Who we rely on

Why this page exists

It would be easy to let an answer read as though the site had worked something out. It has not. Somebody spent a decade following a cohort; somebody else pooled forty trials and argued about the statistics; a statistics agency published its methodology so that a figure could be checked at all. What we add is retrieval, synthesis and a structure you can argue with — which is worth something, and is not the same as doing the work.

So every answer carries a credit line, and this is the full register behind it. Where the evidence comes from is not a footnote to an answer. It is the answer.

What the tiers mean

Each source on an answer carries a tier. It describes what kind of document it is — how much scrutiny it has already survived — and not how much anyone likes the organisation that published it.

1. T1Systematic reviews, evidence synthesis, major scientific assessments, national statistics agencies.
2. T2Peer-reviewed primary research, government and IGO reports.
3. T3Expert bodies, established fact-checkers, quality journalism with named primary sourcing, preprints.
4. T4General news, trade press, aggregators without primary sourcing.
5. T5Blogs, advocacy organisations, social posts, press releases. Evidence that somebody holds a position — never that the position is right.

Two consequences worth stating outright. Preprints sit at tier 3, however good they are, because they have not been reviewed — that is exactly the distinction the ladder exists to draw. And a weak paper in a strong journal is still tier 2: the tier describes the process, and no rule about a website can read the paper. That is what the evidence strength, the caveat and the steelman are for.

This is not a list of approved sources

Nothing here is blocked and nothing here is required. An answer can and often should cite an organisation this register has never heard of — no list can hold every national statistics office or the right specialist body for every question.

A fixed roster of permitted sources would also be the wrong thing to build. It would bias what we look at toward a set of institutions we chose in advance, which is a distortion we would be poorly placed to defend — it is the very move this site asks you not to make, accepting a claim on the strength of who said it. The register labels sources. It does not choose them.

It does have one enforcement job. Where it knows a domain, it checks the tier the model claimed against the tier the domain actually warrants, and the weaker of the two wins. A press release cannot be relabelled as peer-reviewed research.

The register

139 organisations. Coverage is honestly uneven — strongest in health, climate and UK and US official statistics, thinnest outside the anglophone world. That gap is real, and it gets closed by adding to the list rather than by calling it complete. Suggest an organisation.

If we have used your work

We quote briefly, link to the original, and send readers to it rather than replacing it — an answer here is meant to be the thing that makes somebody go and read the source. If you believe we have misrepresented a finding of yours, that is the most serious mistake this site can make, and we would rather hear it from you than not. Tell us and quote the line.