This morning the Chronicle of Higher Education reports on a new report from the College Board on the demographics of its Advanced Placement exams. Once again (similar results last year), the overall numbers are encouraging; the numbers for blacks are not. And the College Board blames, well, everybody (except for the tests and the test-takers).
First the numbers:
… 24.9 percent of the 2.8 million students who graduated from American public high schools in 2007 took at least one AP test, and 15.2 percent of them earned a score of 3 or higher on at least one test. Those numbers are up slightly from the previous year…. [NOTE: This isn’t completely clear, but a check of last year’s article reveals that the 15.2 percent refers to all high school graduates, not 15.2 percent of the 24.9 percent who took AP classes – jsr.]
Underrepresentation of African-Americans
However, only 3.3 percent of the students who scored 3 or higher on a test were African-American, despite the fact that black students represented 14 percent of all high-school seniors last year….
African-American students also are less likely than their peers to take AP classes….
Black students accounted for only 7.4 percent of AP test takers last year, according to the report. White students, by contrast, accounted for 61.7 percent of test takers and 64 percent of graduating seniors.
In many states, American Indian and Hispanic students’ participation matched their representation in the student body. Nationally, Hispanic students made up 14.6 percent of the high-school-senior population, and 13.6 percent of them scored at least a 3 on an AP test.
In short, black students were significantly “underrepresented” among AP test takers and among those doing well on the test.
Trying to explain this “underrepresentation” is a great challenge to our country, requiring the efforts of, among others, our best scholars (such as Abigail and Stephan Thernstrom in their impressive book, NO EXCUSES: CLOSING THE RACIAL GAP IN LEARNING). What doesn’t help, however, is the moralistic finger-pointing engaged in almost reflexively by representatives of elite institutions such as Trevor Packer, the director of the College Board’s Advanced Placement program.
The board sees a “true and startling lack of equity,” Mr. Packer said. “African-American students in particular are not receiving encouragement and support.”
Does Mr. Packer have any evidence, beyond the “underrepresentation,” that black students “are not receiving encouragement and support”? Who does he believe is guilty (and if there truly is a “true and startling lack of equity,” it is guilt we are talking about) of not providing the missing “encouragement and support”? Teachers? School administrators? Parents? Peers? If you’re going to point your finger at shortcomings in equitable treatment, it at least ought to be clear whom you’re pointing at.
Finally, it would be nice to know whether Mr. Packer believes that Asian-Americans, who are no doubt “overrepresented” among the high achievers, have been receiving a disproportionately and hence inequitably high level of “encouragement and support.”
Perhaps what the College Board should propose is an Equiable Support and Encouragement Redistribution Act, taking some equitable treatment away from those who receive an excess of it and redistributing it to those who are not given enough.
John said,
[Does Mr. Packer have any evidence, beyond the “underrepresentation,” that black students “are not receiving encouragement and support”? Who does he believe is guilty (and if there truly is a “true and startling lack of equity,” it is guilt we are talking about) of not providing the missing “encouragement and support”? Teachers? School administrators? Parents? Peers? If you’re going to point your finger at shortcomings in equitable treatment, it at least ought to be clear whom you’re pointing at…..Finally, it would be nice to know whether Mr. Packer believes that Asian-Americans, who are no doubt “overrepresented” among the high achievers, have been receiving a disproportionately and hence inequitably high level of “encouragement and support.”…..Perhaps what the College Board should propose is an Equiable Support and Encouragement Redistribution Act, taking some equitable treatment away from those who receive an excess of it and redistributing it to those who are not given enough.]
John, it is interesting that you make note of the Asian Americans regarding this item, since the article did not highlight the performance of Asian Americans, but only blacks,Latinos and whites. Asian Americans are “overrepresented” in the numbers of the HIGHEST performers on the AP Tests, NOT BECAUSE, they have been receiving a disproportionately and hence inequitably high level of “encouragement and support.”. I agree with you. Now, the AP Tests will be devalued as a metric for evaluating proficiency and achievement, as are all of standardized testing, simply because they cannot achieve equal results or equal outcomes among the racial and ethnic groups by using the same lame and apologetic EXCUSES.
Please read the NY Times article on this today.
http://www.nytimes.com/2008/02/14/education/14exam.html?th=&emc=th&pagewanted=print
Larger Share of Students Succeed on A.P. Tests
ON A RELATED TOPIC, HERE ARE THE ACTUAL RACIAL TEST SCORE GAP NUMBERS ON THE SAT I TESTS FOR 1995. THESE RACIAL GAPS HAVE NOT DECREASED IN RECENT YEARS, BUT IN FACT HAVE INCREASED.
