The left pie of Figure 2 is redrawn to scale. The wedges are ordered by size, starting at 12 o’clock and going clockwise. Thanks to Richard Heiberger for suggesting this chart for my comments.

Sample Size Neglect is a cognitive bias studied by Amos Tversky and Daniel Kahneman. It consists of drawing false conclusions from statistical information, due to having not considered the effects ...

So the formula I use is (sample/600) ^ 0.5 e.g. the poll’s sample size, divided by 600, and taken to its square root.