By The Content Of Their Character

June 29, 2009 8:28 PM

Today, on the last day of this year's term, the Supreme Court of the United States issued its opinion in Ricci v. DeStefano, the latest in the Court's line of decisions on Title VII and the role of race in employment decisions. The famous case centers on white firefighters' claims of race discrimination following the town of New Haven's decision to scuttle a promotion exam after white test takers performed disproportionately better than black firefighters.

The case has been covered extensively in blogs and the mainstream and local media. In addition to the high-stakes questions at issue in the case itself, Supreme Court nominee Sonia Sotomayor sat on the panel of the U.S. Court of Appeals for the Second Circuit that summarily decided the case at the intermediate appellate level.

posted by Law Talkin' Guy (89 comments total)

A good background look at the case. The local alternative press reacts.
posted by availablelight at 8:51 PM on June 29, 2009

I'm of the opinion that this was the right overall decision. For whatever reason, out of that small pool of participants, one group did disproportionately well, and the City threw away the test result because they didn't like the skin color of the people who passed. That's about as racist as it gets.
posted by Malor at 8:53 PM on June 29, 2009 [9 favorites]

That's about as racist as it gets

Not if the test itself was "racist".
posted by @troy at 8:58 PM on June 29, 2009 [1 favorite]

How was the test racist though? I've been following this a bit and I just read the Slate article and it sounds like the city bent over backwards to avoid any bias at all.
posted by fshgrl at 9:04 PM on June 29, 2009 [3 favorites]

For whatever reason, out of that small pool of participants, one group did disproportionately well, and the City threw away the test result because they didn't like the skin color of the people who passed.

They threw away the test result because they felt the test result wouldn't stand up under law, which considering that the test's relevance to performing the job in question - as compared to other tests where minorities did not perform poorly - was low to negligible was a reasonable course of action considering the Civil Rights Act would make any lawsuit resulting from harm suffered by failing the test essentially a walk in the park.

The Supreme Court's Ricci decision isn't much more than the conservative wing protecting those poor oppressed white people as per usual. Given that the decision essentially requires judicial interference with legislative intent - or what conservative judicial critics traditionally call "judicial activism" when they think it's being done for liberal intent - you'll pardon me if I feel the need to sit back and take in the rampant hypocrisy on the part of originalists.
posted by mightygodking at 9:04 PM on June 29, 2009 [1 favorite]

Malor that's a pretty slanted way of portraying the case.
posted by oddman at 9:05 PM on June 29, 2009

I thought this was a pretty intelligent commentary on the test and the city's motivation:

So here's what I'm saying: The city had reason to believe the test was flawed, not only because how it was constructed favored things less applicable to actual firefighting, it advantaged people with more resources and personal connection to the department, and there are other ways to construct the test in which not only have more relevance to the job, they actually have less of a disparate impact. If the city had arbitrarily tossed out the white firefighters' scores and promoted less qualified minority candidates, that would have been one thing. But they didn't do that, the threw out all of the results--blame should be put on the city for having created a flawed test in the first place, but it's not fair to say that the white firefighters were denied their jobs based on the kind of longstanding racial assumptions about black intelligence and diligence that still often govern hiring decisions.

posted by ibmcginty at 9:06 PM on June 29, 2009 [6 favorites]

Question: have those who have commented to this point read the decision? I mean, the whole decision?

Just wondering. I'm a fast reader and I have read maybe 15% of it.
posted by bz at 9:11 PM on June 29, 2009 [3 favorites]

While I can certainly understand issues of racial/cultural bias as they pertain to general aptitude tests like the SAT, ACT, etc... I wonder how much of that could apply to tests that are so vocation specific.

That's an honest question btw, not an open piece of rhetoric. Well not intentionally anyway.
posted by Senor Cardgage at 9:11 PM on June 29, 2009

And then I saw ibmcginty's post.
posted by Senor Cardgage at 9:12 PM on June 29, 2009

From the slate.com series I linked, what an unbiased test might look like:

If New Haven could start over again, what would it do? How would the city decide which firefighters to promote? Lt. Gary Tinney and the black Firebirds Society favor an alternative called an "assessment center." Assessment centers are designed by testing professionals to evaluate the particular skills needed for a certain job. Rather than distributing written questions and following up with an oral exam, this method intersperses a combination of interviews, group discussions, written exercises, oral presentations, role play, and emergency scenarios in order to closely simulate the job environment. Assessment centers are now used by fire departments from Ohio to the District of Columbia. They have been shown to reduce greatly the disparate impact that New Haven's test had in eliminating black promotional candidates.
posted by availablelight at 9:16 PM on June 29, 2009 [4 favorites]

Given that the decision essentially requires judicial interference with legislative intent - or what conservative judicial critics traditionally call "judicial activism" when they think it's being done for liberal intent - you'll pardon me if I feel the need to sit back and take in the rampant hypocrisy on the part of originalists.

What in the world are you talking about? You do realize that the concept of "disparate impact" is a judicially-created gloss on Title VII, right? Please read Griggs v. Duke Power. Thx.
posted by Slap Factory at 9:25 PM on June 29, 2009 [2 favorites]

Assessment centers are designed by testing professionals to evaluate the particular skills needed for a certain job.

As were the written and oral tests in New Haven.

Rather than distributing written questions and following up with an oral exam, this method intersperses a combination of interviews, group discussions, written exercises, oral presentations, role play, and emergency scenarios in order to closely simulate the job environment.

I am mindful of the various complexities and historical inequities at play here, and am a supporter of affirmative action - but I fail to see how you can eliminate (real or apparent) racial bias by replacing a multiple choice exam with a subjectively assessed series of exercises.
posted by moxiedoll at 9:25 PM on June 29, 2009

have those who have commented to this point read the decision? I mean, the whole decision?

I have a heuristic that whenever I see Roberts, Alito, Scalia, Thomas, and Kennedy together chances are 100.0% that I'm going to disagree. So far this has held.

FWIW, I took this Which Supreme Court Justice Are You Test online test and Kennedy came out #1 at 68% match, followed closely by the liberal wing, and dismal matches to the other 4.

Now, being a left libertarian there are cases here & there where I have agreed with some of the conservatives. Like Raich, Lopez, and the "off-the-wall" IR surveillance case.
posted by @troy at 9:27 PM on June 29, 2009

@troy, you're not qualified to comment on this matter unless you've read Title VII and read the opinion. Get back to me when your level of constitutional commentary gets beyond "I'm a left libertarian and don't like Scalia."
posted by jock@law at 9:32 PM on June 29, 2009

To me, the most shocking thing about this decision is Alito's concurrence. It's completely out-of-character for him and I would imagine offers a rare insight into something he obviously feels very strongly about. Alito is in many ways the most reserved jurist from a very technical perspective. His opinions and dissents are largely air-tight. It doesn't mean that I agree with them, but he almost never over-reaches and his writing is clear and concise. The concurrence is just a rant against of urban politics and black power structures. It's bizarre and when I read it, I was just stunned.

But honestly, a 5-4 opinion for the firefighters with Kennedy writing the opinion? Totally unsurprising. And the thing itself is a mess, Ginsberg's claim that this is unlikely to have a lasting impact is probably right. It's clear that there were at least three, if not four votes to throw out Griggs v. Duke Power and Title VII's application to Disparate Impact cases, but Kennedy would never get on board for it. I doubt Roberts would either, but we don't know. So instead of an ultra-right clusterfuck we get a meandering center-right opinion that's largely groundless both legally and from a policy perspective. It's unclear where this came from, other than an obvious compromise between justices, driven largely by Justice Kennedy.