From the book, “America in Black and White: One Nation, Indivisible” by Stephan and Abigail Thernstrom, Simom and Schuster, 1997 in Chapter 14, titled “Higher Learning”, in Table 4 labeled, “Number and The Racial Scoring Gap in the SAT I Math and SAT I Verbal Tests (Black-White, Black-Asian, and White-Asian SAT I Test Score Gaps) are below for 1995 and these Gaps have not narrowed since.
Percent of Black, White, and Asian Students with High SAT Scores, 1981 and 1995”: Source; The College Board, Ethnic Data on Scoring, 1981 and 1995, the figures and percentages for each score level are charted.
For example, in 1995, for 103,872 Black test takers of the SAT 1 Test, in the Math, 107 Blacks scored between 750 and 800, 509 Blacks scored between 700 and 749, 1,437 Blacks scored between 650 and 699. Total > 650 for Blacks was 2,053 or 2.0% of all Black test takers. Total > 700 was 616 or 0.6% or six tenths of 1 percent. Total > 750 was 107 or 0.1% or one tenth of 1 percent.
In 1995, for 103,872 Black test takers, in the Verbal, 184 Blacks scored between 700 and 800, 465 Blacks scored between 650 and 699, and 1,115 Blacks scored between 600 and 649. Total > 600 was 1,764 or 1.7% of Black test takers. Total > 700 was 184 or 0.15% or less than two tenths of 1 percent.
In 1995, for 674,343 White test takers of the SAT 1 Test in the Math, 9,519 Whites scored between 750 and 800, 29,774 Whites scored between 700 and 749, and 51,306 Whites between 650 and 699. Total > 650 for Whites was 90,599 or 13.4% of all White test takers. Total > 700 was 39,293 or 5.8%. Total > 750 was 9,519 or 1.4%.
In 1995, for 674,343 White test takers of the SAT 1 Test, in the Verbal, 8,978 Whites scored between 700 and 800, 19,272 scored between 650 and 699, and 36,700 Whites scored between 600 and 649. Total > 600 was 64,950 or 9.6%.Total > 700 was 8,978 or 1.3%.
In 1995, for 81,514 Asian test takers of the SAT 1 Test in the Math, 3,827 Asians scored between 750 and 800, 7,758 Asians scored between 700 and 749, and 9,454 Asians scored between 650 and 699. Total > 650 for Asians was 21,039 or 25.8%. Total > 700 was 11,585 or 14.2%. Total > 750 was 3,827 or 4.7%.
In 1995, for 81,514 Asian test takers of the SAT 1 Test in the Verbal, 1,476 Asians scored between 700 and 800, 2,513 Asians scored between 650 and 699, and 4,221 Asians scored between 600 and 649. Total > 600 was 8,190 or 10%. Total > 700 was 1,476 or 1.8%.
Therefore, in reference to the above data for 1995, Asians out perform the other two groups at the highest levels of the SAT 1 scores in terms of rate of attainment or percentage of the total group at each score level above 650 and 700 and above in both the Math and the Verbal of the SAT 1 Test. In fact, in the 1999 data given by the College Board: Performance by Ethnic Groups, the rate of attainment or percentage of the total group at each score level above 650 and 700 and 750 and above has risen for the Asian group both independent of and relative to the other two groups.
In 1995, there were only 107 Blacks with a Math score of 750 or above or 0.1% (one tenth of 1 percent) of the total number of Black test takers.
There were 9,519 Whites with a Math score of 750 or above or 1.4% of the total number of White test takers.
There were 3,827 Asians with a Math score of 750 or above or 4.7% of the total number of Asian test takers.
Asians out perform Whites at 3.4 times the rate at which they score 750 or above (4.7% vs. 1.4%).
Asians out perform Blacks at 47 times the rate at which they score 750 or above (4.7% vs. 0.1% or one tenth of one percent).
