Criterion | Score | Weight | Points |
|---|---|---|---|
| Evidence Quality | 4.0 | 20% | 0.80 |
| Trustworthiness / Integrity | 4.5 | 20% | 0.90 |
| Expertise | 3.0 | 15% | 0.45 |
| Transparency | 4.0 | 15% | 0.60 |
| Consistency | 4.0 | 10% | 0.40 |
| Bias | 3.5 | 20% | 0.70 |
Weighted total 3.85 ÷ 5 × 100 = 77, which falls in the Moderately credible band.
What the score means
What every review shows
- The overall score and its credibility band.
- All six criterion scores, so you can see where a source gained or lost points.
- A summary and a detailed review explaining the reasoning.
- Where relevant: fact-check sources you can follow yourself, potential blind spots and key takeaways.
Academic basis
Derived from information science frameworks including Ohanian (1990), Metzger & Flanagin (2013), Fogg (2003), McKnight & Kacmar (2006), Heuer et al. (2024), and others.
Think we got a score wrong? Tell us.