How we research

Make the pathinspectable.

A conclusion is only as serious as the chain connecting source, transformation, interpretation, uncertainty and review.

EvidenceReasoningCorrection
An evidence laboratory containing records, samples, source cards and analytical instruments
Method / evidence before confidence

THE STANDARD

Intellectual ambition does not excuse evidentiary indiscipline.

The Foundation crosses disciplinary boundaries because reality requires it—not to escape the standards of any discipline. Every enquiry must make its sources, transformations, assumptions, uncertainty and institutional interests visible enough to be examined.

OPERATING PRINCIPLES

Eight obligations carried into every enquiry.

01

Primary sources first

Official datasets, regulatory filings, public records, academic literature and direct evidence wherever possible.

02

Source transparency

Claims should remain traceable to the evidence, transformations and assumptions that produced them.

03

Interdisciplinary analysis

Questions are examined across technical, economic, institutional, historical and social boundaries.

04

Quantitative + qualitative

Measurement reveals patterns; institutions, incentives, history and human behaviour explain their meaning.

05

Probabilistic reasoning

Uncertainty is represented explicitly rather than hidden behind false precision.

06

Red teaming

Material conclusions should be challenged through competing explanations and adversarial review.

07

Corrections

Substantive errors should be corrected visibly and without institutional defensiveness.

08

Independence

Funding may support inquiry. It must not purchase conclusions.

Researchers examining archival, computational and material evidence together
The evidentiary table / source before synthesis

ORDER OF EVIDENCE

Begin as near to the source as the question permits.

01

Primary evidence

Official records, original datasets, filings, legislation, judgements, technical documentation, field observation and direct testimony.

02

Scholarly record

Peer-reviewed research, authoritative monographs, archival scholarship and established technical literature.

03

Institutional evidence

Company disclosures, policy documents, professional records and material produced by accountable organisations.

04

Interpretation

Expert analysis, journalism and commentary used with attribution and never permitted to conceal the underlying evidence.

LANGUAGE OF CONFIDENCE

Keep uncertainty visible.

These categories are not mathematical probabilities. They are editorial disciplines intended to prevent presentation from implying more certainty than the record can support.

Known

Directly supported by evidence considered sufficiently reliable for the claim.

Probable

The best present account, with uncertainty and competing explanations disclosed.

Contested

Material evidence or interpretation remains unresolved between credible positions.

Asserted

A consequential claim exists in the field but has not been independently established.

CORRECTIONS & REVISION

A serious institution must be capable of changing its mind in public.

Material errors should be corrected visibly. Revised datasets, methods and publications should preserve version history where practicable. Changes affecting the interpretation of a finding should be distinguished from typographic or presentational corrections.

Authors remain responsible for accuracy. Review strengthens that responsibility; it does not replace it.

Submit a research correction