The liberal judges will likely win this in the long run. Avoiding a disparate impact in hiring ought not to be seen as disparate treatment. It may be hard to construct a judiciable test for this principle, but context matters and the lower courts have slowly gotten used to it. These were unsympathetic facts for New Haven and one has to wonder if things would have been different had Ricci not been dyslexic. Still, it would really be astounding, on a whole level beyond Parents Involved, for the Court to gut Title VII as much as Scalia implies it could, were the debate to be honest. I mean, his concurrence is interesting too, insofar that he shows his hand and makes it clear that he thinks that Disparate Impact is violative of the 14A. He knows it won't happen and there's no surprise in Scalia pulling a stunt like that in a concurrence. Indeed, I'd be surprised if we ever get another justice on the court who wants to overturn Duke Power, this is probably the last gasp. Still, I feel like we learned something very disconcerting about Justice Alito today. That's my takeaway from the whole thing.
posted by allen.spaulding at 9:34 PM on June 29, 2009 [6 favorites]

you're not qualified to comment on this matter unless you've read Title VII and read the opinion. Get back to me when your level of constitutional commentary gets beyond "I'm a left libertarian and don't like Scalia."

oh blow me. Any 5-4 decision from this court is a legal jump-ball. And will get revisted.
posted by @troy at 9:36 PM on June 29, 2009 [2 favorites]

Has anyone seen any discussion on the reliability and validity of the assessment? It seems hard to believe that the test itself would be an effective measure of aptitude in the absence of rigorous test construction...something that i haven't heard or read brought up (but I very well may have missed something).

It seems to me that constructing a flawed test is the simplest explanation as to why certain groups (that, otherwise, were realistic candidates) would fail the test.
posted by Hypnotic Chick at 9:38 PM on June 29, 2009

I didn't mean it negatively. I thought maybe you all had read it because there is no simple conclusion in the post and no link to a cliff's note version and I was assuming you must have read the decision and I just couldn't see how anyone could have gotten through it so quickly.

It's sheer girth surprised me.
posted by bz at 9:38 PM on June 29, 2009

Not if the test itself was "racist".
posted by @troy at 11:58 PM on June 29

Then maybe the city should have offered at least one tiny piece of evidence that would suggest that. Which they did not. In fact the record suggests the city went to great pains to construct a racially neutral test, and only threw it out after the results came back.

The Supreme Court's Ricci decision isn't much more than the conservative wing protecting those poor oppressed white people as per usual. Given that the decision essentially requires judicial interference with legislative intent - or what conservative judicial critics traditionally call "judicial activism" when they think it's being done for liberal intent - you'll pardon me if I feel the need to sit back and take in the rampant hypocrisy on the part of originalists.
posted by mightygodking at 12:04 AM on June 30

In other words, you didn't actually read the opinions. From the dissent:
New Haven maintains that it refused to certify the test results because it believed, for good cause, that it would be vulnerable to a Title VII disparate-impact suit if it relied on those results.

The ditched the results because they were afraid of a lawsuit. They mistaken believed that the risk of a lawsuit only comes from one group and not another. The majority opinion held that fear of a lawsuit is not an acceptable reason to ditch the results.

Furthermore, if anyone is inclined to accept the dissent's statement of the facts at face value, I invite you to consider Alito's concurring opinion, in which he lays out all of the other the facts that the dissent chose to omit.
posted by Pastabagel at 9:40 PM on June 29, 2009 [2 favorites]

mightygodking: They threw away the test result because they felt the test result wouldn't stand up under law

Actually, no. That's the whole crux of the case. They could have thrown them out if they believed it wouldn't "stand up under law." But they threw them out because they feared litigation.

They were afraid of hypothetical lawsuits, and that worry is insufficient to justify the race-based decision to throw the tests out. What would have been sufficient is a strong basis in evidence that there would have been actual liability.

From the opinion's Syllabus:

Under Title VII, before an employer can engage in intentional discrimination for the asserted purpose of avoiding or remedying an unintentional, disparate impact, the employer must have a strong basis in evidence to believe it will be subject to disparate-impact liability if it fails to take the race-conscious, discriminatory action.

Believing oneself in actual violation of the law is subtly but importantly different from believing oneself a likely target of litigation claiming you were in violation of the law.
posted by jock@law at 9:46 PM on June 29, 2009 [2 favorites]

Then maybe the city should have offered at least one tiny piece of evidence that would suggest that. Which they did not. In fact the record suggests the city went to great pains to construct a racially neutral test, and only threw it out after the results came back.

"Great lengths"? Bullshit.

New Haven, the record indicates, did not closely con-
sider what sort of “practical” examination would “fairly
measure the relative fitness and capacity of the applicants
to discharge the duties” of a fire officer. Instead, the City
simply adhered to the testing regime outlined in its two-
decades-old contract with the local firefighters’ union: a
written exam, which would account for 60 percent of an
applicant’s total score, and an oral exam, which would
account for the remaining 40 percent

-- Ginsburg's dissent
posted by @troy at 9:48 PM on June 29, 2009 [1 favorite]

What many find most interesting about this is that none of the Justices agreed with Sotomayor, not even Ginsburg. That's sure to come up in her confirmation hearings.
posted by Chocolate Pickle at 9:49 PM on June 29, 2009 [3 favorites]

Regarding Alito's statement of the facts, Emily Bazelon at Slate notes:

"(Alito) sees a city that screwed a group of hapless white firefighters because the mayor was trying to curry favor with a "politically important racial constituency" and, in particular, with Boise Kimber, a black reverend in the city who spoke out against promoting an almost entirely white slate of fire officers. Boise Kimber has plenty of unsavory bits in his past. Alito runs through many of them. Because in his view, the evil to be protected against in New Haven is black political power."

I think this dovetails with alan.spaulding, point about Alito's potentially out of character opinion.
posted by Hypnotic Chick at 9:50 PM on June 29, 2009 [2 favorites]

I have a heuristic that whenever I see Roberts, Alito, Scalia, Thomas, and Kennedy together chances are 100.0% that I'm going to disagree. So far this has held.

I have a heuristic that people who make blankets statements like this aren't as smart as Roberts, Alito, Scalia, Thomas, and Kennedy, considered together or separately.

And yes, I read all of the opinions.

To allen.spaulding, I think the reason Alito wrote what he did is because Ginsburg's opinion recited a selection of facts that case the fire department as institutionally racist, when the record offered plenty of examples of the seven-term (!) mayor and his cronies exhibiting blantant racism of their own. I think he felt the need to demonstrate that the court considered all of the facts.

I agree with you that this opinion won't have much precedential value, but not for the reasons you mention. I think the particular and somewhat unique situation in New Haven makes this decision very fact-specific, with little broader application.
posted by Pastabagel at 9:52 PM on June 29, 2009 [3 favorites]

alien.spaulding: The liberal judges will likely win this in the long run.

I doubt that very much. Scalia and Ginsburg both hint at the notion that the "disparate impact" test and the Equal Protection Clause are headed toward a showdown in cases (like here) of government employers. The Equal Protection Clause -- being, you know, part of the Constitution? -- will assuredly win.

Equal treatment is great. Equal results are great. But sometimes you can't get both. And equal treatment is Constitutionally mandated.
posted by jock@law at 9:53 PM on June 29, 2009 [1 favorite]

@troy, I urge you to read the opinion from start to finish before making a further fool of yourself.

