Which Of The Following Is Are True

You're staring at a multiple-choice question. So naturally, " Your palm sweats. Now, four options. Worth adding: one is right. Maybe "all of the above.Now, maybe two. You've read the passage three times and still can't tell which statement actually holds up.

Sound familiar? Still, that feeling — the gap between looking* true and being* true — doesn't vanish after school. It just moves to your inbox, your feed, your group chats, and the articles you share before reading.

What Is Claim Evaluation Really

At its core, evaluating whether something is true means separating verifiable claims from opinions, assumptions, and noise. Practically speaking, a verifiable claim can be checked against evidence. "Water boils at 100°C at sea level" — testable. "This policy will destroy the economy" — a prediction wrapped in hyperbole, not a fact you can look up.

Most of what crosses your screen daily isn't a clean true/false proposition. It's a mix: a real statistic framed misleadingly, a genuine quote stripped of context, a correlation presented as causation. The skill isn't spotting lies. It's spotting structure* — seeing how a claim is built and whether the foundation holds.

The Three Buckets Every Claim Falls Into

Empirical claims — statements about the world that evidence can confirm or refute. "Unemployment is at 3.8%." "This drug reduces symptoms in 60% of patients." These live or die by data.

Analytic claims — true by definition or logic. "All bachelors are unmarried." "2 + 2 = 4." No experiment needed. But watch out: people often dress up empirical claims in analytic clothing ("It's just common sense that...").

Normative claims — value judgments. "We should* raise the minimum wage." "That movie was bad." These aren't true or false in the evidentiary sense. They're arguments about priorities. Confusing them with empirical claims is where most fights start.

Why It Matters More Now Than Ten Years Ago

The volume of claims you encounter has exploded. Still, the friction to publish has vanished. The incentives to distort — clicks, engagement, algorithmic reward — have sharpened.

A 2021 MIT study on Twitter (back when it was still called that) found false political claims spread faster and reached more people than true ones. Because false claims are often more novel*, more emotionally charged, and simpler to package. Not because people are stupid. Truth is frequently messy, conditional, and boring.

And the cost of getting it wrong isn't just a bad grade. It's health decisions based on a misread study. On top of that, financial moves driven by a viral thread that skipped the methodology. Votes cast on a claim that collapsed under scrutiny three days later — but the correction never went viral.

How to Actually Check a Claim

You don't need a research degree. You need a checklist you actually use.

1. Isolate the Core Claim

Strip the rhetoric. And " The core claim is testable. "Experts warn that new regulation will crush small businesses" becomes: "Will this regulation reduce small business revenue/survival rates?The wrapper ("experts warn," "crush") is persuasion.

Write it down. That said, one sentence. If you can't, the claim is too vague to evaluate.

2. Ask: What Would Convince Me Otherwise?

Before hunting evidence, define your falsification criteria. "I'll believe X if a randomized controlled trial shows Y.Now, " "I'll change my mind if three independent agencies report Z. " This prevents moving goalposts later.

If nothing could change your mind, you're not evaluating. You're defending.

3. Trace the Provenance

Where did this originate? Not who shared it — where was it born*?

• A peer-reviewed paper? Check the sample size, p-values, replication status, conflicts of interest.
• A government report? Look for the methodology appendix. Political appointees sometimes rewrite summaries.
• A think tank? Funders matter. "Independent" often means "funded by interests aligned with this conclusion."
• A blog post citing a study? Click through. Half the time the study says the opposite.
• A screenshot of a tweet? Assume it's fabricated until you find the original.

4. Check the Context Frame

A stat without a denominator is a prop. Here's the thing — "50 people died" — out of how many? Over what period? Compared to what baseline?

"Crime doubled" — from a historic low to still-pretty-low? Also, or from high to crisis? The frame decides the meaning.

Relative risk vs. absolute risk is the classic trap. In practice, "Drug X increases heart attack risk by 50%! " Absolute risk went from 0.002% to 0.003%. Technically true. Practically meaningless.

5. Look for the Missing Comparison

Every causal claim implies a counterfactual. "Since the policy passed, outcomes improved.The pre-existing trend? A similar region without the policy? " Compared to what? A synthetic control?

