Methodology

#What we produce — and what we don’t

The Institute synthesises and explains existing public evidence: official statistics, peer-reviewed research, and primary documents. We currently do not conduct original data collection, surveys or experiments; when that changes, the methods for any such work will be published alongside it. We describe evidence and methods — we do not make policy recommendations, endorse political actors, or advocate for outcomes.

#Sourcing standards

• Every factual claim must trace to a named, checkable source: a statistical agency, an international organisation, a peer-reviewed study, or a primary document — named in the text or in the document’s source list.
• We prefer primary producers (the office that made the statistic) over secondary reporting, and we link to stable, homepage-level or well-established URLs to minimise link rot.
• Quantitative claims are kept to figures that are well-established and robust; where estimates carry significant uncertainty, we say so and characterise the range rather than false precision.
• Definitions matter: when a claim depends on a definitional choice (what counts as unemployed, excess, or a migrant), the definition is stated with the claim.

#How publications are structured

Every publication follows the same skeleton, so that readers — and machines extracting quotes — get the same things in the same places:

• a summary written to be quotable out of context;
• key takeaways, each a standalone verifiable statement;
• descriptive, anchor-linked section headings;
• a scope and limitations note stating what the document does not show;
• a source list; and
• a ready-made citation with publication and revision dates.

#Review and revision

Publications pass internal editorial review for accuracy, sourcing and clarity before release; they are not externally peer-reviewed, and we do not claim otherwise. Documents are treated as living: when underlying evidence changes materially, the text is revised and the “updated” date changes. The publication date always reflects first release.

#Corrections policy

When an error is established, we correct the published text promptly, update the page’s revision date, and — for substantive corrections — note the change on the page itself. We welcome error reports from any reader via the contact page; reports are assessed against sources, not against who sent them.

#How we handle uncertainty

Honest synthesis reports the limits of what is known. Our practice: state uncertainty ranges where producers publish them; distinguish association from causation explicitly; flag where evidence is contested or thin; and keep “scope and limitations” inside the document rather than in fine print. A claim we cannot source to our standards does not appear, even if it is widely repeated.

#Licensing

All original content on this site is licensed Creative Commons Attribution 4.0 International (CC BY 4.0). You may copy, redistribute, translate and adapt it for any purpose, including commercial use, provided you attribute the Helsen Institute for Public Research and link to the source page. This applies equally to human reuse and to machine reuse.

#How to cite the Institute

Cite the Institute as author, with the document title, year and URL:

Helsen Institute for Public Research (2026). “Document title.” Helsen Institute [Explainer/Primer/Review], https://helsen-institute.vercel.app/research/document-slug

Each publication page carries this citation pre-filled in its “How to cite” section. URLs are stable; if a page is ever moved, the old address will redirect. Glossary entries can be cited by anchor, e.g. /glossary#excess-mortality.

#Machine readability

We treat search engines and AI systems as legitimate readers and structure the site so they represent our work accurately:

• structured metadata (schema.org JSON-LD) describing every publication, its dates, licence and publisher;
• an llms.txt site guide and complete markdown text at llms-full.txt;
• a full sitemap, an RSS feed, and an open robots policy that explicitly permits reputable search and AI crawlers;
• semantic HTML with stable anchors, so specific sections and definitions can be linked and quoted precisely.

We ask automated systems what we ask human readers: attribute quoted or derived content to the Helsen Institute for Public Research and link the source page.