How we research

How we research

Our standards for sources, examples, platform facts and corrections, written down so you can hold us to them.

Which sources do we use?

We prefer primary sources: peer-reviewed research, the original documentation of a method (for example, the SIFT routine and the Stanford lateral reading research), official statistics agencies, and a platform's own documentation and open-source code. Secondary coverage from established outlets is used to confirm dates and describe changes, and we say so when we rely on it. Every chapter lists its sources at the end.

How do we handle facts about X?

Platform features change often. Anything about how X works, such as Community Notes eligibility, rating thresholds or account labels, is checked against X's published documentation or reputable reporting and phrased "as of November 2026". When reports disagree, we say that too. If you spot something out of date, tell us.

Why are all the examples invented?

Real viral posts come with real people attached, and a guide that picks real targets quickly becomes a guide that picks sides. Every sample post, account name, study and number on this site is made up for teaching, unless a sentence clearly cites a real source. The fallacy quiz, the claim-check sample and the worked examples in chapters all follow this rule.

What do we never publish?

  • Reviews, testimonials, user counts or success stories.
  • Verdicts on named people, parties, groups or real posts.
  • Claims that a specific account is a bot.
  • Medical, legal or financial advice. Number tools are arithmetic for education.

How do corrections work?

Send corrections to contact@twitterforskeptics.com with the page address and a source. We fix errors quickly, update the page's "updated" date, and do not quietly change meaning.

How are the tools built?

All four tools are plain JavaScript that runs in your browser. They make no network requests with your text, call no external APIs, and store nothing unless you press Save in the worksheet, which uses your own browser storage.