After reviewing bids from various consultants, the City hired Industrial/Organizational Solutions, Inc. (IOS) to develop and administer the examinations, at a cost to the City of $100,000. IOS is an Illinois company that specializes in designing entry-level and promotional examinations for fire and police departments. In order to fit the examinations to the New Haven Department, IOS began the test-design process by performing job analyses to identify the tasks, knowledge, skills, and abilities that are essential for the lieutenant and captain positions. IOS representatives interviewed incumbent captains and lieutenants and their supervisors. They rode with and observed other on-duty officers. Using information from those interviews and ride-alongs, IOS wrote job-analysis questionnaires and administered them to most of the incumbent battalion chiefs, captains, and lieutenants in the Department. At every stage of the job analyses, IOS, by deliberate choice, oversampled minority firefighters to ensure that the results—which IOS would use to develop the examinations—would not unintentionally favor white candidates.
....
IOS submitted the assessors’ resumes to City officials for approval. They were battalion chiefs, assistant chiefs, and chiefs from departments of similar sizes to New Haven’s throughout the country. Sixty-six percent of the panelists were minorities, and each of the nine three-member assessment panels contained two minority members.

posted by jock@law at 9:56 PM on June 29, 2009 [1 favorite]

In fact the record suggests the city went to great pains to construct a racially neutral test, and only threw it out after the results came back.

Agreed.

"'Great lengths'? Bullshit."

Read the decision which has a lengthy description of the lengths and job analyses that IOS, the company contracted to create the test, went to to create a test that would not favor the white candidates, including a deliberate oversampling of minority firefighters.
posted by bz at 9:59 PM on June 29, 2009 [1 favorite]

Ooops. Beaten.
posted by bz at 10:01 PM on June 29, 2009

"For security reasons, Department officials had not been permitted to check the content of the questions prior to their administration. Instead, IOS retained a senior fire officer from Georgia to review the exams “for content and fidelity to the source material."

The department off-loaded the test design to IOS. pastabagel's "great pains" claim above is completely false and misleading.
posted by @troy at 10:01 PM on June 29, 2009

Yeah, @troy, that quote you pulled from Ginsberg's dissent is a bit misleading out of context - the "two decades old testing regime" bit refers to the 60/40 weighting of the written and oral exams... not to the actual content of the exams.
posted by moxiedoll at 10:02 PM on June 29, 2009 [1 favorite]

YFWGI
posted by jock@law at 10:03 PM on June 29, 2009

went to to create a test that would not favor the white candidates

What is at issue here was, and will be again someday, whether the test itself, regardless of how allegedly well it was designed to be race-neutral comes under the Griggs ruling. If the test itself favors firefighters with familial or otherwise insider connections to the department then it is in effect racially discriminatory since the racial makeup of these insider connections is decidedly non-minority at this time.
posted by @troy at 10:06 PM on June 29, 2009 [1 favorite]

Malor that's a pretty slanted way of portraying the case.

I really don't think it is, in that New Haven didn't even vaguely approach the problem properly. They didn't even allow for the possibility that their small-sampling-size hiring pool might really have talent distributed in that way. That thought didn't, as far as I can tell, even cross their minds -- that the test might be right.

What they should have done was analyze the test. It's okay to test for actual job skills; if a job needs skill X, and testing for skill X means that fewer people of some skin color pass, that doesn't make it a racist test. But if the test is miswritten so as to also require irrelevant, culturally-biased information, then the test needs fixing. And that's what they needed to find out. Were they actually testing for the right things, or were they unfairly testing for irrelevant knowledge?

But instead of doing that, they just assumed that their test was bad, when they'd gotten pretty similar results from prior exams. They didn't like the fact that out of their hiring pool, the non-white candidates weren't doing as well. As long as you're testing for the right things, if that's the result you get, that's the result you get. That's why you have tests!

But minorities were doing poorly, and that wasn't politically acceptable, so instead of trying to determine the truth, they discriminated against the white people because they weren't some other skin color.

It's also worth pointing out that if, as so many of us believe, people really are equal taken across broad sampling groups, then you can't have a racist test that doesn't directly involve testing skin color. You can only have tests that are culturally biased. It's not possible for a knowledge-based test to be written so that people with black skin can't pass it -- it's only possible to write a test so that people from a particular cultural background will have trouble.

That's the mindset we need to be approaching the analysis with -- in addition to the skills we're explicitly testing for, what implicit knowledge, what cultural values and experiences, are required to understand the questions? As long as the actual cultural content is truly neutral, and the test itself is truly only examining the job skills, then if minorities do badly, the test is doing its job, and showing you the truth.

Wrecking the test to get a palatable answer is stupid and shortsighted -- it's the ostrich approach to fixing the problem, sticking our heads in the sand and hoping it goes away. If the test is right, and minorities aren't passing it, that means the PEOPLE need fixing, not the test. The measurement is showing us something important that needs doing. We damage society by ignoring that message, and promoting the incompetent based on their skin color.

We need to be deploying resources to get them up to speed, not destroying the validity of the exams to get results we find more aesthetically pleasing. Otherwise, why even have tests?
posted by Malor at 10:09 PM on June 29, 2009 [6 favorites]

Boise Kimber has plenty of unsavory bits in his past. Alito runs through many of them. Because in his view, the evil to be protected against in New Haven is black political power."

Let's be clear: Kimber doesn't hold an official position. He's a local demagogue that inexplicably has amassed tremendous influence with a mayor who apparently has been appointed for life. So how can attacking Kimber be considered "protect[ing] against...black political power," unless Kimber weilds power that he isn't supposed to have? New Haven needs protection not from black political power (which is utterly nonsensical) but from its shitty leadership, their cronies and hacks. But as I said, that isn't what Alito is talking about.

Alito runs through the facts in order to balance Ginsburg's skewed portrayal of the record. But apparently when he does it it's racist, even though when she does it, it isn't. Nice.

Read pages 47-49 of the pdf. Especially the "Listen to the Klansmen behind us" line from one of the idiots who didn't score well enough on the exam to get a promotion, and only after learning he flunked alleged, without any proof whatsoever, that the white firefighters had cheated on the test.

On preview, okay, so @troy wants to play. The test was structured as it was because of a decades old union contract. Everyone taking the test knew what was going to be tested.

Let me explain how arguing Supreme Court cases works - you don't get to cherry pick arguments from the Justices who share your politics and ignore the arguments from justices whose politics you don't. Ginsburg is certainly a brilliant jurist. But so are Kennedy, Scalia, and Alito. In understand what Ginsburg was trying to do, but I also understand that this wasn't the case for her to do it in-it's simply unfortunate for her that this was the case that ended up on the docket. She's attempting to introduce the notion that institutional history and occupational inheritance can be as pernicious in their racism as explicit or even implicit institutional policy. But there isn't any proof of that racism here, merely an outcome that is racially unbalanced and a total lack of facts that explain why that is (plenty of unsupported conjecture, however).

FYI, I don't think Sotomayor is anywhere near the jurist that Ginsburg was when she was nominated, let alone now.
posted by Pastabagel at 10:16 PM on June 29, 2009 [3 favorites]

If the test itself favors firefighters with familial or otherwise insider connections to the department then it is in effect racially discriminatory since the racial makeup of these insider connections is decidedly non-minority at this time.

I completely agree. Which is why I don't understand why this fact:

For security reasons, Department officials had not been permitted to check the content of the questions prior to their administration. Instead, IOS retained a senior fire officer from Georgia to review the exams “for content and fidelity to the source material."

is something you complained about. Keeping the content of the exam secret from other firefighters was an attempt to prevent "insider connections" from advantaging white applicants.
posted by moxiedoll at 10:18 PM on June 29, 2009

then you can't have a racist test that doesn't directly involve testing skin color. You can only have tests that are culturally biased.

Under Griggs, that doesn't matter at all. It's the effect that's considered:

Yet that is the choice the Court makes today. It is a choice that breaks the promise of Griggs that groups long denied equal opportunity would not be held back by tests “fair in form, but discriminatory in operation.”

Having a test that you can only be promoted if an ancestor was a firefighter is completely race-neutral.

