DR-301e · Module 2
Cross-Source Contradiction Patterns
3 min read
Contradictions reveal patterns when analyzed in aggregate. If Source A and Source B consistently disagree on revenue figures, with Source A always reporting higher, the pattern reveals a systematic bias in one or both sources. If three analyst firms agree on a company's trajectory but the company's own filings contradict the consensus, the pattern suggests either analyst groupthink or strategic misdirection by the company. Cross-source contradiction patterns are meta-intelligence — they tell you not just what the data says, but how reliable your sources are relative to each other.
Do This
- Track which source pairs produce the most contradictions — persistent disagreement reveals systematic bias
- Analyze whether contradictions cluster in specific domains — a source may be reliable on technology but unreliable on financials
- Use contradiction patterns to update source reliability scores — repeated inaccuracy should degrade the score
Avoid This
- Treat each contradiction as an isolated event when a pattern exists
- Assume a source is uniformly reliable or unreliable — reliability can be domain-specific
- Ignore persistent contradiction patterns because each individual instance was resolved