How to Evaluate a Source Before You Trust It

CREDIBILITY CHECK FRAMEWORK ============================ 1. AUTHOR EXPERTISE - Who wrote it? Named author or anonymous? - Do they have credentials, relevant degrees, or a track record in this field? - Have they published elsewhere (peer-reviewed, reputable outlets)? - Is their expertise actually in the topic, or adjacent/borrowed? - Red flag: authority claimed via title only, no verifiable history. 2. PRIMARY EVIDENCE - Does the piece link to or quote original data, documents, studies, or raw observations? - Can you reach the underlying source yourself, or is it summarized secondhand? - Are numbers, quotes, and statistics verifiable against the original? - Red flag: claims with no source, vague references ("studies show"), or data that only exists inside the article. 3. INCENTIVES - What does the author/outlet gain if you believe this? (ad revenue, political agenda, product sales, reputation) - Is the piece sponsored, affiliated, or paywalled-to-sell? - Does the author have a financial or ideological stake in the conclusion? - Red flag: conclusion conveniently serves the author's interest; no disclosure. 4. DATES - When was it published, and when was the underlying evidence gathered? - Is the claim time-sensitive (statistics, medical guidance, legal facts) and still current? - Was it updated after major developments, or left stale? - Red flag: old data presented as current, or undated content. 5. CITATIONS - Are sources named precisely (author, title, outlet, date, URL)? - Do the citations actually support the specific claim, or are they decorative? - Do cited sources exist and say what the article claims they say? - Red flag: broken links, no links, or citations that contradict the text. 6. CORROBORATION - Do independent, unrelated sources reach the same conclusion? - Does the claim survive a search for counter-evidence, not just supporting hits? - Are the corroborating sources genuinely independent, or all quoting each other? - Red flag: only one source exists, or all sources trace back to a single origin. VERDICT SCALE: Strong / Plausible / Unverified / Doubtful / Reject Score each of the six areas pass/fail; two or more fails = downgrade one level. EXAMPLE ------- Claim: "Vitamin C megadoses cure the common cold" from a wellness blog by a self-described nutrition coach. 1. Author: self-described "nutrition coach" — no verifiable medical credentials, no publication history. FAIL 2. Evidence: cites "a study" but gives no author, journal, or year; no link to primary data. FAIL 3. Incentives: sells branded vitamin C supplements; conclusion boosts his product. FAIL 4. Dates: article from 2019, cites no recent trials; cold research has evolved. WEAK 5. Citations: one vague reference, unverifiable; fails to mention large RCTs showing no cure effect. FAIL 6. Corroboration: mainstream medical consensus (Cochrane reviews) says vitamin C does not cure colds, only marginally shortens duration. FAIL Verdict: REJECT. Five of six areas fail; the one plausible point (minor duration effect) is contradicted by the stronger consensus evidence the article omits. The correct claim, corroborated by independent systematic reviews, is "vitamin C may slightly shorten a cold, it does not cure or prevent it."