If the test is right, and minorities aren't passing it, that means the PEOPLE need fixing, not the test.

Nah, to have minorities perform so poorly on this test makes such categorical statements about how great the test is at identifying the best candidates for promotion suspect:

More striking still, although nearly half of the 77 lieutenant candidates were African-American or Hispanic, none would have been eligible for promotion to the eight positions then vacant. The highest scoring African-American candidate ranked 13th; the top Hispanic candidate was 26th.

Now, that's statistically odd, don't you think?
posted by @troy at 10:19 PM on June 29, 2009 [1 favorite]

is something you complained about.

I didn't "complain about it", I used that fact to counter pastabagel's incorrect assertion that the Department took "great pains" in creating a fair test.
posted by @troy at 10:20 PM on June 29, 2009

To close the loop, @troy's comment quotes me thusly:

the record suggests the city went to great pains to construct a racially neutral test, and only threw it out after the results came back.

And then quotes Ginsburg:

New Haven, the record indicates, did not closely con-
sider what sort of “practical” examination would “fairly
measure the relative fitness and capacity of the applicants
to discharge the duties” of a fire officer.

Do you see how the two have absolutely nothing to do with each other? There could be a racially neutral test that is not a practical examination and does not fairly measure the relative fitness and capacity of the applicants; likewise there could be a racially biased test that is practical and does fairly measure the relative fitness etc. It could just be racially biased in some other way.

According to the record, the test was written and administered in accordance with a long-standing union contract. An African-American Assistant Fire Chief believed that the results should be certified (PDF, p. 46; Alito's concurring opinion, p. 5).

Seriously, they write these opinions for a reason.
posted by Pastabagel at 10:27 PM on June 29, 2009 [1 favorite]

"Now, that's statistically odd, don't you think?"

Winning the lottery is statistically odd yet it is won with regularity.
posted by bz at 10:29 PM on June 29, 2009 [2 favorites]

I didn't "complain about it", I used that fact to counter pastabagel's incorrect assertion that the Department took "great pains" in creating a fair test.

I think you're missing the point. Keeping the test secret from the department WAS ONE OF the actions the Department took to create a fair test.
posted by moxiedoll at 10:39 PM on June 29, 2009 [1 favorite]

Having a test that you can only be promoted if an ancestor was a firefighter is completely race-neutral.

You should be embarrassed about posting something that stupid.
posted by Malor at 10:45 PM on June 29, 2009 [2 favorites]

@troy, seriously... read the opinion, pay attention to the evidence, forget the filler, make a judgment. When you're done, I think you'll read your posts so far and slap yourself.
posted by JakeEXTREME at 10:46 PM on June 29, 2009 [1 favorite]

"Now, that's statistically odd, don't you think?"

It's only "statistically odd" if you control for other factors like age, experience, education, time with the organization, number of times each candidate took the test previously etc. Which obviously no one did. If the results came back that fire fighters under the age of 30 scored worse overall, or those with less overall experience and that correlated to ethnicity due to recent changes in hiring practices that meant more young fire fighters were minorities would you look at the result differently?
posted by fshgrl at 10:57 PM on June 29, 2009 [1 favorite]

I didn't "complain about it", I used that fact to counter pastabagel's incorrect assertion that the Department took "great pains" in creating a fair test.
posted by @troy at 1:20 AM on June 30

Okay, I see. Here's where you're going astray. Sorry if I got snippy earlier.

You can't argue facts as you or I imagine them. The hogher courts have to take the facts as they stand in the record of the lower court - appeals to higher courts, esp. the Supreme Court, can't introduce new facts. To critique the opinions, you have to argue the facts that the parties introduced in court. Maybe the all the white guys got free study guides from their dads in the department. Maybe that's true. But that fact is not in evidence, so you can't argue it. If that is true, hypothetically, your beef is with the city for putting on a crappy case.

The record is what the majority opinion relied on when it held "There is no genuine dispute that the examinations were job-related and consistent with business necessity. The
City’s assertions to the contrary are “blatantly contradicted by the record.”" (PDF, p. 32) or "there is no evidence—let alone the required strong basis in evidence—that the
tests were flawed because they were not job-related or because other, equally valid and less discriminatory tests were available to the City." (PDF, p. 37).

I am not asserting anything. I can't speak for what the Department did--I have no personal knowledge. I asserted that the record showed they went through great pains to ensure racial neutrality, and nothing in the record contradicts this:

"At every stage of the job analyses, IOS, by deliberate choice, oversampled minority firefighters to ensure that the results—which IOS would use to develop the examinations—would not unintentionally favor white candidates. " (PDF, p. 8) The process IOS went through is described in detail at PDF pp. 8-10.

In fact, the dissent only finds issue with the results of IOS's test, not with how they wrote it. But there could be many reasons for the racial disparity in the results that have nothing at all to do with the test or the fire department. The dissent only brings up that IOS testified that the City never asked it to try to develop an alternative testing method. But that doesn't imply that the method they used was racially biased.

The city had the burden to introduce in the lower court whatever facts could help it makes it's case. But the record showed, and the majority relied upon, the fact that the City left many of its key arguments unsupported by any facts.

Remember that this was a decision on summary judgment, and as the District Court noted "the parties strenuously dispute the relevance and legal import of, and inferences to be drawn from, many aspects of this case, the underlying facts are largely undisputed.” (PDF, p. 7)
posted by Pastabagel at 11:02 PM on June 29, 2009 [3 favorites]

I'm of the opinion that this was the right overall decision. For whatever reason, out of that small pool of participants, one group did disproportionately well, and the City threw away the test result because they didn't like the skin color of the people who passed. That's about as racist as it gets.

First of all, that's not as racist "as it gets". Second of all, the lower courts job was to determine if what the city did was legal or not, not whether or not it was "racist".
posted by delmoi at 1:09 AM on June 30, 2009

What many find most interesting about this is that none of the Justices agreed with Sotomayor, not even Ginsburg. That's sure to come up in her confirmation hearings.

All Sotomayor did was affirm a lower court ruling, along with three other judges (who were unanimous).

Put another way, 11 out of the 21 federal judges to rule on Ricci ruled as Sotomayor did. It's perfectly reasonable to argue that she ruled erroneously, but it's definitively unreasonable to claim that her Ricci ruling places her on some sort of judicial fringe.

Anyway, these 5/4 decisions since Roberts got on the court are all pretty absurd. The idea that these judges are neutral arbiters of justice, crunching through the law like emotionless computers is ridiculous.

It's fairly obvious that a good lawyer can come up with arguments for whatever position they need to take for their job. Some people seem to think that ability goes away when someone gets onto the Supreme court, and that's obviously not true.
posted by delmoi at 1:24 AM on June 30, 2009 [2 favorites]

I'm still confused as to how a test for a position in a Fire Department can be racially biased...
posted by PenDevil at 2:09 AM on June 30, 2009

More striking still, although nearly half of the 77 lieutenant candidates were African-American or Hispanic, none would have been eligible for promotion to the eight positions then vacant. The highest scoring African-American candidate ranked 13th; the top Hispanic candidate was 26th.

Now, that's statistically odd, don't you think?

If you actually know what you're talking about then you know it's impossible to judge whether that's statistically odd or not without more information. It's statistically odd if the minority candidates were as equal in knowledge as the white candidates. It's not necessarily statistically odd if the minority candidates were not as knowledgeable as the white candidates. The greater that knowledge differential, the less it looks odd.

So the question is whether the firefighting and other knowledge being tested was equal between the white and minority candidates. Since the only evidence we actually have is the performance on the test I don't know how to resolve that question.