More recent data of this nature is not released by the College Board to the general public any more because this data is racially sensitive and *politically incorrect*, but it is the damn truth.
http://www.lagriffedulion.f2s.com/testing.htm#APPENDIX%20B
More SAT data may be found in Appendix B. There, you will discover that Asians mostly sit on top of the heap; that whites, Mexican Americans and blacks follow in that order. Some details prove interesting. For example, whites enjoy a verbal advantage over Asians that disappears at high levels of income and social advantage. Regrettably, the College Board no longer discloses these data. In 1996, they stopped publishing performance by income and parental education disaggregated by race and ethnicity
I wonder when minority community leaders will cease the unholy trade of race preferences in lieu of education. I say leaders because 75% of black parents support vouchers, they recognize this abombination. But minority community leaders don’t have their contituents interests at heart, they have their own. Their actions prove they are perfectly comfortable failing to educate children as the price of maintaining their electoral coalition which is hugely dependent on the education system for votes, money, activists, and indocrination.
I wonder if Obama’s campaign may effect this in the black community. Black leaders disproportionately endorsed Clinton, primarily due her institutional power and past political favors. Yet those endorsements have virtually no effect on black voting patterns. Maybe this is a step toward replacing the black leadership in this country which is out of step with its constituents.
So how does one account for the advantage enjoyed by a mixed white and Chinese student? Are they doubly advantaged?
E wrote: “More recent data of this nature is not released by the College Board to the general public any more because this data is racially sensitive and *politically incorrect*, but it is the damn truth.”
That is only partly true. The raw data or the decile cohorts are no longer published, but close assumptions of the data can be extracted from the published information via the statistical data given.
Further more, it is reasonable to presume that the data is no longer disclosed because it might be used against the methods and activists in affirmative action politics today. The reports have always shown a race gap for as long as comparative data has been collected. It has always been politically incorrect and has been published nontheless in order to proove the necessity for action.
I believe it is no coincidence, that the change in strategies occured in a time when criticism of special need politics ceased to be a taboo. For as long as any criticism of black and minority politics was an untouchable taboo the protagonists were safe from being challenged on the topics of effectivity and objective measures. The complaints about the existence of the race gap were seen as proof of further need and nobody dared to question the methods implied. Now that criticism is no longer a taboo, the apparatchicks face danger from two sides: the question of justification in respect to the 14th amandment and the question of effect on behalf of the people whom they are supposed to help. Minority politics have always been argued for as a crude but effective method where the violation of the civil rights of white and asian people have been justified as necessary to help black people and other minorities overcome the rage cap. But what if that is just another lie, what if minority politics is in fact not helping minorities?
It seems that is a painful question, too painful to be answered. In a strange coincidence the minority poilitcs representatives and instituions applying the criticised methods have almost simultaniously started to hide away any information that might help proove just how useful minority politics really are at about the time when critics were starting to tear down the taboo.
In the most recent publication of the Collegeboard, the Total Profile Report of 2007, there is enough Information to partially retrieve the numbers you were looking for. Of course, this cannot replace the actual raw data, but it will give a reasonable estimate of what to expect.
http://www.collegeboard.com/prod_downloads/about/news_info/cbsenior/yr2007/national-report.pdf
On the ethnic analysis page we find the results for the three subtests as averages and standard deviations (table 8).
Critical Reading:
Asians: Avg 514 SD 124
Whites: Avg 527 SD 102
Blacks: Avg 433 SD 97
Mathematics:
Asians: Avg 578 SD 123
Whites: Avg 534 SD 102
Blacks: Avg 429 SD 97
Writing:
Asians: Avg 513 SD 121
Whites: Avg 518 SD 100
Blacks: Avg 425 SD 93
The large sample sizes of 72109 Asians, 828038 Whites and 159849 Blacks allow for a reasonable precision with the statistical standard errors of .46 .11 and .23 respectively for asians, whites and blacks diminishing when building larger clusters.
When looking at the top end of the distributions, it is important to not forget, that a lower average can easily be compensated by a greater standard deviation. This happens to the asian group in the critical reading and writing disciplines. In both cases the difference in average ammounts to 10,4% and 4,1% of the Asian Standard deviation and therefore the “rank” is reversed when focusing on the top end of the scale.
Further more it is noticable, that the standard deviations are very consistent and that the black group average is consistantly about one standard deviation below the white group average.
