Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The number increments get more significant the further from 100 you get. 75 IQ is under the fifth percentile -- definitely not common.


I recognize that this is kinda subjective, but I'm a bit baffled that someone wouldn't find 1-in-20 to be "common".

Though it's not homogenous, that's very conservatively the equivalent of one child in every classroom. It's millions of people, even in a small country.


To me, "common" means roughly 50% or more, with less than that being some degree of "rare."

1-in-20 I would definitely call rare.


Your definition of "common" is out of step with most people's. It means not-rare, or widespread, for example jobs in trucking, retail, and tech are all common in America without that implying any of them make up 50% of the country's jobs. Or another example, in the UK the NHS classes probability of side effects into groups where "very common" means "more than 1 in 10" and "common" means around 1 in 10+ (with the next grade being "rare" for 1 in 100+).


>Your definition of "common" is out of step with most people's.

This is uncharitable. Adjectives can be extremely contextual, and "common" is very much one of those words. The biggest factors are relativity to other things in the same domain, and deviation from popular perception. E.g. if most instances of X in a given domain are ~1 in N, people will generally use words like "common" and "unusual" for Y if it is greater than or less than 1 in N, regardless of what N is. Or say Y is exactly 1 in N, but people at large misestimate it at 2 in N. Now you might say it's "rare" despite whatever its relativity to the domain is. Now note that X and Y usually belong to effectively infinite domains, throw in the ambiguity of conversation, and it gets very fuzzy. This is basically a long way of saying that this is a semantic argument, and therefore may be divorced from the original implication of the adjectives being contested.


I once saw a website I think where histograms of each of such words were plotted, can't find it anymore though


You're probably thinking of this HBR article by Andrew and Michael Maboussin

https://hbr.org/2018/07/if-you-say-something-is-likely-how-l...

which is an updated take on the methodology of Dick Heuer's Psychology of Intelligence Analysis

https://prasadcapital.com/2018/07/03/book-summary-psychology...

, which itself was driven by the research into words of estimative probability by CIA legend Sherman Kent.

https://en.wikipedia.org/wiki/Words_of_estimative_probabilit...


Yes!


Yes, it's 4.77, about 1/20.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: