Fact-Checking Policy

What we mean by fact-checking

Fact-checking at RTD JOURNAL is not a separate department that stamps a story on its way out. It is what reporting is, done properly: every claim of fact that carries weight in a story is checked against a source we can point to before the story runs. This page describes how that works, so you can judge the process rather than take the result on trust.

The core idea is simple. A claim is only as good as the evidence behind it, and the best evidence is the primary source: the original document, the original data, the person who was actually there. We work back toward that source rather than resting on somebody else’s summary of it, because summaries drift and the drift always seems to help whoever is doing the summarising.

Verifying a claim against the primary source

When a story says a court ruled a certain way, we read the ruling, not the press release about it. When it cites a number, we find the report, the filing, or the dataset that the number comes from and check that we are quoting it correctly and in context. When a public figure is said to have said something, we go to the full transcript or recording, because a quote that is accurate to the word can still be a distortion if it is lifted out of what surrounded it.

Not every source is equal. An official record outranks a summary of it. A named expert speaking within their field outranks an anonymous one speaking outside it. A contemporaneous document outranks a memory of what it said. Where sources conflict, we do not simply average them or pick the convenient one; we say what each says, weigh their reliability, and, if the honest answer is that the matter is unresolved, we tell you that.

Claims we treat with extra suspicion

Some material is right to distrust on sight until it is checked. A screenshot with no link. A statistic with no source attached, which is often a statistic that will not survive being traced. A striking quote that is too perfectly shaped for the argument it is being used in. A photo or video that could be old, staged, or from somewhere else entirely. A claim that arrives already viral, because the fact that something spread tells you nothing about whether it is true.

For images and video we check where and when they were taken before we use them, and we do not present old footage as new or unverified footage as confirmed. If we cannot establish that a piece of media is what it claims to be, we do not run it, however compelling it looks.

Numbers, studies, and the fine print

Figures deserve their own caution because they carry an air of certainty they have not always earned. We check whether a number is a count or an estimate, whether it is preliminary or final, what it is being compared against, and whether the comparison is fair. A percentage with no base, a total with no timeframe, and a record with no like-for-like are all ways to mislead with true numbers.

With research, we look past the press release to the study itself: what it actually measured, how big it was, who funded it, and whether its authors claim what the headline claims on their behalf. A single early study is a signpost, not a settled conclusion, and we try to say which we are looking at.

When we cannot fully confirm something

Sometimes a story matters and the confirmation is not yet complete. We do not solve that by pretending to a certainty we do not have. We attribute the claim clearly to whoever is making it, we say what we have been able to verify and what we have not, and we label estimates and unconfirmed reports as exactly that. On a fast-moving story this labelling changes as the picture firms up, and our corrections policy explains how those changes are marked.

The alternative, publishing a shaky claim cleanly and hoping it holds, is how misinformation gets a respectable byline. We would rather show the seams.

Putting claims to the other side

Checking a fact is not only about documents. When a story turns on a contested allegation, part of verifying it is putting it to the person or body it concerns and taking their answer seriously. Their response can change what we publish, add a fact we lacked, or occasionally sink a claim that looked solid. This is a verification step, not a courtesy, and it sits alongside the rest of our editorial standards.

When a check fails after publication

Checking before publication reduces errors; it does not abolish them. When something we verified turns out to have been wrong, a source who misled us, a document we read too quickly, a number we misunderstood, we correct it openly on the page, dated and described, under our corrections policy. We also look at how the error got through, because a fact-checking process that never examines its own failures is not really a process. If you can show us that something we published does not hold up, write to support@rtdjournal.com with the story link and your evidence, and we will check your check as seriously as we should have checked our own.

Last updated: August 2026