Risk of Mass Atrocities in India
The 490% figure is the most consequential unverified statistic in the calibration corpus. The original text said 'reportedly' — a hedge that has been stripped in virtually every downstream citation. The USHMM's institutional authority has amplified an unverified figure into an established fact in policy discourse.
What This Report Is
The US Holocaust Memorial Museum’s Simon-Skjodt Center published this policy brief in 2024. It assesses the risk of mass atrocities in India. It uses a statistical model from Dartmouth College called the Early Warning Project.
What We Looked At
How a study is done determines whether you can trust what it found. The Citation Integrity Dashboard (which scores research methods, not conclusions) evaluated this brief as a Policy Report. That means it pulls together other people’s research instead of collecting its own data. We scored it on whether it accurately represents the sources it cites and whether readers can verify the claims it makes.
What We Found
The 490% problem. Verification Standards (a dimension that measures whether an outside reader can check the report’s claims) scored 5 out of 10. The brief’s most widely cited statistic says hate speech by Indian public figures “reportedly increased by 490%.” That word “reportedly” matters. It means the authors themselves were not sure where the number came from. They did not name the source, explain how “hate speech” was defined for that count, or describe the counting method. That single statistic has since appeared in Congressional testimony, UN submissions, and dozens of news articles — almost always without the “reportedly” caveat. The original authors flagged their own uncertainty. Everyone who cited them afterward dropped it.
Missing limitations. Counter-Evidence (a dimension that measures whether a report addresses challenges to its own findings) scored 5 out of 10. A brief this influential should tell readers what it cannot prove. It does not. The Early Warning Project model has been challenged in academic literature. Those challenges go unmentioned. India’s functioning democratic institutions — an active Supreme Court, opposition parties winning elections, a large free press — are not discussed as factors that reduce atrocity risk. The brief lists warning signs. It does not list reasons the warning signs might not lead where they point.
Strong institutional credibility. Transparency and Governance (a dimension that measures whether an organization’s funding, leadership, and structure are open to public review) scored 8 out of 10. The Holocaust Museum is a federally funded institution with public governance, named staff, and Congressional oversight. This is the strongest governance score among non-survey reports in our calibration set (a group of ten reports we scored first to test the rubric). The gap between this score and the verification score is the core finding: strong institutions can still circulate weak statistics.
Adequate scope alignment. Coverage Symmetry (a dimension that measures whether a report’s actual content matches what its title promises) scored 7 out of 10. The brief covers India-level atrocity risk, which matches its title. The Early Warning Project model covers many countries, suggesting the Museum applies similar methods elsewhere. An open question remains: does the Simon-Skjodt Center apply the same evidence standards to comparable cases of religious violence in Bangladesh, Pakistan, or Nigeria?
The Bottom Line
This report scored 6.2 out of 10 — an Adequate grade (which means usable with caveats, but with gaps present). No non-compensatory cap (an automatic score limit triggered by critical failures) was applied. Under a weighting scheme that emphasizes verification more heavily, the score drops to 5.6 — which falls into the Deficient range (which means significant gaps that compromise reliability). That instability matters. The report’s policy influence rests on specific statistics that cannot be independently verified. The score reflects methodology only. The report’s conclusions about risk in India may be correct even though its methods for one key claim have a gap.
Citation Context
How this report's findings have been cited or applied after publication. Severity reflects the gap between what the report establishes and how it was represented.
What was claimed: The Holocaust Museum documented a 490% increase in hate speech in India
What the report actually says: The brief said 'reportedly increased by 490%' — the source and methodology of this figure are unidentified
Congressional citations that attribute '490%' directly to USHMM strip the 'reportedly' caveat and present an unverified figure as institutionally established. This is the most consequential citation ecosystem failure in the calibration corpus by policy impact.
What was claimed: Independent empirical evidence of atrocity risk in India
What the report actually says: Policy brief synthesizing secondary sources — not primary empirical research
Citing this brief as independent empirical evidence misrepresents its document type. It synthesizes existing reports and statistical models; it does not conduct original research.
What was claimed: USHMM research documenting India atrocity risk
What the report actually says: Policy brief — not peer-reviewed research
Academic citations that do not disclose the document type misrepresent the brief's evidentiary status.
What was claimed: International institutional warning about BJP governance
What the report actually says: Risk assessment policy brief — Early Warning Project model produces probabilistic assessments, not predictions or political judgments
Use of USHMM's institutional authority for domestic Indian electoral purposes is a scope escalation the brief's authors could not have anticipated and would likely not endorse.
Organization Response
US Holocaust Memorial Museum, Simon-Skjodt Center has been invited to respond to this assessment. If a response is received, it will be published here in full and without editing.
Status: Pending