← Checking What You Hear, in Practice
Numbers and charts · 39 of 73
Ask who was asked
Some evidence
Do this
- See “most Nepalis think…” or “a survey shows”? Ask who was asked, and how they were chosen.
- Chosen at random from everyone, or just whoever saw the link and felt like answering?
- Ask how many answered, and when.
- An online poll anyone can join? Enjoy it, but don’t quote it as what people think.
Why it matters
Good surveys choose people at random, so that every group has a fair chance. Polls that people opt into can go wrong: Pew Research found about 4 to 7% bogus answers in some US opt-in online polls, against about 1% in random samples recruited by address, and those answers tilted the results a little. And a page’s followers are rarely like the whole country. Some evidence: Pew’s numbers are American, and we found none for Nepal.
Try it today
Find a poll on a page you follow. Write down who could have answered it, and who couldn’t. Three minutes.
Where these facts come from
- Assessing the risks to online polls from bogus respondents, Pew Research Center, 2020.
General information, not professional advice. No page can fit every situation. You use it at your own risk: see our terms. Spotted a mistake, or know something better? Tell us.