For more on this topic, read our article on how many seconds are in 5 days or check out organisms that produce their own food.

If the piece doesn't address the counterfactual, it's not analysis. It's storytelling.

6. Verify the Expertise Match

A Nobel physicist commenting on epidemiology is not an expert source. Because of that, a cardiologist opining on macroeconomics is not an expert source. Credentials are domain-specific.

Also: consensus > lone genius. If 97% of relevant specialists agree and one dissenter gets amplified, ask why that* voice reached you.

7. Run the "Steel Man" Test

Can you articulate the strongest version of the opposing case — without straw-manning? If not, you don't understand the claim well enough to evaluate it. You're just picking a team.

Common Mistakes / What Most People Get Wrong

Mistaking citation for verification. A footnote doesn't make a claim true. Check the cited source. Often it doesn't say what the author claims.

Confusing "peer-reviewed" with "replicated." Peer review catches obvious errors. It doesn't guarantee the finding is real. Replication does. Many landmark studies — power posing, ego depletion, priming effects — passed peer review and failed replication.

Treating "no evidence of harm" as "evidence of no harm." Absence of evidence is not evidence of absence. This shows up constantly in health and environmental debates.

Overweighting anecdotes. "My uncle smoked three packs a day and lived to 90" tells you nothing about population risk. Humans are wired for narrative, not statistics. Fight the wiring.

Ignoring base rates. A test for a rare disease is 99% accurate. You test positive. Your chance of actually having the disease? Still low — because the disease is rare. Base rate neglect makes people panic over false positives and ignore real risks. Not complicated — just consistent.

Believing corrections backfire. The "backfire effect" — where corrections strengthen false beliefs — is real but rare*. Most people update when presented with clear evidence, especially from trusted sources. Don't skip the correction because you heard it "doesn't work."

Practical Tips / What Actually Works

Build a personal "trusted sources" list — but keep it short. Two or three outlets per domain (health, economics, tech, policy) with transparent corrections policies, clear methodology, and ideological diversity. Rot

Rotate them annually. Familiarity breeds trust, and trust breeds blind spots.

Use the "24-Hour Rule" for breaking news. The first draft of history is usually wrong. Early reports conflate speculation with fact, confuse correlation with cause, and amplify the loudest voices, not the most informed ones. Wait. The signal emerges from the noise only with time.

Check the funding and the incentives. Who paid for the study? Who benefits if you believe the claim? Who loses? "Follow the money" sounds cynical until you realize it’s just basic experimental design: identify the confounding variable.

Learn the handful of statistical concepts that do 90% of the work. Statistical significance vs. practical significance. Relative risk vs. absolute risk. P-hacking. P-value thresholds. Confidence intervals. Selection bias. Survivorship bias. You don’t need a degree. You need a mental checklist.

Pre-register your own conclusions. Before you click the headline, ask: "What would I need to see to change my mind?" If the answer is "nothing," you’re not evaluating evidence. You’re defending identity.

Teach it to someone else. The Feynman technique works for media literacy, too. If you can’t explain why a study is weak or a claim is strong to a smart 15-year-old, you don’t actually understand it. You’re just repeating the abstract.

Conclusion

Information hygiene isn’t about becoming a cynic. Cynicism is lazy — it assumes everything is fake so you don’t have to do the work of sorting. Skepticism is disciplined. It assumes some* things are true, and treats the gap between claim and evidence as the only place where understanding lives.

The tools above aren’t academic. They’re survival skills. The modern information environment is adversarial by design: algorithms optimize for engagement, not accuracy; bad actors flood the zone with plausible noise; and your own biology — confirmation bias, availability heuristic, tribal epistemology — is the exploit they target.

You cannot outsource your epistemic immune system. Because of that, no fact-checker, no platform label, no trusted brand will catch every error before it reaches you. The filter has to be yours*.

The goal isn’t perfect certainty. The goal is calibrated confidence: knowing how you know, knowing where* the uncertainty lies, and refusing to let the volume of the claim substitute for the weight of the evidence.

That’s not media literacy. That’s intellectual adulthood. And in a world drowning in assertion, it’s the only life raft that holds.