Skip to main content

"A Bigger Sample Is a Better Sample"

This video presents the same text shown beside it, spoken and on screen. It adds nothing the text does not say.

State

Sample size cannot repair a biased sample: how members were selected governs whether a finding generalizes, and no quantity of the wrong people fixes it.

Show

A magazine poll of two million of its own readers describes its readers, not a country.

Watch for

Large biased samples are more persuasive and no more correct.

Builds on

Unlocks

  • Nothing yet depends on this.