5 ms·

This comment reminds me somewhat of the Oxford and Cambridge interviewing processes. They have names and educational backgrounds to hand, and it's well known t

This comment reminds me somewhat of the Oxford and Cambridge interviewing processes. They have names and educational backgrounds to hand, and it's well known that all things being equal, tutors would always rather go for a kid from a state echool / underprivileged / non traditional background over one that went to a top public school. It's not just a matter of hitting targets - the alternative is that everyone who gets in is from a wealthy family or went to a top fee paying school. If everything is blind, you get less of an opportunity to engage in this kind of affirmative action.

manfredo 8y ago
It's a double edged sword. In tech, it's been found that blind hiring processes where candidates' name, education, and past employers are obscured (a third party broadly describes the education and work experience but doesn't give any explicit names) does result in more hires from less prestigious universities and companies. But it does make the hired population more male and more white & Asian. I think triplebyte did a study.
- taway90210 8y ago
  >> But it does make the hired population more male and more white & Asian Is that a reflection of bias or a reflection of the graduate hiring pool? If 90% of the CS graduates are male and your hiring process hires 90% male, is that bais or simply an extension of some problem upstream?
- depressedpanda 8y ago
  How is that in any way double edged? Or am I wrong in thinking that hiring policies should strive to be as meritocratic as possible, even if that means more white/Asian men?
  - manfredo 8y ago
    Well, in the Bay Area yes. Many do think meritocratic systems are bad because they often produce results that don't reflect the broader population. Some go so far as to say that advocating a meritocratic system is an act of racism and sexism. To me at least, I don't think it's fair to make the blanket statement that meritocracy is bad. It's all a question of values, namely whether someone values equality of the process itself or equaity of the process's outcome. Different people have different values so they'll inevitably have different opinions on meritocratic systems.
sct202 8y ago
It's kind of unfair to compare raw stats of kids who went to elite prep schools vs poor public schools. Like I was thinking about the amount of activities, tutors, private school my SO and I could pay for if we had a kid. And it was kind of insane to compare that versus what we grew up with, and how different our own HS resumes would have been with the same level of resources.
umanwizard 8y ago
I think you should point out what "public school" means in Britain, otherwise there is a 100% chance that any Americans who aren't aware of that definition will misunderstand your comment. (In the US, the term "public school" means what you mean by "state school", afaict).
Sacho 8y ago
> It's not just a matter of hitting targets - the alternative is that everyone who gets in is from a wealthy family or went to a top fee paying school. This doesn't follow. The alternative is getting whatever the ratio of <people from wealthy families>/<people from underprivileged/nontraditional backgrounds> is. So following your argument, the people from underprivileged backgrounds that meet the "all things being equal" test must be an insignificant amount.

manfredo 8y ago

It's a double edged sword. In tech, it's been found that blind hiring processes where candidates' name, education, and past employers are obscured (a third party broadly describes the education and work experience but doesn't give any explicit names) does result in more hires from less prestigious universities and companies. But it does make the hired population more male and more white & Asian. I think triplebyte did a study.

taway90210 8y ago
>> But it does make the hired population more male and more white & Asian Is that a reflection of bias or a reflection of the graduate hiring pool? If 90% of the CS graduates are male and your hiring process hires 90% male, is that bais or simply an extension of some problem upstream?
depressedpanda 8y ago
How is that in any way double edged? Or am I wrong in thinking that hiring policies should strive to be as meritocratic as possible, even if that means more white/Asian men?
- manfredo 8y ago
  Well, in the Bay Area yes. Many do think meritocratic systems are bad because they often produce results that don't reflect the broader population. Some go so far as to say that advocating a meritocratic system is an act of racism and sexism. To me at least, I don't think it's fair to make the blanket statement that meritocracy is bad. It's all a question of values, namely whether someone values equality of the process itself or equaity of the process's outcome. Different people have different values so they'll inevitably have different opinions on meritocratic systems.

sct202 8y ago

It's kind of unfair to compare raw stats of kids who went to elite prep schools vs poor public schools. Like I was thinking about the amount of activities, tutors, private school my SO and I could pay for if we had a kid. And it was kind of insane to compare that versus what we grew up with, and how different our own HS resumes would have been with the same level of resources.

umanwizard 8y ago

I think you should point out what "public school" means in Britain, otherwise there is a 100% chance that any Americans who aren't aware of that definition will misunderstand your comment. (In the US, the term "public school" means what you mean by "state school", afaict).

Sacho 8y ago

> It's not just a matter of hitting targets - the alternative is that everyone who gets in is from a wealthy family or went to a top fee paying school. This doesn't follow. The alternative is getting whatever the ratio of <people from wealthy families>/<people from underprivileged/nontraditional backgrounds> is. So following your argument, the people from underprivileged backgrounds that meet the "all things being equal" test must be an insignificant amount.