But as others have said that's not really the issue. The issue was simply whether New Haven could toss out the test simply because they feared litigation in the absence of a real belief that the test was biased. You seem to be glossing over the fact that New Haven didn't toss out the test because they were convinced it was a racist test, they threw it out because they thought the minority candidates would sue them if they went with the results. They may well have been correct in that belief; instead they got sued by the white candidates.

Moral of the story: Sucks to be New Haven.
posted by Justinian at 3:54 AM on June 30, 2009

I offer an anecdote on one detail alone -- that of an employer realizing weird results on something means a problem with the test.

I've been an office temp for years. A few months ago I re-tested with my main temp agency -- you have to be tested periodicaly on various computer skills, typing, etc. They had a different type of test they were using this year, they told me. It, like most others, was an interactive computer-based test, where I was given a mock copy of Word or Excel or whatever I was testing and asked to perform a series of commands.

And -- it was unusually hard. I know Word forwards and backwards, and I was still striking out on this test. I did okay on it, well enough to get work, but I was still stunned that so many of the things I thought knew how to do were suddenly wrong.

My typing test was way down too -- I had previously tested at 90 words per minute, but in this test, I tested at 75. I even asked to re-take the typing test, because that just seemed so wrong. I tested at 70 that time. I let it stand.

A month later, I got offered the chance to re-test again, with a different test. "We noticed that a lot of people were testing way differently than they usually did," my agency said, "and it was an unusually low number of people who were passing. We checked out the test and saw that it was unusually difficult -- it only had one acceptable way to perform each of the commands in the different programs -- like using the keyboard shortcuts for various commands. It actually marked you wrong if all you did was use the mouse and a pull-down menu instead."

Granted, I don't know that there was any racial influence on their decision -- it seemed to have been based only on, "whooo, a whole lot of people seem to be flunking this, what's going on?" But that is at least one instance where an employer saw weird results on a test and threw out the test.
posted by EmpressCallipygos at 4:45 AM on June 30, 2009 [1 favorite]

> What they should have done was analyze the test. It's okay to test for actual job skills;
> if a job needs skill X, and testing for skill X means that fewer people of some skin color
> pass, that doesn't make it a racist test. But if the test is miswritten so as to also
> require irrelevant, culturally-biased information, then the test needs fixing.

There are other factors besides culturally-biased information that can strongly affect the outcome. For instance, differences in work and study habits that separate groups of candidates by cultural background. From this morning's New York Times report:

Many of the plaintiffs in the case — 18 white firefighters, one of them Hispanic — studied
intensively for the test, giving up second jobs and missing family celebrations. The lead
plaintiff, Frank Ricci, who is dyslexic, said he studied for 8 to 13 hours a day, hiring an
acquaintance to tape-record the study materials.

For perfect outcome-neutrality we'll need to be sure the revised test doesn't pick up any differences between candidates who study their asses off and candidates who don't.
posted by jfuller at 5:11 AM on June 30, 2009

I'm still confused as to how a test for a position in a Fire Department can be racially biased...

1. While serving hors d'œuvres at a local fundraiser, you hear screams of "fuego!" from the kitchen. Do you:

a. grab a bucket of aqua and head to the kitchen
b. find your main man, tell him the shit's on and he best grab his shit
c. locate the nearest portable mechanical device for distributing chemical expellant
d. throw down and cowboy up, partner
posted by Civil_Disobedient at 5:16 AM on June 30, 2009 [3 favorites]

I'm still confused as to how a test for a position in a Fire Department can be racially biased...

Indeed, I would love to actually see and/or take this test. Anybody have a link to a PDF? Just how pernicious were these questions, and how related/unrelated to firefighting are they?
posted by billysumday at 5:42 AM on June 30, 2009

this decision doesnt solve the issue of whether the promotion test was biased or not. what this decision did was to say "we dont care if the test is racially biased or not. you cant change it now because it is unfair to take away the competitive advantage you are giving the white firefighters after administering the test".

do you get that? it's why ginsburg says this decision will not hold. but the court of john roberts doesnt give a shit as long as it protects those conservative memes of "narrow" and "strict" interpretation of the law and constitution.

john roberts needed to be stopped back in 2005 way more than alito. this is the anti-equality, anti-civil rights shit we're going to get from this bigoted moron.
posted by liza at 5:50 AM on June 30, 2009

What many find most interesting about this is that none of the Justices agreed with Sotomayor, not even Ginsburg. That's sure to come up in her confirmation hearings.

In the sense that Ginsburg's tenth footnote involves a procedural quibble with the appeals court, yes. But the conservative meme that all nine justices disagreed with Sotomayor is just goofy posturing. Rush Limbaugh went so far as to describe the decision as "nine-zip" yesterday.

It's like conservatives have forgotten how to count to four.
posted by EarBucket at 5:56 AM on June 30, 2009 [2 favorites]

In a deal the city reached with the union, the tests were weighted so that 60 percent of each score was based on written questions and 40 percent on oral ones. Both parts of the exam were highly technical, designed to test memorization from classes and books rather than leadership or on-the-ground skills...The reading was later announced along with the date of the test, giving applicants a sense of what they should study. But Tinney says that white firefighters had access to this information earlier—through a "network" of friends, family, instructors of the extra training courses they'd taken, and connections at testing companies.

The vast majority of the white firefighters were raised in suburban areas with much better school systems; the vast majority of the minority firefighters are from New Haven where--to put it mildly--the public school system doesn't really prep anyone to test well. So, in addition to favoring those who, like most of the white candidates, come from generations of firefighters and have been raised within the "system" and network, the test as it stands also favors those who learned better how to game/ace standardized tests. (If you're one of those folks who thinks SAT tests are the same as IQ tests, I'm wasting my breath on this.)

I'd rather have a firefighter who went through an "assessment center" style evaluation (I know that state EMT testing is weighted more heavily towards on-the-spot knowledge and assessment).
posted by availablelight at 5:58 AM on June 30, 2009 [4 favorites]

[whoops-- that first block of text was lifted directly from here-- link didn't go through]
posted by availablelight at 5:59 AM on June 30, 2009

available light, just rock on. this bears repeating:

The vast majority of the white firefighters were raised in suburban areas with much better school systems; the vast majority of the minority firefighters are from New Haven where--to put it mildly--the public school system doesn't really prep anyone to test well. So, in addition to favoring those who, like most of the white candidates, come from generations of firefighters and have been raised within the "system" and network, the test as it stands also favors those who learned better how to game/ace standardized tests. (If you're one of those folks who thinks SAT tests are the same as IQ tests, I'm wasting my breath on this.)

sure, in this case "race" is actually social class with a twist: "race" is the network of family and "same-colored skinned" people that gives you a competitive advantage. and here in the US the majority of the better-off are considered white.

what is interesting to me is that this lawsuit was brought up by an italian (Ricci) and int his country italiants were not really considered white until as recently as the 1970s. The cultural proof is in movies like Saturday Night Fever, Taxi Driver and Karate Kid; where italians are portrayed "white outsiders" wanting desperately to belong.
posted by liza at 6:42 AM on June 30, 2009

I'm of the opinion that this was the right overall decision. For whatever reason, out of that small pool of participants, one group did disproportionately well, and the City threw away the test result because they didn't like the skin color of the people who passed. That's about as racist as it gets.

It's not racist if the test had zero validity in determining who would be a good choice to lead a group of firefighters.
posted by jonp72 at 6:53 AM on June 30, 2009

[Scalia] shows his hand and makes it clear that he thinks that Disparate Impact is violative of the 14A [14th Amendment].

