Classification defined only by examples — no operational criteria
'Hindutva hate' is defined through a curated list of examples with no operational definition that would allow independent classification. Borderline cases have no decision rule. A trained coder cannot replicate classifications using only the published report.
No codebook — classification criteria not disclosed
No codebook was published. No reliability testing was reported. The number of coders is not stated. Classifications cannot be verified or replicated by outside researchers.
No documented inclusion criteria for 312-organization list
The report claims to identify 312 organizations but does not document the search methodology, geographic coverage criteria, or inclusion and exclusion rules. There is no null data — no accounting of what was searched for but not included.
No equivalent scrutiny of opposing groups or incidents
The report monitors one side of documented communal conflicts without applying equivalent scrutiny to opposing organizations or incidents. The Swap Test fails: the methodology as described would not be applied symmetrically. This asymmetry is not acknowledged as a scope limitation.
⚑ 73% of citations trace to CSOH-affiliated organizations — undisclosed
India Hate Lab and Hindutva Watch are cited throughout as independent sources. All three organizations share the same founder, established through IRS 990 filings. This relationship is not disclosed in the report, the author bios, or the acknowledgments. 73% of endnote citations trace to this affiliated network.
Source posts not archived — independent verification not possible
The underlying incident data is not available for review. Source posts are not archived with accessible permalinks. Independent verification of even a small sample is not possible from the published report.
No funding disclosure — founder relationship not disclosed
Funding sources are not listed. The CSOH/IHL/Hindutva Watch shared founder relationship is not disclosed. IRS 990 filings document the governance overlap that the report's sourcing structure depends on concealing.
No limitations section — no denominator reporting throughout
The report contains no limitations section. Percentages are cited throughout pages 12–34 without sample sizes, making them uninterpretable. No engagement with prior critiques of the classification methodology.