Hacker News new | past | comments | ask | show | jobs | submit
Group level differences are of little to no value when evaluating a specific candidate.

It is widely understood and accepted that males and females differ in their physiology in ways that have dramatic impacts on their capabilities. However, the two groups form overlapping bell curves, and if you're seeking someone for a task you'd be a lot better off focusing on the attributes of the individual, which may be at either end of their group bell curve or anywhere in between.

Put differently, my wife, when she was a serious triathlete, would never have been able to beat the best males at any distance. But she could beat most of the males in a half ironman. So if you were interviewing her and some male to do something like a half ironman, you'd better make sure you ask a lot more than "what sex are they?". You'd better find out if the male is in the top X%, because if not, you should be hiring her instead.

All of that is true despite the group differences being real and significant.

Hiring is never about groups ... unless you're a racist/sexist/*ist ...

> Group level differences are of little to no value when evaluating a specific candidate.

Somehow that doesn't go for trying to determine how (dis)advantaged someone is though?

I mean, all of this is obvious. Group-level differences will still lead to the composition of individuals in a given profession differing from the composition of the general population, even if no hiring managers discriminate.
That's not necessarily true, for many kinds of complex sociological, economic and demographic reasons. The nature of the working population is different than the general population. The skills required from the working population vary across time and space, and may very well consist of a set in which different groups vary only slightly. Etc. etc.

Frankly this just reads like a cover story for "I don't want to have to care about this".

Well yeah, it’s an extremely complex system. If we’re just going to leave it at that, then you have no basis for insisting that working populations are proportional to the general population. But you seem to want to have it both ways.