I can think of no better evidence that originalist jurisprudence is complete horseshit that makes total hash of constitutional history. If Scalia knew history, he would know that the 14th Amendment was ratified at the same time you had the Freedman's Bureau (which was deliberately designed to redress "disparate impacts" against black people through racially targeted programs) and racial segregation in D.C. public schools (which gives the lie to the idea that the original intent of the 14th Amendment was "color-blind"). This is just twisting constitutional history so that it fits a right-wing anti-affirmative action consensus.
posted by jonp72 at 7:03 AM on June 30, 2009 [2 favorites]

So, in other words, no matter how damaged someone is by their upbringing, no matter if they're even literate, we should promote them anyway, because their skin is black.

How about fixing the fucking schools instead? All this approach nets us is a bunch of idiots in jobs they're not qualified for. Putting unqualified people in jobs is a disaster, because then they get promoted and start hiring and truly making policy decisions themselves, and there's no guarantee whatsoever that they'll ever pick up the necessary skills to actually do the job.

And the one thing that incompetent people is hire more incompetent people. Just one bad hire can, over the long term, do immense damage to an organization -- get a number of them, like a whole batch of starting firefighters at the same time, and it can be effectively ruined.

Firefighting is a damn dangerous occupation, and you'd better know what the hell is going on in a fire, or you could get people killed. Promoting incompetent firefighters, ones that don't really understand the mechanics of how fire works, makes it much more likely that some of those firefighters are going to die or be seriously injured, not to mention the additional property damage if the fire isn't attacked properly in the beginning. If there's one guy or gal on the team that needs to be a walking dictionary about fires, it's the foreman. If they direct the team wrong, people can die.

Don't fucking enshrine incompetence in the name of "racial equality", fix the fucking schools, and get remedial training for these guys until they DO pass the test.

It's not, after all, like they're not capable. You can always learn. Expecting -- no, demanding -- competence in that kind of high-risk occupation is the only sane choice.

Promoting them without demanding full competence means that we're accepting and enshrining that minorities are less capable. The anti-racists are putting racism on a pedestal and worshipping at its feet.

They're just as capable as anyone else, and demanding the same performance isn't even vaguely unreasonable.
posted by Malor at 7:04 AM on June 30, 2009 [2 favorites]

How about fixing the fucking schools instead? All this approach nets us is a bunch of idiots in jobs they're not qualified for.

Hai, I think you skipped the part where the black candidates didn't fail the test-- candidates were competing against each other in a scoring system where just a point or three less could push you out of the promotion pool.

Ever taken a hyper-competitive science or math course where an 89% could get you a "C-" if the curve was killed by anxious pre-meds? Did that C- make you an unqualified idiot who didn't master the subject matter?
posted by availablelight at 7:16 AM on June 30, 2009 [1 favorite]

liza: "this decision doesnt solve the issue of whether the promotion test was biased or not. what this decision did was to say "we dont care if the test is racially biased or not. you cant change it now because it is unfair to take away the competitive advantage you are giving the white firefighters after administering the test". "

No. What the opinion says is that New Haven can't throw out the test just because they think they might get sued.

If an employer believes that a test is actually illegal — actually a violation of the law, not just oops-this-looks-bad-we're-gonna-get-sued — they can still throw it out. This case doesn't change that, because it never really came up.

What people seem to be missing is that the City of New Haven never really argued that the test was unlawful, or that they thought it was a violation of the law. They just tried to toss the results because it looked really bad, and they thought they'd get sued.

From the second page of the opinion:

Under Title VII, before an employer can engage in intentional
discrimination for the asserted purpose of avoiding or remedying an unintentional, disparate impact, the employer must have a strong basis in evidence to believe it will be subject to disparate-impact liability if it fails to take the race-conscious, discriminatory action.
The Court’s analysis begins with the premise that the City’s actions would violate Title VII’s disparate-treatment prohibition absent some valid defense. All the evidence demonstrates that the City rejected the test results because the higher scoring candidates were white. Without some other justification, this express, race-based decisionmaking is prohibited. The question, therefore, is whether the purpose to avoid disparate-impact liability excuses what otherwise would be prohibited disparate-treatment discrimination. (Emphasis mine)

All this says — all of it — is that you need something more than the "threat of litigation" (like an actual belief that you've broken the law) before throwing out results and engaging in discriminatory behavior, for or against any group. You can blame New Haven for not trying harder to construct a case for the test being actually illegal, but they didn't do that. Their justification for throwing out the results was fear of litigation, and the case operated around that.
posted by Kadin2048 at 7:30 AM on June 30, 2009 [2 favorites]

I question why a written test is being used at all, regardless of any racial bias it may or may not contain. The positions in question are captain and lieutenant, both of which are primarily tactical (as opposed to Chief officers who are more often involved with strategy). It involves understanding how the fire is acting and factoring in the building conditions (for interior attacks), available personnel and suppression capability (how much water, how many hoselines, etc.). You don't want to know how well they do on a written test, you want to know how they're going to act in a fire: Are they going to fall apart? Are they capable of leading a team of disparate personalities? Do they act with safety in mind or are they reckless? Do the firefighters trust them?

I understand this is a civil service job and the firefighter's union has forced the use of the written test but surely there's a better way to do this.
posted by tommasz at 7:33 AM on June 30, 2009 [1 favorite]

Maybe the all the white guys got free study guides from their dads in the department. Maybe that's true. But that fact is not in evidence, so you can't argue it. If that is true, hypothetically, your beef is with the city for putting on a crappy case.

Maybe that's the crux of the disagreement here. It looks like New Haven was more motivated by avoiding lawsuits than by racial goals, whether those goals were motivated by affirmative action, racial justice, reverse racism, or whatever your point of view thinks is really at issue. If the black firefighters have the opportunity to file their own countersuit against the test, the results could be very different, simply because different facts will be introduced into evidence.
posted by jonp72 at 7:33 AM on June 30, 2009

It looks like New Haven was more motivated by avoiding lawsuits than by racial goals

It doesn't just look like it, it's the linchpin of the whole issue.
posted by Kadin2048 at 7:35 AM on June 30, 2009

So from what I'm reading, the test itself wasn't racist (or culturally biased... whatever). What was racist was that a test was given at all, because people seem to believe minorities are incapable of passing a test (or at least doing as well as a white guy). The belief of which is, ironically enough even more racist no matter how you try to explain it away. Nice.
posted by Nyarlathotep at 7:55 AM on June 30, 2009

Moral of the story: Sucks to be New Haven.

That was the moral of the story long before this case.
posted by oaf at 8:20 AM on June 30, 2009 [1 favorite]

Nyarlathotep, if I understand correctly, the moral of the case is more that we didn't even get to have the disccussion and/or see the facts about whether the test was racist because New Haven pre-emptively threw the test out. They did not look at the test itself, but instead threw it out based on fear of litigation, which the court rules against. As jonp72 and Kadin2048 note.
posted by lillygog at 8:25 AM on June 30, 2009 [1 favorite]

All this says — all of it — is that you need something more than the "threat of litigation" (like an actual belief that you've broken the law) before throwing out results and engaging in discriminatory behavior, for or against any group. You can blame New Haven for not trying harder to construct a case for the test being actually illegal, but they didn't do that. Their justification for throwing out the results was fear of litigation, and the case operated around that. posted by Kadin2048 at 10:30 AM on June

Kadin, do you get that the results were deemed enough of a reason for the city of New Haven to deem the test invalid because they saw in the results the potential of a lawsuit? doesn't mean New Haven didnt screw up royally. they should have proved why the test was bad enough to make them believe they could get sued over it.

what i am wondering is why did the white officers sue for "reverse discrimination" if the racist tenor of the testing wasnt the issue. because, really, this is the crux of the matter. they sued because they believe they were being discrimated for having a competitive advantage for being white. they didnt sue because the test was tossed out for being racist.

in other words, they sued to keep whatever competitive advantage those tests had for them --regardless of whether the test unfairly gave them that competitive advantage. in their eyes it was a bum rap for the city they got stuck with a test that gave the inner circle firefighters (all of them white) an advantage over their black and brown peers. they feel discriminated because the results were tossed afterwards and their results should have been held valid.

it's almost as if it were a "breach of contract" case, really.
posted by liza at 8:37 AM on June 30, 2009

if I understand correctly, the moral of the case is more that we didn't even get to have the disccussion and/or see the facts about whether the test was racist because New Haven pre-emptively threw the test out.

exactly.

although, am wondering if a lawsuit can be brought anyway against New Haven to prove the test was racist in the first place --because this case doesnt seem to deal at all with this issue.
posted by liza at 8:41 AM on June 30, 2009

although, am wondering if a lawsuit can be brought anyway against New Haven to prove the test was racist in the first place --because this case doesnt seem to deal at all with this issue.

