Distribution of 12th Grade TIMSS Math Scores

The top 5% of an organization is where the rubber meets the road. The top 20% does more work than the lower 80%. This is why it's so significant that the top fifth percentile of American 12th grade students scored 101 points lower than Sweden's top fifth percentile in TIMSS General Math--and 63 points lower than Switzerland's, 31 points lower than Slovenia's, 64 New Zealand, 83 Netherlands, 62 Iceland, 23 Hungary, 31 Germany, 34 France, 68 Denmark, 27 Czech Republic, 53 Canada, 32 Austria, and 63 points lower than Australia's. Our top fifth percentile scored higher than Lithuania and Italy, but their median scores were, respectively, 16 & 21 points higher than our median scores. The only two countries whose median and top fifth percentile scores were lower than ours were South Africa and Cyprus, whose top fifth percentile scored 89 and 49 points lower than ours, but even Cyprus' lowest fifth percentile scored higher than our lowest fifth percentile--almost a clean sweep, so to speak.

Comparing median scores can be misleading. Sweden had a median score which was 92 points higher than the US, but it's more significant that their top fifth percentile scored 101 points higher than ours. Comparing the top fifth percentile can also detect aberrations which need to be explained. It is important to know why the 228 point gap between the median scores of South African and Netherlands was larger than the 83 point gap between their top fifth percentiles. The distribution curves provide a possible explanation:

14% of the South African population is White. Most of them are descended from the Netherlands whose top fifth percentile scored 704. If South African Whites constituted the top fifth percentile, then blacks constituted the majority if not all of the seventy fifth percentile. Removing Whites from this score provides a bell curve where the top fifth percentile is at a score of 420 and which is as symmetric as those for the US and Sweden. This is 284 points lower than the fifth percentile for Netherlands, which is larger than the 228 point gap in their median scores, which is a gap consistent with all other scores around the world:

From Australia to the Netherlands, countries score consistently higher than the US at each percentile, which scores consistently higher than South African blacks at each percentile.

It is unlikely that the other 860 million blacks in Africa scored higher than the top fifth percentile of blacks in South Africa, which means that the average math skills of the top 5% of black Africans is about equal to the average math skills of the lowest fifth percentile of Whites in most European nations. In other words, almost all of the Whites in most European nations scored higher than ALL 900 million blacks in Africa. If the academic skills of the 12% of the American population which is black is equivalent to the blacks in South Africa, then it would also be true that all of the Whites of America score higher than all of the American blacks.

5 | 25 | 50 | 25 | 5 | |

S. Africa, 14% Whites | 264 | 304 | 337 | 380 | 532 |

US | 325 | 395 | 454 | 521 | 621 |

Australia | 357 | 459 | 523 | 585 | 684 |

Austria | 393 | 461 | 515 | 573 | 653 |

Sweden | 396 | 483 | 546 | 620 | 722 |

S. African Blacks | 264 | 304 | 337 | 380 | 420 |

Denmark | 406 | 487 | 548 | 609 | 689 |

France | 392 | 468 | 523 | 578 | 655 |

Switzerland | 395 | 478 | 539 | 601 | 684 |

Netherlands | 407 | 498 | 565 | 622 | 704 |

Belgium, Hong Kong, Japan, Korea, and Singapore didn't participate in the 12th grade TIMSS, but at the 8th grade level their top fifth percentile scored, respectively, 58, 90, 121, 128, and 139 points higher than ours and 285, 317, 348, 355, and 366 points higher than South African blacks. Singapore's lowest fifth percentile scored slightly higher than our median score and 71 points higher than the top fifth percentile of South African blacks. In other words, there was no overlap between South African blacks and Singapore. The very best black students in South Africa scored lower than the very worst students in Singapore by an amount which is almost as big as the gap between American girls and boys. Within this gap (430 to 501) fall the median scores of the US, Latvia, Spain, Iceland, Lithuania, Cyprus, Portugal, Scotland, Romania, and Greece. The 2-12 point overlap between the top fifth percentile of South African blacks and the lowest fifth percentile of Belgium, Hong Kong, Japan, and Korea could be explained by intermarriage with or immigration of blacks who would fill those lower ranks.

Washington, DC, is proof that money doesn't solve the problem. Their ACT Composite score is consistently the lowest in the nation, even though students in Washington, DC are almost the highest costing students in the country. Washington, DC spends twice as much per student as Utah and Idaho, yet Utah and Idaho score almost as high as the Asian or White average. Alabama and Arkansas spend half as much for education but their students score more than 3 points higher than students in Washington, DC. The least costly students in the country are home schooled students, yet their average score was higher than any state's average score, including Oregon who scored the highest in ACT. The only group which scored lower than Washington, DC are blacks who scored lower than Washington, DC by an amount equal to the the amount by which Mexicans scored higher than Washington, DC. If money was the answer, Washington, DC, would have the highest scores in the nation, but the experience in Washington is proof that the problem is attempting to educate blacks, which can't be accomplished by any level of education expenditure.

The only reason that Washington, DC scored 0.8 points higher than the average score for blacks is that 7% of the population is Whites. Spending too much money for education created a bureaucratic nightmare which may have destroyed even the meager opportunity which young blacks had to learn in the first place, which might explain why the crosslinking studies between NAEP and IAEP show that Mozambique and South Africa scored even higher than the average black American score of 17 on the ACT.

While the distribution curves of the ACT data is not known, the following graph provides the most likely scenario for how the top scoring blacks in the nation compare to the lowest scoring Whites. The highest score for blacks is 19.2 which is a score which less than 10% of Whites fall below. Where the gap between the median scores of blacks and Whites is 4.8, the gap at the 95th percentile is 6.9. The lowest scoring Whites score higher than the median score for blacks. The academic skills for the highest scoring blacks in the country are lower than Whites who flunked out of or never graduated from high school.

The distribution curves for the NAEP Science show that the top two percentile of blacks score in the range of the median scores for Whites, suggesting that this highly politicized test from those folks who don't want us to know how poor the real intellectual skills of blacks are has been given some special attention.

The "liberal" theory that increasing the percentage of blacks will increase their academic skills or or improve their work productivity is belied by the fact that a ten percent increase in the percent of blacks by state accomplishes nothing less than reducing ACT scores of that state by one point.

Conversely, increasing the cost per student for education has almost zero correlation to an improvement in ACT scores. If there is any relationship, it is that an increase in spending is followed by a decrease in ACT scores.

The distribution curves for the Graduate Record Exam might be the most revealing of all the tests because this details the difference in peformance of students AFTER they have been educated under affirmative action programs. Women who majored in education and took the GRE towards the end of their undergraduate education scored just slightly higher than black students in general. But the upper two percentile of both of these groups scored lower than the median score of men who majored in engineering and in the range of the lowest two percentile of men engineering majors.

Expecting blacks with such poor academic skills to perform like Whites in schools, government, law, industry, or universities is an exercise in futility for all involved. Taxpayers are frustrated because their $7.7 trillion extra expenditures for education, which put the country into great debts, failed to narrow this gap even a tenth of a point. Blacks are frustrated because none of the promises of higher education could possibly come true. Whites are frustrated because qualified Whites are displaced by unqualified blacks in every segment of society. Families are frustrated because this contributed to a two thirds drop in family incomes. Businesses are frustrated because endless racial discrimination and harassment litigation has failed to produce a nickel's worth of product at the same time that it robbed R&D and capital expansion funds from entire industries. Government is frustrated because it will never give up on a failed education program just because it failed.