Toobsllik,
I fully stand by my assertion that statistics can be manipulated to show whatever you want. I could take the same set of data from your link that shows such disparity between races, and manipulate it in such a way that shows that any disparity is statistically insignificant. Or even introduce alternative explanations for any disparity. No one is arguing that standardized test disparities exist. It is the INTERPRETATION of the test data, and an attempt to explain what it shows that is in question here. Considering your link is from an admittedly racist site, the conclusions drawn must be taken with some skepticism. By the way, I tried to find verifications for the numbers listed in your linked site, and the actual study from which the numbers and breakdown supposedly came from, and I cannot verify that the numbers used (including the breakdowns based on wages) exist anywhere away from that site.
Because I am technologically impaired, I am unable to provide links to the various quotes I am including. In addition, one is from a recent sociological textbook for which there is no internet link (or at least none that I am aware of). Again, I state that there are other conclusions from "bona fide" studies that directly contradict the conclusions from your rascist site.
________________________________________________
"The misuse of test scores for purposes beyond which they have been validated have had a systematic adverse impact on minority applicants to higher education. The differences in the performance of black and white test takers are a magnitude of approximately one standard deviation in each of the admission testing programs. Much of this significant difference in performance can be attributed to environmental and societal factors that neither reflect an individual's level of achievement nor his/her capacity to achieve if given the opportunity."
_________________________________________________
"Demands for allegedly race-neutral criteria of "merit" and "standards" fail to recognize this country's historical and ongoing record of institutionalized discrimination. (Current examples are plentiful, including unequal school funding, red-lining and housing segregation, and discrimination in job promotion.) Moreover, the standardized tests used to determine "merit" are notoriously biased in favor of those with higher socio-economic status or those whose parents have a higher level of education. Skin color and circumstance of birth, not merit and ability, are undeniable factors in such tests."
__________________________________________________
The strongest claim that can be made for standardized tests is a correlation between performance and scores at the very highest and lowest levels. In the vast middle range, where the majority of students fall, the predictive value is no more reliable than the flip of a coin.
_______________________________________________
"The only objective part of most standardized tests is the scoring, when it is done by machine. What items to include on the test, the wording and content of the items, the determination of the "correct" answer, choice of test, how the test is administered, and the uses of the results are all decisions made by subjective human beings ("FairTest: What's wrong with standardized tests?" May 5, 1998, p. 1). "
__________________________________________________
"An assessment test is considered biased when it does not predict the performance of specific groups in comparison to the performance of the predominant group (Garcia & Pearson, 1994). Although overt bias in tests, to a certain extent, is being addressed by test producers, most forms of bias are subtle and tend to remain in the very structure of tests. Essentially, how the test is normed, what questions are asked, how the test questions are asked, and who determines the correct answers produce a biased process through which people are ranked, sorted, differentiated, and then compared.
Norming is the process of developing a standard to which test-takers can be compared. Often, in applying strict probability samples of the nation, for norming purposes, only small numbers of people of color are represented (Garcia & Pearson, 1994). Consequently, the process of norming has a tendency to reflect the dominant culture (Garcia & Pearson, 1994).
In the norming process, specific trial questions are tested for their correlation with the total test score. Those questions that have a low correlation with the total score are removed based on the assumption that "they provide a poor estimate of the phenomenon being measured" (Garcia & Pearson, 1994, p. 343). Invariably, the questions that are eliminated are those questions on which a small number of low-scoring individuals do relatively well (Garcia & Pearson, 1994). Not surprisingly, people of color are over-represented in the group of low-scoring individuals (Garcia & Pearson, 1994). Therefore, people of color are apt to do poorly on the authorized final version of the test instrument.
When the Stanford-Binet test (the United States version) was developed, it was normed on middle-class white
_________________________________________________
In the vast middle range, where the majority of students fall, the predictive value is no more reliable than the flip of a coin"
IT is very dangerous to make conclusions based on race when ALL factors are not considered. Many teachers in Germany were all for Hitler's reforms because he was able to manipulate data to "prove" that Aryans were a superior race. But then, Tooblsllik, you probably agree with him.
Also, on a sidenote, what is your fascination with the word "anachronistic" about? You use the word often in your posts, but unfortunately it doesn't make sense in the context in which you use it.
