I think the terms you're looking for are local maxima and minima.
Re:Dutch men confirmed as world's tallest
by SideburnsOfDoom on Tuesday July 26, 2016 @05:17AM
(#12164333)
I have looked at the reference for this (1) and I don't think it captures the idea that statistically, small groups vary more.
Ah, I think I'm with you now. My stats is rusty, but I think what you're asking about is one of the fundamentals of variance.
I think this[1] Stats StackExchange answer frames it well.
[1]: http://stats.stackexchange.com/questions/129885/why-does-inc...
Re:Dutch men confirmed as world's tallest
by SideburnsOfDoom on Tuesday July 26, 2016 @07:34AM
(#12164846)
That is helpful, thanks. Especially this part:
> I believe that the Law of Large Numbers explains why the variance (standard error) goes down when the sample size increases. https://en.wikipedia.org/wiki/Law_of_large_numbers
So what I was looking at is the inverse of the law of large numbers.
Is there a name for that phenomenon? I saw a similar thing in some data that we were looking at recently, and it took a while to work out that the reason why one group's number was consistently a bit smaller than another group's number was that we were simply taking the max of a smaller group. it's obvious in retrospect, but can catch you out if you're not expecting it.
I'd like to have a name for it, a reference to send people to.