Are you in anyway familiar with Cobb County, Georgia?

Why do you believe he was chosen to evaluate the test?

Maybe Newt Gingrich recommended him. Or maybe originally they wanted the outgoing police commissioner in Dekalb but he shot the incoming police commissioner. It's tough getting evaluators in Atlanta.

Glenn, this is the sub-title of your post:

“Four Supreme Court Justices agree with Sotomayor, including the one she is to replace.”

It reminds me an old joke. The athletes A and B participate in a competiton. The athlete A wins. But one of the “objective”reporters presented it to the public this way: the athlete B took the second place.

If Obama had nominated a white male to the Supreme court, and that white male had made the same decision on Ricci case as Sotomayor, do you think the right wing republican would be making a big fuss over this case? I doubt it.

One way to do it is to ban anyone from taking practice tests.

....but does anyone here think the CIA acted without Obama's knowledge in orchestrating and carrying out the Zelaya coup?

In what way does taking a practice test, or allowing test takers to take a practice test, bias a test?

It does not bias the test, it bias the outcome. In most cases it allows people with more money to spend money practicing for tests, while those who cannot afford it cannot spend the money taking tests. Why does anyone need to take a practice test. They should know the material. I feel the same about SAT tests.

Many of teh white guys who took the test in the Ricci case spent money taking practice tests.

Thanks for the clarification on bias. You said, test. You meant, outcome.

Do you see any way to effectively ban practice tests? Or, an effective way to know that a test taker took a practice test?

What's the matter, you don't dig the Empire? The Force Monopoly decides what Fair is. We need a massive bureaucracy to make sure we are all Fair, to each other. The bureaucracy is so into fairness that it is spreading its love of all things Fair far and wide, to all those Unfair. Except for our friends, of course, who are Fair Enough.

That is why we have the Federal Government, working its hardest to uphold the Constitution, a focal point fueled by our infinite passion for Fairness.

I was talking statistics. If the test scores are the only data you have with which to determine whether the test was biased or not, then you need to use a test to determine how likely the result was, given that the exam was unbiased. Having done that, you calculate a confidence interval that represents the possibility that you made an error in that determination, (e.g. because of sample size), and if your actual scores are outside that interval, the exam is biased, in the same way that, say, survey results are accurate. No if's, and's or but's. There are rough measures based on the average size of such test samples and average exams. That's how the government came up with their suggested method in the first place. For Walfisch to claim that just because the sample size was 77 it is impossible to know anything about a statistical bias, as opposed to being able to find a specific question that said, "All fire chiefs should be white, T/F" is just ignorant. If statistics could never be done on samples of size 77, the world would be a much different place.

When Congress prohibited discrimination based on race and gender, employers developed "neutral" criteria to select and promote individuals in a manner that excluded minorities and women. For example, employer created lifting restrictions for jobs that did not require lifting, high school diploma requirements for janitors, etc. To address these types of discriminatory policies (policies that are neutral on their face, but have a disparate impact on an impacted group), courts developed a disparate impact theory of liability. Under this theory, where a certain policy has a disparate impact on a protected class, the employer needs to establish a legitimate, business reason for the criteria that causes the disparate impact. For example, that the lifting restriction or high school diploma is necessary for the job.

Alternatively, some promotion policies in their implementation have a significant disparate impact on a protected group. Plaintiffs cannot identify the exact criteria that causes the impact, but they can statistically exclude all legitimate busines reasons for the difference and still show a stastically significant disparate impact. For a giant education on this type of case, review the 9th Circuit decision in Dukes v. Wal-Mart. In that case the 9th Circuit upheld certification of a class action by all female Wal-Mart employees who claimed that Wal-Mart's promotion policies had a disparate impact on women. While the plaintiffs could not identify the specific criteria that caused the impact (other than manager discretion), they could statistically argue that even after factoring out all legitimate reasons for differences in promotion rates, women were not promoted at a greater rate than men.

Given this background, the City of Newhaven was faced with a selection criteria (a test) that had an adverse impact on a protected class. They could not validate that the test did not have a disparate impact for legitimate reasons closely tied to the job. In addition, they had information that a test more closely tied to the job duties of a captain, etc. would not have an adverse impact. Consequently, to avoid the disparate impact lawsuit, they decided to look at a test that tied more closely with job duties (a practical rather than written test), and that had been validated as not having an adverse impact on minorities.

The City was faced with a true damned if you do damned if you don't situation. No matter what choice they made, a lawsuit was inevitable.

bybstander

Many of teh white guys who took the test in the Ricci case spent money taking practice tests.

Please remember that we use English here.

Even if, in bernbartian, the above sentence would read: "Pseale remmber weh sue Enhglish hure."

You really make things difficult (yuo raely mkea thigngs difiucilt, darm that cmpouter!").