Seems like it could; maybe the black firefighters could go ahead with the suit that the City was afraid of all along, and maybe that would get into the actual merits of the test.

What it seems to me is that the City of New Haven was trying to have its cake and eat it too: they didn't want to admit that their test was racist, but they also didn't want to piss off the black community for political reasons. So they decided to throw out the test not because it was racist — which would have involved admitting a mistake, putting them at odds with the testing company they hired, and probably forcing a renegotiation with the union over future promotion tests — but because they were afraid of getting sued. They wanted to get rid of the test results without admitting that they had spent a lot of money yet still come up with a racist test. In a way, it looks like they were trying to preemptively blame the black firefighters: "hey, don't look at us, we tried to make a fair test, but they were going to sue us, so we had to throw it out."

That's where the Court said 'no dice.' Apparently, if you want to throw out the test results, you have to show why the test itself is unfair and in violation of the law; you can't just decide the results are unpalatable. Maybe New Haven can still do this, I'm not sure.

So yes: the bottom line is that it sucks to be New Haven. Although they do have really good pizza, so that's something.
posted by Kadin2048 at 9:28 AM on June 30, 2009

Lillygog, I understand that part of it. By 'from what I'm reading' I meant the comments on this thread. Not the decision handed down by the judges.

All this talk about 'racially biased' tests on this thread and in the world at large just strikes me as not only nonsense but racist in and of itself. In essence they're saying one ethnicity is mentally incapable of achieving what another can. Thats garbage.

The funniest thing about this mess is that the one person that could possible cry discrimination about how the test was presented was the dyslexic, and he was one of the top scorers!

The people running New Haven are morons but then again you don't have to take a test to run for office.
posted by Nyarlathotep at 10:06 AM on June 30, 2009

Thought this was interesting (from Washington Independent article focusing on right-wing reaction, so the article itself isn't strongly focused on the ruling):

"Justice Ruth Bader Ginsburg, writing for the four justices in dissent, questioned the court’s “newly announced strong-basis-in-evidence” standard and recites in painstaking detail the evidence supporting the city’s decision. She went on to note that since the majority is announcing “a new legal rule,” then it should remand the case to allow the lower courts to apply it, since they didn’t have notice before that that’s what the rule was. “[T]he ordinary course is to remand and allow the lower courts to apply the rule in the first instance,” she wrote, chastising the majority for not following that usual course and instead deciding against the city of New Haven."

What I find interesting is that remanding the case would have allowed exactly that hashing-out of real facts we're sort of doing in this thread. I would have liked to see this done in court, with a real defense from the city, with wide consideration of facts pertaining to the test.
posted by lillygog at 10:17 AM on June 30, 2009 [2 favorites]

Context. At the turn of the century, states and the federal government employed Literacy Tests - tests that required you to show that you knew how to read, and a lot about civics, in order to vote. These tests appear race-neutral; after all, the ability to read and know civics is not determined by race, after all there are plenty of educated non-whites out there.

Effectively, Literacy Tests were employed to deny black suffrage, as blacks were very much less likely to know how to read or know civics than whites. The people who employed the tests knew as much, so this was in fact a racist test.

You have to understand the relationship between racist intent and effect, and how easy it is hide intent, but still have the desired effect; and how, in the legal system, you cannot always determine intent. But, if you see the effects, and they are striking, then it is often the only way to determine whether actions are racist.

Here is the problem: you have a test, and 118 candidates. 50 are not white. Statistically, 42% of the people who are at the top should also be not white, right?

In fact, you know that there are a lot of very qualified non-white test takers that you expect to place at the top of the test, because you have worked with them, you know they know their stuff, and they are better than even the top white guys.

When the test results come in, only two brown guys makes it for promotion. Two out of 50. 13 of 68 white guys make it. That's a 13% / 87% split, not the 42% / 58% you were expecting. Not even close.

Well, shit. Was the test racist? I don't know; but damn if it isn't a coincidence that only 2 brown people guy made it, and 13 white guys did, and not one black guy. The test may not be intentionally racist, but it looks like it could be effectively racist.

Well, what do you do? (A) We could just let the test results stand. Or, (B), we scrap the test because maybe it's not properly testing and emphasizing what's important in firefighting. Because the test results look flawed - and maybe test is flawed too. Better safe than sorry, right?

Nope. Fuck the brown folks. Dumb brown peole can't pass the damn test, and you know what, a bunch of white folks did. So what? That was a fair test - fair to the guys that made it to promotion, sure enough! So what if we all just coincidentally happen to be white? Maybe brown people need to get better educations! Not our fault they can't pass the test, it's their own damn fault.
posted by jabberjaw at 11:02 AM on June 30, 2009 [1 favorite]

I'm reading a few threads about this over at Volokh Conspiracy (I'm not a legal expert, though, and completely willing to acknowledge that some of this goes over my head), and it seems that the case wasn't ever argued in full trial. Also, the test in question had had the same suspect statistically-weighted results before, but in those years there had been enough openings that two or three candidates of color were promoted, and no-one had ever thought to look twice at the test.

This comment in particular rings true to me: that after the test first started showing results that seemed racially weighted in some way, the city should have done immediate due diligence, investigating the test, looking for alternatives, making sure that it wasn't unfair.

(For those not familiar with Volokh Conspiracy, I believe it's considered to be pretty conservative, and in the comments particularly there's often sharp partisan rhetoric for one side or the other, but I really appreciate the posts/comments that get into legal reasoning. They're not always views I'd agree with, but quite interesting.)
posted by lillygog at 11:30 AM on June 30, 2009 [1 favorite]

Here is the problem: you have a test, and 118 candidates. 50 are not white. Statistically, 42% of the people who are at the top should also be not white, right?

No, dang it. That isn't how statistics work at all. To name just one problem, you're assuming that all the candidates have equal knowledge, aptitude, and study ethic or at least that these things are exactly equal across lines of ethnicity. That may well be the case but you're simply begging the question, and it's the central question as to whether disparity in outcome can be seen as a prima facie case for disparity in opportunity. Of course if you assume it the test must have been racially biased. But whether it's actually the case is exactly the question in dispute.

I do think people are eliding the great importance of book knowledge in being a fire captain or lieutenant. Yes, it's very important to be a good leader of men. But you damn well better know, for example, technical details of how various building materials burn in a fire. You better know, quickly and without much thought, various technical details about the equipment your men are using. You better know quite a lot of sciency stuff about fire itself. The safety of your men depends on it. There are a LOT of critical, testable facts that you need to know and the willingness to pooh pooh that strikes me as disingenuous.
posted by Justinian at 11:40 AM on June 30, 2009 [5 favorites]

All this talk about 'racially biased' tests on this thread and in the world at large just strikes me as not only nonsense but racist in and of itself. In essence they're saying one ethnicity is mentally incapable of achieving what another can. Thats garbage.

