A survival skill for navigating our world
Misinformation, fabricated news stories, deepfakes, TikTok debates, hyperpartisan news sites and misleading political claims are part of the information we encounter every day. Source evaluation is more than an essay-writing skill—it is a survival skill for navigating our world. It helps you check what is accurate, recognise how information is framed and decide what deserves your trust.
Objectivity means applying the same standards of evidence and reasoning to your own beliefs—and sources you agree with—as to those you disagree with. You do not have to reach the same judgement about every side. You have to use the same standards to judge them.
1. Is this even real, or remotely true?
Fact-check the claim and examine the source. Is the material genuine? What evidence supports it? Who is behind it, and what are they trying to achieve?
Is it real?
Check the account, quotation, image or video.
Find the original post or full recording. A screenshot may be edited; a clip may be altered or generated. Even genuine footage can mislead when it is old, cropped or given a false caption.
Is the claim supported?
Find the evidence behind what is being said.
Check the date and context. Does the evidence actually support the claim? Look for independent confirmation: several sites repeating one report are still relying on one source.
Look beyond the page
Open a new tab and investigate the author or organisation. What do other credible sources say about its expertise, ownership and funding? Its own About page is a useful lead, not independent confirmation.
Use fact-checks—but check them too
Look for a fact-checker that shows its evidence, explains its methods, discloses its funding and corrects mistakes. Read the reasoning behind its verdict, and notice which claims it chooses to check.
A checkable claim is not necessarily true. Opinions need reasons; predictions need evidence and uncertainty. If someone says one thing caused another, ask whether other explanations have been considered.
2. Trace the evidence—even behind AI and TikTok
Check the sources behind AI answers and fact-checks too. Is the information coming from an original study, a news report, a Reddit comment—or something else?
Follow the claim back
Answer, video or post → cited source → original evidence → your judgement
Open the source. Does it exist? Does it say what is claimed? Is it suitable evidence for this question?
Can a TikTok video be a good source?
Yes, it can contain accurate information. When it makes a factual claim, go to the source of the video’s information. A confident speaker, lots of likes or a convincing presentation does not replace evidence.
A Reddit thread may contain useful leads, or even a firsthand account. Check who is speaking, what they can know and whether their claims can be confirmed. The platform alone does not settle reliability.
What about AI?
AI can invent references or misrepresent real ones. A citation that looks convincing is not enough: find the source and read it. Asking AI to confirm its own answer is not independent fact-checking.
If you cannot trace a claim to reliable evidence, it remains unverified.
For an image, try reverse image search. For video, search a clear still frame and look for the original upload. Compare the event date with the upload and repost dates. Search results can reveal earlier uses, but do not by themselves prove authenticity.
3. Beyond accuracy: bias, framing and agenda
Even when the facts are accurate, how is the information framed? Does the source favour a particular viewpoint? Does it have a political agenda—something it wants you to believe or support?
Bias is a tendency
Information may be selected or presented in ways that favour a viewpoint.
Bias can be unconscious. Look at what is emphasised, downplayed or left out, and whether opposing claims receive the same scrutiny.
An agenda is a goal
A source may aim to influence beliefs, decisions or action.
Look for evidence of its aims: stated positions, campaigns, ownership or funding. A slanted headline alone does not prove intent.
Bias and agenda can overlap, but they are different. Neither automatically makes a claim false. A neutral tone does not automatically make it reliable.
A quick example · fictional
A reviewer reports eight hours of headphone battery life, then calls them “the perfect choice for every student”. The reviewer earns commission from sales.
The measurement may be accurate. The recommendation is a judgement, and the commission is an incentive to consider. Check the test conditions and independent tests before accepting either conclusion.
Notice your own response too: would you question the same omission, emotional appeal or unsupported claim if it came from a source you usually trust?
How is it trying to persuade you?
Logos · reasoning and evidence
Does the evidence support the conclusion?