Looking at the cohorts of +700 and +750:
1. Presuming normal distribution, the cohort 700+ is situated at a x SD for each group and discipline:
Critical reading:
Asians: 1.5
Whites: 1.7
Blacks: 2.8
Mathematics:
Asians: 1.0
Whites: 1.6
Blacks: 2.8
Writing:
Asians: 1.5
Whites: 1.8
Blacks: 3.0
Therefore the percentage of each group in the 700+ cohort can be assumed at:
Critical reading:
Asians: 6.7 %
Whites: 4.5 %
Blacks: 0.3 %
Mathematics:
Asians: 16.1 %
Whites: 5.2 %
Blacks: 0.3 %
Writing:
Asians: 6.1 %
Whites: 3.4 %
Blacks: 0.2 %
And in absolute numbers of students who achieved more than 700 points according to group sample size:
Critical reading:
Asians: ~ 4800
Whites: ~ 37200
Blacks: ~ 470
Mathematics:
Asians: ~ 11600
Whites: ~ 42900
Blacks: ~ 410
Writing:
Asians: ~ 4400
Whites: ~ 28500
Blacks: ~ 250
2. Presuming normal distribution, the cohort 750+ is situated at a x SD for each group and discipline:
Critical reading:
Asians: 1.9
Whites: 2.2
Blacks: 3.3
Mathematics:
Asians: 1.4
Whites: 2.1
Blacks: 3.3
Writing:
Asians: 2.0
Whites: 2.3
Blacks: 3.5
Therefore the percentage of each group in the 750+ cohort can be assumed at:
Critical reading:
Asians: 2.9 %
Whites: 1.4 %
Blacks: 0.1 %
Mathematics:
Asians: 8.1 %
Whites: 1.7 %
Blacks: 0.05%
Writing:
Asians: 2.5 %
Whites: 1.0 %
Blacks: 0.02 %
And in absolute numbers of students who achieved more than 750 points according to group sample size:
Critical reading:
Asians: ~ 2000
Whites: ~ 11900
Blacks: ~ 90
Mathematics:
Asians: ~ 5800
Whites: ~ 14200
Blacks: ~ 75
Writing:
Asians: ~ 1800
Whites: ~ 8400
Blacks: ~ 40
For the sake of camparison to your variable of “outperforming at the score of 750 or above”:
Critical reading:
Asians outperform whites at a ratio of 2.0:1
Whites outperform blacks at a ratio of 26.5:1 *
Asians outperform blacks at a ratio of 52.4:1 *
Mathematics:
Asians outperform whites at a ratio of 4.7:1
Whites outperform blacks at a ratio of 36.5:1 *
Asians outperform blacks at a ratio of 172.6:1 *
Writing:
Asians outperform whites at a ratio of 2.5:1
Whites outperform blacks at a ratio of 42.8:1 *
Asians outperform blacks at a ratio of 105.5:1 *
* The problem of statistical error invalidates these numbers due to the comparatively very small sample size for african americans. Statistical inacuracies would amount to the asian-black outperformance ratio in mathematics presumably falling somewhere between 100 and 300. This inacuracy thereby renders the values useless and meaningless.
I do not believe these numbers to carry any meaning in detail, but they do show a rough general tendency: that the race gaps are widening- the asian-white gap increasing by roughly one third and the asian-black gap roughly quadrupling within a decade. Although I have derious doubt concerning the accuracy of these ratios, I believe it is irrelevant to the basic problem at what speed the race gap is widening. The basic problem is, that the measures that have been and are being implied for almost half a century either have no effect at all or are contraproductive in helping the majority of african americans.
For the discipline of mathematics, where the gaps have shown to be greatest, the decile scores should be approximately:
Decile | Asians | Whites | Blacks
9. | 736 | 665 | 554
8. | 682 | 620 | 511
7. | 643 | 588 | 480
6. | 610 | 560 | 454
5. | 578 | 534 | 429
4. | 547 | 509 | 405
3. | 514 | 481 | 379
2. | 475 | 449 | 348
1. | 421 | 404 | 305
The shocking aspect of this is not the difference in the top deciles, but rather that the white-black difference remains in low deciles. The differences in the first deciles of whites and asians converge to a presumable minimum at about 400 Points. The tragedy is, that about 40% of african americans fall below that minimum and that the cohort of 350 points and below is almost exclusively african american. The real problem here is, that minority programs have always just concentrated on exploiting the low average of african americans to attaining a bonus for the elite whilst ignoring the broad community. I know this is a grave accusation, but the expectable results of this would be a thriving black elite that is well established in society and ecconomy profiting from minority politics and a large protion of black people who are in social, ecconomical and educational aspects literally falling off the bottom end of the scale. There, I believe, lies the rub.
Andrew, I commend you for your outstanding statistical analysis of recently published racial SAT I score numbers by the College Board. I also agree with you on the importance and relevance of your findings.