It may be a little more subtle. First, because as some people up thread have mentioned, it may be as much a class issue as a race issue, and to hear that people of lower socioeconomic status have lower test scores is less contentious (than scores varying significantly between races) because it can be ascribed to environmental factors rather than some intrinsic quality of the people being tested.

Second, there can be other factors in play, like stereotype threat. In this case, if the testing process failed for whatever reason to eliminate some stereotype threat that may have influenced those being tested, that could be considered a kind of implicit bias. A company that exists to do personnel testing should be aware of these issues and try to deal with them, but that does not guarantee that they will be successful.
posted by benign at 11:43 AM on June 30, 2009

I'm not really interested in getting to the underlying issues here, but there is one thing worth pointing out about the test. The questions were drawn directly from the study materials, which had also included answers. If there was ever an example of teaching to the test, this was it. It's a particularly stupid way to measure anything, independent of all other concerns. It also allows for partisanship in obvious ways (all it takes is someone to tell another "make sure you really memorize those questions, ahem.")

Nonetheless, this is a very technical and fact-specific issue. There's pretty much no way to stop this from devolving into accusations and retorts about the value of affirmative action, testing, remedial efforts to address institutional racism, etc. At the very least, let's just remember that this test was incredibly stupidly designed, even if great efforts went into its construction. The majority focused on the efforts that went into making it race-neutral, the dissent focused on the impact. Yet in some sense this fact pattern is particularly ill-suited for this distinction. I mean, literally using the questions from the practice materials is just not at all acceptable. I don't know where to draw the line, but really, all this measured was rote memorization. It's entirely likely that were this test administered again in other context, minority candidates were strongly outperform White candidates, with equally little validity.
posted by allen.spaulding at 11:54 AM on June 30, 2009

allen.spaulding: has there ever been a case where a written test was biased against white people?
posted by Iax at 12:13 PM on June 30, 2009 [1 favorite]

allen.spaulding: has there ever been a case where a written test was biased against white people?

If you're taking the same standard used in this thread where disparity in outcome is viewed as ironclad evidence of disparity in opportunity then yes, absolutely, without a doubt.
posted by Justinian at 12:35 PM on June 30, 2009

But where and when?
People have listed a bunch of reasons here why this test and others have been biased against minorities. Basically all the reasons just seem to boil down to class issues; not knowing some vocabulary or not having enough free time to study, stuff like that.
How would someone write a test to fail out well off white people?
posted by Iax at 12:44 PM on June 30, 2009

Well, that isn't what you asked. You asked if there has ever been a case where a written test was biased against white people, full stop. Assuming you mean a test where white people scored significantly lower than another ethnic group that's an easy one: Pretty much every single standardized test measuring mathematical ability given to students in the United States shows a significant gap in the scores between Asian-American and Caucasian students. By the standard presented in the thread (difference in outcome is a prima facie case for difference in opportunity), these tests obviously are biased in favor of Asian-American students.

Now of course in the real world we would argue that this isn't the case and the difference is due to other factors. Which is why to be intellectually honest we have to allow for that fact in the firefighter's case as well. Maybe the white firefighters simply studied harder. I don't know and neither does anyone else in this thread.
posted by Justinian at 1:26 PM on June 30, 2009 [1 favorite]

"allow for that fact" should obviously be "allow for that possibility", of course, since we don't know whether it is a fact or not. Without having access to the test and having more information about the availability of study materials, etc, we simply don't know why the results came out the way they did.
posted by Justinian at 1:27 PM on June 30, 2009

This issue is so interesting. I don't really know much about or follow legal matters but this has definitely captured my attention.

New Haven should have utilized one of those Assessment Centers, which would have given a better idea of overall of each man's ability. Many people (including the dyslexic man, Ricci, who scored very well), don't test or write well. I don't know much about firefighting, but I would assume that paper-based tests are not a frequent job duty. I'm sure paperwork is, but I doubt it needs to be completed under the same level of stress that accompanies test-taking. A firefighter needs to be clear-headed when faced with the stress of fighting fires, and an assessment to determine promotion should take that into account much more than anything written on paper.

The desire to to have a racially-balanced group of Lts and Cptns (I believe those were the ranks) is absolutely admirable. But . . . to throw out the results of the test because you did not like them is not fair to the men who did score the highest.

I am an advocate for Affirmative Action. I believe that universities should have some kind of "quotas" or at least incentives to accept more minority and/or economically disadvantaged (which is probably all of us these days) students. But in a case like college admissions, acceptance is based on much more than one single number. While SAT and GPA scores count, recommendations and the reality of the student's background (along with what they made of it) can also be considered. College admissions are represented as being multi-faceted, and therefore can take more issues into account when deciding who to admit.

The decision of which fire-fighters to promote was not multi-faceted (though I think it should have been) nor was it represented as such. Everyone who took the test knew this. The parameters were strictly drawn. With no ability to prove that the test was "intentionally racist" (in jabberjaw's words), I believe the results must stand.

However, the test may have been "effectively racist" (that's hard to judge from where I am, though, and depends on who "really deserved" the promotions -- and that may be hard to judge from anywhere). At the very least the test was not the best way to determine promotions. The method of picking officers needs to change, and I've got to believe it will after this (does anyone know how often these tests are given? Perhaps it was in there somewhere and I missed it).

But for now I believe it's right to allow the men who scored highest to receive their promotions. According to the paradigm in place, they won them fair and square.

In the future, New Haven, change the damn paradigm, for heaven's sake.

(Though I'm sure that will lead to more problems . . .)
posted by imalaowai at 2:02 PM on June 30, 2009

Here is the problem: you have a test, and 118 candidates. 50 are not white. Statistically, 42% of the people who are at the top should also be not white, right?

> No, dang it. That isn't how statistics work at all. To name just one problem, you're assuming that all the candidates have equal knowledge, aptitude, and study ethic or at least that these things are exactly equal across lines of ethnicity. That may well be the case but you're simply begging the question, and it's the central question as to whether disparity in outcome can be seen as a prima facie case for disparity in opportunity. Of course if you assume it the test must have been racially biased. But whether it's actually the case is exactly the question in dispute.

Agreed - which is why statistics wasn't the only factor I mentioned. Other factors are that some people that were expected to succeed, did not. I think the point is that - nobody really knows whether the test was discriminatory or not. Can you really quantify that? With a disparity in expected outcome, which happens to walk down a racial line, can you reasonably infer that the test was potentially discriminatory? I think yes.

People make assumptions that it was not, and that white people got shafted. Other may assume that with the unbalanced outcome, it must have been. What New Haven's authorities did was perhaps the lazy thing: instead of actually trying to figure that out, they took all of these factors (including the potential threat of litigation) and said fuck it, let's have a do-over.

A new test altogether. There was no guarantee that their new test would necessarily be any less discriminatory; only that by slightly shifting the emphasis from written-testing to on-your-feet testing, blacks and hispanics might do better.

Perhaps this shift was ultimately the right thing to do; not because it favored one race over another, but because it actually properly tested the skills necessary for promotion. Perhaps it took this disparate impact for the authorities to take a second look at a bad test.
posted by jabberjaw at 2:43 PM on June 30, 2009

Well if one thing comes out of this I hope it's that people reconsider the crazy-ass things they do to avoid getting sued, even if they actually would have won the lawsuit.
posted by delmoi at 5:03 PM on June 30, 2009

This is why I'm choosing to raise my kids white.
posted by Eideteker at 5:54 PM on July 1, 2009

As Frank Ricci prepares to testify against Sonia Sotomayor at next week's confirmation hearings, Dahlia Lithwick questions whether he can be characterized as a serial plaintiff.
posted by the littlest brussels sprout at 8:21 PM on July 10, 2009