In his very first talk, the anecdote was that he polled Swedish professors at his university, i.e. his colleagues, and got lower than 25%, and he consequently compared them to chimpanzees in the same way.

Really, the whole chimpanzee thing is for helping explain why "lower than X%" means "worse than random chance" and nothing else.

The important paragraph in the article was this:

> The fact that humans do worse than chimps shows the problem is not a lack of knowledge, but the result of having preconceived ideas, Rosling says - ideas that are years, or sometimes decades out of date.