Chicago Fire fighters affirmative Action

And this was based on, since we don't see that in the article, the fact that percentages of a certain group scored noticeably low? Is that all it takes? I mean, we don't know the total number of applicants here, but suppose I open a job, and give everyone a test that is, for all practical matters, without any possible racial bias. Suppose the candidates from a particular race score particularly low. Not because the test sucks, or not because it is skewed to somehow weed out a certain group, but... just because they do. And let's say those numbers support your 3 step process to victory. What then?

Granted, if 6000 were wronged, the numbers in this must be pretty high overall, and perhaps that's enough for the skew to be determined as more than coincidence... but even if so, where is that line?

Just curious. Where there is smoke, there is probably fire (firefighter pun intended) but still seems a rather simplified way to determine something like that.

 

I wanted to first comment on your "just because they do" point.

There are 2 reasons to why a particular race would score low on an exam.
1) They are genetically dispositioned to do bad on the exam. (i.e. race in country is born dumb)
2) They did bad on the exam due to social and economic disparity compared to another race. (i.e. race in country has been purposefully roadblocked from natural social evolution)

What reason do you think it is?

I'm not really gonna make you answer that (feel free to do so though). My point is that some people believe #1 is true and the whole 4/5ths rule is out the window regardless of where it is being applied. Hard to argue against ignorance. But if you think #2 and your hypothetical exam was for positions of police detectives instead of firefighters, your company should be able to prove that ability to score high on aptitude is critical to job performance. Plantiffs would lose the case against your company.

The process has 3 steps. All 3 must show that the exam was discriminatory. In my opinion, the discrimination in the firefighter case was inadvertant. As I said, they had too many candidates and just looked for ways to reduce the number of people that would qualify. I highly doubt that the city wanted only whites as firefighters.

 

Well, my example wasn't to be applied here specifically, it was a general scenario question.

You left out 3) just because that day, these people might do poorly and a certain number of them (again, where's the line ?) could coincidentally be of a certain race. I agree that with a sample the size of this firefighting example, that would be hard to argue against those numbers....and so was giving a different example to examine ... well... where the line is. 100 applicants? 1000? 10000? How small a group is subject to this formula? Just curious.

I would agree that it was probably inadvertent in this case to a certain extent, not that we'll know really, but seems a rather ruthless formula minus any other details.

Critical to job importance? What about just hoping to hire those who were smart enough to do well on the test?

I don't want to really get caught up in the firefighter plight as I think we've examined it and I appreciate your insight on the process. I was more thinking "Wow, it's not hard to imagine a company getting hosed on this formula" depending on a few variables... not that I discount the intent, or even success of it in whatever situations.

 

why is that anyone's fault? Best scores should advance.

 

Good point, I believe there is a minimum number at which the four-fifths rule can't apply statistically. I just don't know enough about the Disparate Impact theory. It would be good to see all court decisions that used the rule and see how it's been applied.

That's where plantiffs lose out. Now we don't know if discrimination was purposeful or not. I don't think it was but say for example it was. It is plausible to have a scenario where the hiring manager was racist and knew that to eliminate a lot of white candidates, he only had to make the test based on something he knew whites would do poorly on such as an exam on hip-hop/rap lyrics. He would be deliberately discriminating while masking it with some non-applicable requirement. Some whites would do well but a significant portion would do poorly.

Yes, I understand what you are saying. I don't have a clear idea why the four-fifths was chosen instead of another ratio like 5/8ths or something. Another thing is I don't know how well the 4/5ths rule would hold out if this was not a govt position. But there is a more precise method called Fisher's Exact Test (http://en.wikipedia.org/wiki/Fisher's_exact_test).

 

Agreed.

Agreed and completely understandable, though it's hard to imagine a test that would be so blatantly skewed that it would garner the hypothetically intended results. Obviously it can happen, but this discussion makes me want to know exactly what about the test was so powerful yet subtle enough to be used.

Great info. Thanks.

 

i guess this case is along the lines of making the poll tax illegal.

 

So I think I understand the idea that the test was not applicable to the job - (no offense to firefighters, but it does not seem like the kind of job that requires you to be terribly intelligent).

However, the idea that a test of intelligence is discriminatory to African Americans seems like a racist idea in itself.

Then again, I have not read any of the case in detail, so this could all be conjecture - I'm just going off the posts in the thread...

 

Maybe it WAS applicable to the job. Again, might be interesting to know what was on said test. Anyway, calling something a fair test of intelligence and it actually being a fair test of intelligence can be different things.

 

Yeah - it's hard to evaluate this objectively without knowing what's on the test. I don't care enough to go looking though so...bleh.

 

lol man

 

but that contradicts what you said.

if firemen don't have to be smart, then it should be easy for anybody. correct?

 

Uhhhh...no contradiction, that was my point.

 

Actually it takes a fair amount of intelligence to be a firefighter. You need knowledge of building construction, the properties of materials under high temperatures, medical knowledge and etc..

That doesn't mean that the test actually determined those things but firefighting is definitely not unskilled labor.

It depends on how the test was worded and administered.

Neither have I so I don't really have an opinion either way about this particular case.

 

Fair enough, I retract my clumsily worded earlier statement.

Let me rephrase:

If it was a generic aptitude test (like a small-scale SAT), I think the idea that it discriminates against black people is, in itself, discriminatory. That being said, I recall plenty of studies that indicated black folks have a (probably society-induced) tendency to score lower on standardized tests (I remember a particularly amazing study that showed a correlation of lower test scores among black test-takers if "race" was asked vs if it was not!). Point being, while I may think it a debasing foundation to start from, that may be unfair given specific racial social context. Meh.

If it was a test of firefighting knowledge and acumen (which, given the judge's comments appears to not be the case), I would think the accusation of discrimination baseless. Those better qualified should be hired.

 

each child in American has rights to a public education

 

It's not because of the lower scores. Again, it was because the exam was not a valid requirement to the job.

In your example (assuming that if failed the 4/5ths metric), if the defendants fail to prove that doing good on the exam resulted in better job productivity AND the prosecutors could provide that an alternate exam exists or could be created to better gauge job productivity, then it should count as discrimination against whites.

 

Fortunatly for you, the chinese in america weren't forced into slavery for 300 years.