a few minutes times thousands of prospective applicants is a lot of time. Depending on where you believe the averages lie it could easily make economic sense to use the race heuristic for hiring.
Wouldn’t most group-based heuristics be useless when self-selection is involved?
Let’s assume we care about strength, and that men are significantly stronger than women (I’m cutting corners for simplicity’s sake).
If you’re looking for people to mug in the street, and all else being equal prefer a weak victim, then you’re better off targeting women than men—it’s a useful heuristic.
However, if you’re hiring people for a job that requires strength, people will only apply if they think they have a chance, then you should expect about the same distribution of strength among the man and women who apply—any sex-based heuristic is useless.
You’d think so, but from my experience screening resumes tons and tons of people who are not qualified or obviously don’t fit very simple criteria still apply to jobs. I wasted time reading resumes of people who shouldn’t have even applied, and I wasn’t even doing any interviewing.
Strength is also a lot less fuzzy than intelligence/capability in general. Warehouse jobs can say “Must be able to lift x pounds to y height regularly” or whatever so it’s obvious what the job requires and easy for the person to know if they can do it. It’s harder to quantify being smart, or having knowledge of certain fields, or ability to sell products. This means that it’s more effort on the part of people applying to know whether they should (which means they’ll probably just apply because you might as well) AND more effort on the part of the hirer to figure out who is qualified.
In theory, yes. If I am hiring a nanny and I have 300 applicants, I can walk down the line and keep only the 20 best groomed of the applicants, and while I can be virtually sure to have sent home the “best” applicant on many very sensible but detailed and difficult to measure criteria, I can also be sure that many of the 20 I kept will perform very well, and that a more intensive look at the remaining 20 than I could have afforded applying to the 300 will reliably identify a few very good candidates.
But these are empirical questions about what practices actually dominated, not theoretical questions about how it might have been. In actual practice, there were broad job classifications for which a black person would not be hired because they were black. In actual practice, the strike against them was that many white people wouldn’t work with them. You would lose many more qualified white person from you hiring pool than you would gain in qualified black people, and so your average hiring and wage costs would go up. That is, most whites would highly value not working with blacks and that would have to be figured in your hiring costs.
Certainly, the reasons given for this social judgement against black people were things like they were lazy, dishonest, stupid, ignorant, dirty, among other things. And indeed, given their exclusion from schools, many social institutions, and their lack of income from the existing social equilibrium, there was much truth to these generalizations.
In this case, in the U.S., the situation was massively changed by the imposition of federal laws on the states, industries, and institutions where this occurred. Armed militia were deployed to protect a small number of very bright and very brave black students who attended government funded but previously closed to them universities. Courts ordered the mixing of the races in public schools.
Over the decades, the real differences between the races arising from the real differences in opportunities and resources available on the basis of race has declined immensely, and most of the generalizations have been abandoned by most of the population.
The actual way it happened in almost all cases is an essential aspect to understanding these things and in reasoning about them, and the policies that seem to have changed them. The theoretical possibilities, in a vacuum that did not include the gigantic social agreement among whites that blacks were not wanted, are useless and even distracting to a clear understanding of what was going on.
Over the decades, the real differences between the races arising from the real differences in opportunities and resources available on the basis of race has declined immensely,
This seems dubious. If you look at intelligence (as measured by say SAT scores) or crime rate, you will find that there’s still a very large difference between the races.
Is there an accessible publicly available explanation on the argument that crime rate is linked to poor intelligence?
The argument for a correlation between poverty and intelligence is quite straightforward, and it is clear that crime rate (especially violent crime and trade in illicit substances) is correlated with poverty, which is also correlated with race.
But the reason that race continues to be correlated with poverty is exactly the question that is under dispute, so use of the correlation between crime rate and race that rely on the correlation between poverty and race assumes the conclusion that is under dispute.
For those who did not click the link, the article asserts that (1) crime is often caused by irrational cognitive bias, (2) people with less intelligence show more cognitive bias.
(2) is an empirical claim that makes quite a bit of sense.
I’m not sure that (1) is a more important effect than classical economics accounts of crime. If you will starve unless you eat the seed corn, achieving goals that don’t involve starving eventually are out of your reach regardless of how rational you are.
a few minutes times thousands of prospective applicants is a lot of time. Depending on where you believe the averages lie it could easily make economic sense to use the race heuristic for hiring.
