I mentioned in an old Linkluster post regarding an old court ruling that allowed a police department to discriminate against people that scored really well on their variant of an IQ test. This spawned a conversation between Kirk, Brandon, Phi, and myself.
“It’s not okay to discriminate against dumb people, so why is it okay to do it to those who are smart?” -Kirk
“Technically, it’s not illegal to discriminate against people with low IQs. But in practice doing so has a disparate impact on another demographic which it is illegal to discriminate against. You don’t have that problem with discriminating against smart people. ” -Brandon
“Brandon called it. If discriminating against low IQs has a disparate impact on blacks, discriminating against high IQs has a disparate impact on whites. Why should one be allowed but not the other?” -Phi
“We see here that they used it to discriminate against people who did very well, but they almost certainly use it against people who did poorly. So somehow or another, they have already justified the disparate impact of the test.” -Trumwill
Reading over another account of the case, I am relatively sure that we all actually missed what’s really going on here. I touched on it in my comment, but half-accidentally. I initially actually believed the departments claims of concerns over turnover due to boredom or that it was a sort of personality profiling. But the more I think about it, this is less likely something despite disparate impact, but rather it was done precisely because of disparate impact.
We all know the legal problems with IQ tests: they have a disparate impact on minorities. This can be overcome, but only with a justification process that can be expensive and arbitrary. So organizations don’t like to do it. However, if you can devise an IQ test that doesn’t discriminate against minorities, then you don’t really have a problem. Therefore, instead of accepting scores above a certain threshold, you accept scores within a particular target zone. That means excluding low IQs (more likely to be Hispanic or black) and high IQs (more likely to be white or Asian). That, to me, makes a lot more sense than the personality profiling (with is self-deprecating in the extreme) or a disdain for high IQs (police departments are more frequently asking for more education rather than less). From a police departments perspective, eliminating a few high-IQ people from consideration is worth the cost of being able to eliminate those at the low end of the spectrum. From a utilitarian standpoint, that actually makes sense to me.
What would make things really interesting is if a bunch of Asian-Americans sued.
—-
This post is going to be treading on dangerous terrain. It’s unavoidable. All I ask is that we avoid derogatory remarks and derogatory references to stereotypes. Let’s assume the following for the sake of this post:
(1) An IQ test, or a test that can be directly tagged to IQ, will have a disparate impact on Non-Asian Minorities.
(2) The reason for #1, be it genetics, education disparities, cultural disparities, or what-have-you, are not particularly relevant to the discussion.
(3) Because of (1), cities are loathe to employ such tests because of the hurdles required to justify the disparate impact. But sometimes they do it anyway because regardless of #2, they see a benefit in excluding people below a certain threshold on such tests.
About the Author
5 Responses to Sidestepping Disparate Impact
Leave a Reply
please enter your email address on this page.
I thought the links between IQ and race were discredited a while back… I always think of IQ test as an old way to really rate thing. Aren’t most IQ test very outdated now… this is bizarre to me in many leves. That said, every single one of those test that I have taken I have done quite above average even for an immigrant who is taking it on his second language.
I will have to talk to my friends that are cops and see how their departments did it. I was under the impression that there were police academies where people just went to once they got a job with a city or something. I am really curious about the process now.
I thought the links between IQ and race were discredited a while back…
It most definitely hasn’t. The only debate at this point is whether the reason for the disparity is completely environmental or partially or mostly genetic. People are still going on and on about that, but for the purposes of these tests it doesn’t actually matter what the causes are.
The relevance of IQ tests is also subject to debate. I tend to give it more. Others say it is meaningless.
Back in Colosse, there were two routes to being a police officer. Bigger departments had their own academies and you would get hired as a cop and then go through it. Smaller departments would require people who went through private academies or academies associated with criminal justice college degrees. My understanding is that the former outnumbers the latter.
I guess it would be hard to really measure in my case, because environment is 11 years in South America and 6 years in North America when it came to schooling. I think there is a genetic factor, but people in South America have a very differen genetic composition than here in the US where it seems to have been more segregated. Even if I wanted to classify myself it would be extremely difficult, I have both caucasian and native blood and don’t even truly know what the percentages are. Two of my grandparents were way more native than the other two, one of them seemed to be a lot more caucasian than the remaining one but percentages are based purely in guesses and the way they looked (which with genetics can be very innacurate because of genes receeding and things I really know only vaguely)
Looking at the catholic school I attended down in Colombia I would say that about 70% of the kids could have scored very high in IQ tests. Recently thanks to facebook I have seen what most of them are doing and a high percentage of them are in good professions that require the use of the brain (and I could see an entry requirement could be an IQ test of some sort). In contrast, the people I went to high school with here in the US don’t seem to be in the same boat even thought they could have been put in the same socio-economic bracket. I don’t keep in contact with any, and there is only like 5% of them that I know are in professions that could test for IQ, but I really don’t have a good sample. You would think that I would keep in contact with more people from the US than those that I went to school with back in 93.
This is brilliant. I wish I’d thought of it.
Trumwill: I should have seconded Brandon when you originally wrote this. Brilliant observation.