Skip to content

R0050/2026-03-31-02/Q001/SRC05/E01

Research R0050 — Journalism Disciplines
Run 2026-03-31-02
Query Q001
Source SRC05
Evidence SRC05-E01
Type Factual

Bellingcat uses structured investigation phases and minimum 3-source corroboration

URL: https://yemen.bellingcat.com/methodology/collection-protocol

Extract

Bellingcat's investigation methodology includes:

  1. Collection phase: Gather primary and secondary source media using text, source, and image-based searches
  2. Analysis phase: Conduct geolocation, software analysis, corroboration, and content descriptions
  3. Integration phase: Combine with supporting evidence from NGO reports and media reports

Key procedural standards: - Cross-reference information with "at least 3 sources" using multiple verification techniques - "A biased or non-credible source can still post content that can be independently verified" — content is assessed on a case-by-case basis regardless of source reputation - Show investigative work in categories of analysis (geolocation, software analysis, corroboration)

Key finding: Bellingcat has a structured investigation process (collect-analyze-integrate) and a minimum corroboration threshold (3 sources), but no formal evidence quality hierarchy, no calibrated uncertainty language, no structured bias checklist, and no source reliability tiering. Notably, Bellingcat explicitly rejects blanket source reliability tiering — a biased source can still provide verifiable content.

Relevance to Hypotheses

Hypothesis Relationship Strength
H1 Contradicts No formal scoring systems; explicitly rejects source-level reliability blanket ratings
H2 Supports Structured phases and minimum corroboration thresholds exist, even without formal scoring
H3 Contradicts The three-phase process and 3-source minimum are clearly structured methodology elements

Context

Bellingcat's approach is notable for its emphasis on content verification over source reputation — a philosophical departure from intelligence community approaches that tier sources by reliability. This is a deliberate methodological choice reflecting OSINT practice.