Reading beyond the headline: a field guide to international news

A headline is an invitation to read, not the full evidence for a claim. Before sharing an international story, look for what the report establishes, what remains uncertain and where its information comes from.

The aim is not to distrust everything. It is to match your confidence to the quality of the evidence.

Read past the opening

Start with the opening paragraphs, but do not stop there if a claim matters. Look for qualifications such as “proposed,” “preliminary” or “according to officials.” A proposal is not an enacted law, and an attributed allegation is not an established finding.

Read the passages that explain how the outlet knows. If the headline seems broader or more certain than the body, use the narrower claim supported by the reporting.

Check when and where

Look for publication and update times, and distinguish them from the date of the event. An old report can be genuine without describing what is happening now. If you cannot establish the timing, avoid presenting it as current.

A dateline provides location information under an outlet’s conventions; it is not a complete map of how the story was gathered. Reuters’ journalistic standards require on-the-ground reporting for its datelines and identify additional contributors separately. Check the byline, reporting credits and attribution rather than treating a dateline alone as proof of reliability.

Notice who is speaking

Attribution lets you distinguish the reporter’s verified findings from someone else’s assertions.

• Named sources: Look for relevant knowledge and access to the events. A name alone does not prove a statement.
• Anonymous sources: Check how the outlet describes their knowledge and why their identity is withheld. Anonymity limits what readers can independently assess.
• Official statements: Treat them as evidence of what an institution says, not automatic confirmation that its account is complete.
• Interested parties: Ask what evidence supports their figures and whether anyone independent has checked them.

Reuters’ standards favor named sources, context about unnamed sources and cross-checking. Those are useful things for readers to look for, not guarantees that every published claim is correct.

Identify the kind of content

Check whether you are reading reporting, analysis, opinion or paid material. Do not assume a missing or hard-to-find label means a piece is straight news.

News reporting seeks to establish facts; analysis interprets them; opinion advances a viewpoint. Each can contain factual claims that need evidence. The University of Washington’s source-evaluation guide encourages checking purpose, authority and accuracy rather than judging a source by appearance alone.

Put numbers in context

Ask what a number measures, where it came from and what comparison makes it meaningful. Check the time period, geographic coverage and denominator. A percentage without a starting value can conceal as much as it reveals.

For money, check the currency and whether figures being compared use consistent units. Do not let a conversion distract from the underlying scale or method.

Look for missing evidence and independent checks

Has the subject of a serious allegation had a chance to respond? Are documents linked, methods explained and uncertainties acknowledged? Berkeley’s evaluation guidance offers questions about authority, evidence, relevance and purpose.

Compare credible coverage, but trace it back. Three sites repeating one wire report or one official statement are not three independent confirmations. Seek separately gathered evidence rather than counting matching headlines.

Give updates and corrections weight

For a developing event, check what later reporting confirms or changes before sharing the most dramatic version. A visible correction is evidence of an accountability practice, not proof that an outlet never makes mistakes.

Look for a clear correction policy and specific explanations of changes. If a claim remains unsupported, the responsible conclusion may simply be that it is not yet established.