How to Evaluate Sources of Information: A Better Method Than Checking the About Page
A polished website can still be a poor source
Professional design is cheap.
A logo, citations and an impressive About page can be created by reputable institutions, commercial marketers, advocacy groups and misinformation sites alike.
That is why modern source evaluation increasingly emphasises lateral reading: leave the page and investigate the source using information from elsewhere.
Research associated with Stanford's Civic Online Reasoning project found that professional fact-checkers were faster and more effective than many students and academics at evaluating unfamiliar sites because they did not spend long inspecting the site itself. They opened new tabs and investigated reputation, ownership and outside evidence.
1. Identify what kind of source you are looking at
Before judging quality, identify function.
Is it:
a government statistical release;
a peer-reviewed paper;
a company filing;
a news report;
an editorial;
a think-tank analysis;
an advertisement;
a personal blog;
a social-media post;
a database entry?
Different sources serve different purposes.
A company filing is strong evidence for what the company officially reported, but weak evidence that management's interpretation is unbiased.
A newspaper can explain a court ruling, but the judgment is better when exact legal wording matters.
2. Investigate the organisation laterally
Open a new tab and search the organisation's name.
Look for independent descriptions.
Who owns it?
How is it funded?
What is its stated mission?
Does it have subject-matter expertise?
Does it correct errors?
Is it known primarily for research, advocacy, marketing, satire or partisan campaigning?
The goal is context, not a purity test.
An advocacy group can publish accurate data. Its role simply tells you to inspect methods and framing carefully.
3. Investigate the author
Check whether the author has relevant expertise.
A Nobel-winning physicist is not automatically an expert in nutrition.
A celebrity is not a financial analyst because they discuss investing.
A lawyer may be qualified to analyse one jurisdiction while speaking loosely about another.
Expertise is domain-specific.
4. Trace claims to original evidence
If an article says “a study found,” find the study.
If it quotes official statistics, find the statistical release.
If it discusses a law, regulation or judgment, find the authoritative text.
If it cites another article, follow the chain until you reach the origin.
This prevents citation laundering, where repeated references make a claim look independently supported even though every link traces back to one weak source.
5. Check whether confirmation is actually independent
Five websites repeating the same press release are one evidence stream.
Ask whether sources gathered the information independently.
Independent confirmation is especially important for breaking news, allegations and scientific claims that have not yet been replicated.
6. Match the source to the claim
Source quality is contextual.
For current population data, an official statistical agency may be strongest.
For the effect of a treatment, a systematic review of high-quality studies may be stronger than one clinic's testimonial page.
For what a politician said, a full speech transcript or recording may be better than a paraphrase.
Do not ask only, “Is this source credible?”
Ask:
Is this source appropriate evidence for this particular claim?
7. Inspect methods, not just institutional prestige
Strong institutions can publish weak analyses.
Look at how the information was produced.
For surveys:
Who was sampled?
How many people?
How were questions worded?
For scientific studies:
What design was used?
Was the sample adequate?
What comparison group existed?
Are limitations acknowledged?
Methodological transparency is a stronger signal than reputation alone.
8. Check date and version
A source can be accurate and obsolete.
Software documentation, laws, prices, officeholders, medical guidance and economic statistics change.
Check:
publication date;
last update;
version number;
whether newer official information exists.
For historical questions, older sources may be exactly what you need. Recency is a requirement only when the claim itself is time-sensitive.
9. Read beyond the headline and excerpt
Headlines compress.
Abstracts compress.
Social posts compress even more.
A study may report an association while the headline says “causes.”
A report may describe one subgroup while the summary generalises.
Open the source far enough to inspect what was actually claimed.
10. Check conflicts and incentives without assuming corruption
Funding and incentives matter because they can influence which questions are asked and how results are framed.
But funding does not automatically make findings false.
Treat conflicts of interest as information that affects the level of scrutiny required, not as an automatic verdict.
A quick source-evaluation workflow
For a consequential claim:
Step 1: identify the source type.
Step 2: leave the page and investigate the source laterally.
Step 3: trace the claim to original evidence.
Step 4: check whether other evidence is independent.
Step 5: inspect method, date and scope.
Step 6: decide how much confidence the evidence justifies.
This method is faster than spending ten minutes studying fonts, logos and About pages.
What to do when sources disagree
Disagreement does not automatically mean both sides are equally strong.
Compare methods, directness of evidence, sample quality, recency and whether one source has better access to the underlying facts.
Sometimes the correct conclusion is that the evidence is genuinely unsettled.
Source evaluation should be capable of producing uncertainty rather than forcing a winner.
11. Distinguish authority from evidence
An authoritative source can save time because expertise and institutional processes are useful credibility signals. But authority should lead you toward the evidence, not replace it.
For example, an official health body may be a strong starting point for public guidance. If the question concerns a contested scientific mechanism, the underlying studies and reviews may still matter.
Likewise, a university affiliation tells you something about a researcher. It does not guarantee that every claim in a lecture, blog post or interview represents a research consensus.
Use authority as a reason to pay attention, then ask what supports the claim.
12. Use a source hierarchy without making it rigid
A practical hierarchy can speed research.
For legal wording, start with legislation and judgments. For company financials, start with audited filings. For scientific effects, start with systematic reviews and major peer-reviewed studies. For current government policy, start with the responsible ministry or regulator.
But hierarchy is contextual. An eyewitness may be the best source for what they personally saw, while being a poor source for the wider cause of the event.
The strongest source is the one closest to the specific question and produced through a method capable of answering it.
Credibility is not binary
Sources are not simply trustworthy or untrustworthy forever.
A source can be reliable in one domain and poor in another. It can publish an excellent investigation today and correct an error tomorrow.
Good evaluation therefore produces calibrated conclusions:
strong evidence;
credible but incomplete;
uncertain;
weakly supported;
unreliable for this claim.
That is more useful than attaching a permanent label to an entire website.
Sources / Further Reading
Stanford Civic Online Reasoning, Civic Online Reasoning curriculum
Stanford Civic Online Reasoning, Teaching Lateral Reading
Suggested Internal Links
How to Question Information Effectively - Planned internal link
The Importance of Media Literacy - Planned internal link
How to Avoid Misinformation Online - Planned internal link
Understanding the Skill of Research - Planned internal link
Understanding the Difference Between Facts and Opinions - Planned internal link