Wouldn’t most group-based heuristics be useless when self-selection is involved?
Let’s assume we care about strength, and that men are significantly stronger than women (I’m cutting corners for simplicity’s sake).
If you’re looking for people to mug in the street, and all else being equal prefer a weak victim, then you’re better off targeting women than men—it’s a useful heuristic.
However, if you’re hiring people for a job that requires strength, people will only apply if they think they have a chance, then you should expect about the same distribution of strength among the man and women who apply—any sex-based heuristic is useless.
You’d think so, but from my experience screening resumes tons and tons of people who are not qualified or obviously don’t fit very simple criteria still apply to jobs. I wasted time reading resumes of people who shouldn’t have even applied, and I wasn’t even doing any interviewing.
Strength is also a lot less fuzzy than intelligence/capability in general. Warehouse jobs can say “Must be able to lift x pounds to y height regularly” or whatever so it’s obvious what the job requires and easy for the person to know if they can do it. It’s harder to quantify being smart, or having knowledge of certain fields, or ability to sell products. This means that it’s more effort on the part of people applying to know whether they should (which means they’ll probably just apply because you might as well) AND more effort on the part of the hirer to figure out who is qualified.
Less useful, not necessarily entirely useless. Depends on who rational the people doing the self-selection are.
In theory, yes. If I am hiring a nanny and I have 300 applicants, I can walk down the line and keep only the 20 best groomed of the applicants, and while I can be virtually sure to have sent home the “best” applicant on many very sensible but detailed and difficult to measure criteria, I can also be sure that many of the 20 I kept will perform very well, and that a more intensive look at the remaining 20 than I could have afforded applying to the 300 will reliably identify a few very good candidates.
But these are empirical questions about what practices actually dominated, not theoretical questions about how it might have been. In actual practice, there were broad job classifications for which a black person would not be hired because they were black. In actual practice, the strike against them was that many white people wouldn’t work with them. You would lose many more qualified white person from you hiring pool than you would gain in qualified black people, and so your average hiring and wage costs would go up. That is, most whites would highly value not working with blacks and that would have to be figured in your hiring costs.
Certainly, the reasons given for this social judgement against black people were things like they were lazy, dishonest, stupid, ignorant, dirty, among other things. And indeed, given their exclusion from schools, many social institutions, and their lack of income from the existing social equilibrium, there was much truth to these generalizations.
In this case, in the U.S., the situation was massively changed by the imposition of federal laws on the states, industries, and institutions where this occurred. Armed militia were deployed to protect a small number of very bright and very brave black students who attended government funded but previously closed to them universities. Courts ordered the mixing of the races in public schools.
Over the decades, the real differences between the races arising from the real differences in opportunities and resources available on the basis of race has declined immensely, and most of the generalizations have been abandoned by most of the population.
The actual way it happened in almost all cases is an essential aspect to understanding these things and in reasoning about them, and the policies that seem to have changed them. The theoretical possibilities, in a vacuum that did not include the gigantic social agreement among whites that blacks were not wanted, are useless and even distracting to a clear understanding of what was going on.
This seems dubious. If you look at intelligence (as measured by say SAT scores) or crime rate, you will find that there’s still a very large difference between the races.
Is there an accessible publicly available explanation on the argument that crime rate is linked to poor intelligence?
The argument for a correlation between poverty and intelligence is quite straightforward, and it is clear that crime rate (especially violent crime and trade in illicit substances) is correlated with poverty, which is also correlated with race.
But the reason that race continues to be correlated with poverty is exactly the question that is under dispute, so use of the correlation between crime rate and race that rely on the correlation between poverty and race assumes the conclusion that is under dispute.
Here.
Thanks.
For those who did not click the link, the article asserts that (1) crime is often caused by irrational cognitive bias, (2) people with less intelligence show more cognitive bias.
(2) is an empirical claim that makes quite a bit of sense.
I’m not sure that (1) is a more important effect than classical economics accounts of crime. If you will starve unless you eat the seed corn, achieving goals that don’t involve starving eventually are out of your reach regardless of how rational you are.