Examine statistics, comparisons and the steps in the argument. Numbers can look convincing while leaving out context or supporting a narrower claim.
Pathos · emotion
Is it appealing to fear, anger, sympathy or pride?
Does the emotion help you understand what is at stake, or does it replace evidence and discourage questions?
Ethos · credibility
Why should you trust the speaker?
Look for relevant expertise and a record of honesty. Confidence, popularity or an impressive title does not establish expertise in every subject.
These appeals are not inherently misleading. Ask whether they strengthen a supported argument or make a weak claim seem convincing.
4. Consider the Media Bias Chart
Explore the Ad Fontes Media chart ↗
Does the outlet’s placement help explain what you noticed—and what can it tell you about this particular article?
Compare political bias on the horizontal axis with reliability on the vertical axis. These are separate judgements.
How it works—and its limits
Ad Fontes combines ratings of sampled articles or episodes into outlet scores. Its published method uses at least three analysts per item, with different self-reported political leanings.
A sample does not cover every article. Human ratings involve interpretation. Ad Fontes is a commercial company selling data and subscriptions; evaluate its methods and incentives as you would another source.
The chart is rooted in US news coverage, but now includes international sources. Ad Fontes says it evaluates international political bias in the relevant country’s context. Do not assume left and right mean the same thing in Sweden and the US.
Explore the chart
- What patterns or imbalances do you notice? Compare reliability on the left and right, as well as political bias.
- Where are the sources you regularly encounter on social media? Search for a few. Are they high or low on reliability, and how politically biased are they rated?
- Are the sources you see most often also the most reliable? What might explain any difference?
- What could explain an imbalance? Examine individual article ratings and the chart’s methods before deciding.
- Does anything challenge your expectations? Apply the same scrutiny to sources you usually agree with.
If a creator or account is not rated, trace its claims to the sources it uses. Being absent from the chart does not establish reliability or unreliability.
Read the methodology and its explanation of international ratings. A placement is background context, not a substitute for checking claims.
5. Compare the headlines
Bias can appear in which facts are selected, whose voices are included and how a story is worded. Start by comparing what these headlines put in the foreground.
Judge the wording before looking at the source labels. Would the same headline seem fair if it described a politician you supported rather than one you opposed?
One event, three headlines
These headline quotations were published on 30–31 May 2024, around Donald Trump’s conviction on 30 May. Compare their framing before revealing the sources. This is a historical example, not a report on the case’s current status.
Headline A
“Guilty on all counts, Donald Trump becomes first US president convicted of a crime”
Reuters (straight-news comparison source)
Headline B
“Trump blasts ‘rigged, disgraceful’ trial after being found guilty”
Fox News (leans right)
Homepage headline recorded by Associated Press
Headline C
“Trump’s ‘Team of Felons’ adds the most important member of all”
MSNBC (now MS NOW) (left-leaning)
Steve Benen · MaddowBlog commentary · 31 May 2024
The political labels describe broad outlet orientations, not accuracy verdicts. They are shorthand, not exact chart categories. Compare Fox’s and MS NOW’s ratings and methods. Straight-news reporting still needs scrutiny.
- Focus: What comes first—the verdict, its significance, or someone’s reaction?
- Language: Which words shape your emotional response?
- Attribution: Is the headline stating a fact or quoting someone’s assessment?
- Omissions: What useful context is absent? Is that explained by limited space, the article’s purpose, or a pattern of selective coverage?
- Clickbait: Does the headline communicate the news clearly, or does it withhold key information or sensationalise the story to make you click?
The links identify the published articles; AP’s contemporary report records the Fox homepage wording. These are historical headlines, and online versions may change.
Check your understanding
Use the evidence supplied, rather than whether you agree with the source’s position. Be ready to revise your judgement when the evidence changes.
Use the supplied evidence to choose the best-supported evaluation. These are fictional scenarios. You can check any selected answers.
Your answers are checked in your browser and are not sent to your teacher.