SAT Math & Verbal Scores by Race
While many politicians and educators lament the poor performance of black Americans in schools and on standardized tests, the magnitude of the differences in the scores between blacks and Whites, and particularly between black women and White men, is rarely mentioned and even less understood. But it doesn't take much investigative research to discover what the magnitude actually is, and a little more to understand what it means. To merely note that, between races and sexes, there's a 174 point difference in *average* SAT Math scores and a 77 point difference in *average* SAT Verbal scores, a 251 point difference which itself is rarely mentioned in the news, is almost meaningless unless the distribution or standard deviation of these scores are also considered.
You don't need to be a mathematician to understand what is meant by "distribution" and "standard deviation". It's easy enough to just remember that two thirds of the test takers score within a "standard deviation" or about 100 points of the median score, 95.4% score within 200 points, and 99.7% score within 300 points. This means that two thirds of black women score between 326-510 in math and 335-533 in verbal, whereas two thirds of Asian men score between 471-713 in math and 383-637 in verbal. It also means that only 16% of black women score higher than 510 in math and 533 in verbal, 2.3% score higher than 602 in math and 632 in verbal, and 0.15% score over 639 in math and 731 in verbal. Conversely, 16% of Asian men score higher than 713 in math and 637 in verbal, 2.3% score higher than 834 in math and 764 in verbal, and 0.15% score higher than 955 in math and 891 in verbal. [note: this is the theoretical score, as the actual top score is 800. By "recentering" the test in 1995, the College Board left thousands of White and Asian students off the scale while leaving a huge gap between 0 and 200].
Of the 91,130 Black women who took the 2003 SAT Math Test, 137 scored over 694. But 11,179 White women scored over 717, 65,416 White men scored over 657 with 9,523 scoring over 763, and 10,221 Asian men scored over 713, with 1,488 of them scoring over 834, a score achieved by none of the Black women. And 11,999 Asian women scored over 676, 744 Mexican men scored over 682, 57 Mexican women scored over 717, 149 Puerto Rican men scored over 679, 11 Puerto Rican women scored over 725, 595 Hispanic men scored over 704, 46 Hispanic women scored over 747, 149 Indian men scored over 712, 11 Indian women scored over 760, 595 "Other Men" scored over 768, and 699 "Other Women" scored over 716. Overall, 101,641 other students scored an average of 712, which is 18 points higher than these top 2 1/2% of Black women.
At the high end, the top 2 1/2% of each race and sex, 621 White men scored 175 points higher, 729 White women scored 123 points higher, 49 Mexican men scored 91 points higher, 57 Mexican women scored 23 points higher116 Black men scored 41 points higher, 97 Asian men scored 261 points higher, 114 Asian women scored 218 points higher, 10 Puerto Rican men scored 90 points higher, 11 Puerto Rican women scored 31 points higher39 Hispanic men scored 120 points higher, 46 Hispanic women scored 53 points higher, 10 Indian men scored 125 points higher, 11 Indian women scored 66 points higher, 39 "Other men" scored 191 points higher, and 46 "Other women" scored 134 points higher. Overall, 2,131 other students scored an average of 124 points higher than the highest scoring 137 Black women students.
Extrapolating the scores shows that the top scoring black woman student would have scored only 700, 12 points below the average achieved by 101,641 other students and 118 points lower than that achieved by 2,131 other students.
Where the sex differences are significant for other races on SAT Math, the sex differences between Blacks are small, so math skills for Black men follow essentially the same pattern as Black women. However, where most men of other races score higher than women on the SAT Verbal test, Black men score 6 points lower than Black women, meaning that the differences in verbal skills between, say, Black men and White men are even greater than the differences above.
With this in mind, you must wonder what it is about Colin Powell, who is known to have been jumped over 32 other more highly qualified generals to become commander in chief, and Condaleeza Rice, that they were appointed to their current positions. With more than 100,000 students taking the SAT each year who are obviously far more qualified, both verbally and mathematically, than both Rice and Powell, what characteristic was it that made them the favored candidates?
Was this decision based on the fact that they WERE unqualified, in order to convey a false impression about the progress of Black Americans, or did they really possess some other skill which made it mandatory to deny millions of other far more qualified Americans this opportunity? Their performance to date suggests to this author that they're even more unqualified for these positions than the above differences in test scores would indicate.
If you were to set a hiring limit on this year's graduates of 700 on the SAT math test, you'd have one Black woman to choose from, but more than 101,641 students of other races and sexes who would be far more qualified. What this means is that the number of other races and sexes who possess easily quantifiable and verifiable skills which are important to employers outnumbers the number of black women who possess such skills by 101,641:1. In a non-discriminatory work environment, where affirmative action is illegal, where the ability to do the job and communicate are the determining factor for employment, and where the employer wants to hire the best 7% of college graduates, you would expect to find no black women at all, particularly if the employer had a commitment to the board of directors to avoid litigous applicants. Black women are historically litigous, which is more than enough reason to refuse to hire them.
Even ignoring this vital point, the notion that black women should represent 6% of any workplace just because they're 6% of the population is clearly a destructive notion. Not only will a company be forced to hire the least competent employees, but they will be prevented from hiring, or forced to fire, far more qualified and competent employees.
IS THE SAT FAIR? YOU DECIDE...
The following data is published by the College Entrance Examination Board, and/or the Educational Testing Service, as reported by FairTest, a wonderful non-profit organization in Boston.
The Gender Gap... Is the SAT fair to women?
Studies have shown that women slightly outperform men in their first year of college. So why does the SAT, a supposedly "fair" test, yield 45-point lower scores to women?
1994 SAT Test Scores for: Women Men Average for ALL races 881 926
The Ethnic Gap... Is the SAT fair to minorities?
Now let's take a look at similar data for ethnic groups. The differences are even larger!
1994 SAT Test Scores for: Asian American 951 Caucasian 938 Mexican American 799 Black 740 Average for ALL races 902
The Income Gap... Is the SAT fair to middle class and poor students?
Perhaps the most interesting data is the correlation between test scores and family income. On the average, well-off students score 234 points higher than students whose familes earn under $10,000 per year.
1994 SAT Test Scores for:
under $10,000/yr 766 $10,000 - $20,000/yr 812 $20,000 - $30,000/yr 856 ... $50,000 - $60,000/yr 929 $60,000 - $70,000/yr 948 over $70,000/yr 1000
We think these numbers (along with other facts publicized by FairTest) mean that your SAT score is a better prediction of your income, race, or gender than it is of your ability to perform in college. Which is precisely why we work so hard to bring low-cost and free SAT preparation methods to as many students as possible at the lowest possible cost.
Modified Tuesday, November 17, 2009
Copyright @ 2007 by Fathers' Manifesto & Christian Party