Good work, Andrew!
http://www.jbhe.com/features/49_college_admissions-test.html
Almost No Blacks Among the Top Scorers
on the Scholastic Assessment Test
It is important to explain how the SAT racial scoring gap challenges affirmative action policies at the nation’s highest-ranked colleges and universities. Under the SAT scoring system, most non-minority students hoping to qualify for admission to any of the nation’s 25 highest-ranked universities and 25 highest-ranked liberal arts colleges need to score at least 700 on each portion of the SAT.
For admission to the very highest ranked, brand-name schools such as Princeton or MIT, applicants need scores of 750 to be considered for admission. Yet, as we shall see, only a minute percentage of black test takers score at these levels. Thus, if high-ranking colleges and universities were to abandon their policies of race-sensitive admissions, they will be choosing their first-year students from an applicant pool in which there will be practically no blacks.
Let’s be more specific about the SAT racial gap among high-scoring applicants. In 2005, 153,132 African Americans took the SAT test. They made up 10.4 percent of all SAT test takers. But only 1,132 African-American college-bound students scored 700 or above on the math SAT and only 1,205 scored at least 700 on the verbal SAT. Nationally, more than 100,000 students of all races scored 700 or above on the math SAT and 78,025 students scored 700 or above on the verbal SAT. Thus, in this top-scoring category of all SAT test takers, blacks made up only 1.1 percent of the students scoring 700 or higher on the math test and only 1.5 percent of the students scoring 700 or higher on the verbal SAT.
If we eliminate Asians and other minorities from the statistics and compare just white and black students, we find that 5.8 percent of all white SAT test takers scored 700 or above on the verbal portion of the test. But only 0.79 percent of all black SAT test takers scored at this level. Therefore, whites were more than seven times as likely as blacks to score 700 or above on the verbal SAT. Overall, there are more than 39 times as many whites as blacks who scored at least 700 on the verbal SAT.
On the math SAT, only 0.7 percent of all black test takers scored at least 700 compared to 6.3 percent of all white test takers. Thus, whites were nine times as likely as blacks to score 700 or above on the math SAT. Overall, there were 45 times as many whites as blacks who scored 700 or above on the math SAT.
If we raise the top-scoring threshold to students scoring 750 or above on both the math and verbal SAT — a level equal to the mean score of students entering the nation’s most selective colleges such as Harvard, Princeton, and CalTech — we find that in the entire country 244 blacks scored 750 or above on the math SAT and 363 black students scored 750 or above on the verbal portion of the test. Nationwide, 33,841 students scored at least 750 on the math test and 30,479 scored at least 750 on the verbal SAT. Therefore, black students made up 0.7 percent of the test takers who scored 750 or above on the math test and 1.2 percent of all test takers who scored 750 or above on the verbal section.
Once again, if we eliminate Asians and other minorities from the calculations and compare only blacks and whites, we find that 0.2 percent of all black test takers scored 750 or above on the verbal SAT compared to 2.2 percent of all white test takers. Thus, whites were 11 times as likely as blacks to score 750 or above on the verbal portion of the test. Overall, there were 49 times as many whites as blacks who scored at or above the 750 level.
On the math SAT, only 0.16 percent of all black test takers scored 750 or above compared to 1.8 percent of white test takers. Thus, whites were more than 11 times as likely as blacks to score 750 or above on the math SAT. Overall, there were more than 61 times as many whites as blacks who scored 750 or above on the math section of the SAT.
In a race-neutral competition for the approximately 50,000 places for first-year students at the nation’s 25 top-ranked universities, high-scoring blacks would be buried by a huge mountain of high-scoring non-black students. Today, under prevailing affirmative action admissions policies, there are about 3,000 black first-year students matriculating at these 25 high-ranking universities, about 6 percent of all first-year students at these institutions. But if these schools operated under a strict race-neutral admissions policy where SAT scores were the most important qualifying yardstick, these universities could fill their freshman classes almost exclusively with students who score at the very top of the SAT scoring scale. As shown previously, black students make up at best between 1 and 2 percent of these high-scoring groups.
Looking to the Future
In the Grutter case upholding affirmative action in college admissions, Justice Sandra Day O’Connor’s decision expressed the goal of eliminating affirmative action over the next 25 years. At the moment there is no evidence that substantial progress toward closing the test scoring gap will occur. Thus, the huge and growing gap in SAT scores, and particularly the scores at the highest levels, becomes one of the nation’s most urgent problems.