A decade of education reform has lifted the academic performance of college-bound seniors of all races and ethnicities - white, black, Asian, Hispanic.
But there's a dark side to this good news: The improvement has been dramatically uneven, with white students raising their scores on college-entrance exams to a greater degree than other students have lifted theirs.
As a result, the performance gap - long a distressing feature of American education - is wider now than it was before "standards" and "accountability" became the watchwords of reformers across the US.
Indeed, SAT scores released this week by the College Board, when compared with test scores from 1991, reveal that the 1990s can be seen as a decade of "educational backsliding" for all minority groups except Asian-Americans, some analysts say.
"The gap doesn't just 'remain.' It has widened dramatically, if you read this report with the right eye," says Seppy Basili, vice president of pre-college programs at Kaplan Inc., an education and test-preparation firm.
At a minimum, the gap suggests that minority students are not as well prepared as whites are for the academic rigors of college, the traditional ladder to upward mobility in US society. More broadly, it is a sign that much more remains to be done to improve minorities' access to excellent schools and teaching.
The College Board, which administers the tests, puts a positive spin on this year's test results. It notes that the 1.3 million SAT takers now entering college include the "largest number of minority students in history" - more than one-third of the total.
Such numbers indicate that a rising share of minority students see college as a viable option. Nearly 364,000 SAT test takers were students whose parents had not attended college, a development that college board officials called "very heartening."
But others offer a less sanguine assessment.
"The irony is that the continuing surge [of minority SAT test takers] shows the aspiration to be highly educated is clearly there," says Hugh Price, president of the Urban League in New York. "What isn't there is reciprocal commitment on the part of public schools and society to provide a quality education to children of color."
The SAT results, he says, reflect "a continuing disinvestment" in the urban school systems that educate a vast majority of children of color.
Part of the problem, analysts say, is continued reliance on local property taxes to fund local schools. That tax base continues to grow in affluent communities, but not as much as in urban school districts.
Other reasons are often mentioned, as well. Some of the test-score disparity has to do with costly SAT test preparation, which many affluent kids in suburban schools receive and less-well-off kids in urban schools don't. Some of it is the rise of intense parent involvement in the more well-heeled communities.
"We know there's virtually no access to coaching and testing in some of those minority communities," says William Hiss, an administrative vice president at Bates College in Lewiston, Maine. "On the other hand, I know upscale communities are just pouring the gas in the carburetor to improve student achievement as measured by both grades and testing."
These differences, analysts say, are showing up in the SAT scores.
Among African-Americans, for instance, verbal SAT scores rose 6 points and math scores jumped 7 points in the past 10 years. But white students' scores rose 11 points for verbal and 18 points for math during the same period, widening an already-existing gap.
"The College Board congratulates themselves for the achievement gap narrowing between women and men," says Mr. Basili. "Why don't they note the performance gap is widening between whites and ethnic minorities? It's the wrong headline."
College Board officials duly note that a gap exists - and they agree with critics that less progress was made toward educational equity during the 1990s than many had believed. "These differences [in achievement] are a powerful illustration of a persistent social problem in our country: inequitable access to high-quality education," said Gaston Caperton, the College Board's president, in a prepared statement.
The SAT has limits as a gauge of social and educational progress. It includes only students who choose to take the test - not all students. In some states where verbal and math scores soared over the past decade, very few students took the test, distorting a state's performance.
Some critics also say the SAT is biased against minorities and women. But even the SAT's fiercest opponents say the growing achievement gap may have a silver lining.
"It could simply be that greater percentages of African-American and Latino students are taking the test, and so more 'C' students might be taking it," says Robert Schaeffer, a testing expert at FairTest in Cambridge, Mass. "That may lower the average score. But it may not be a bad thing if more kids are thinking about college."
Despite the recent foam and fury concerning the SAT - over whether it is a valid indicator of student readiness for college - many states have seen their students' SAT scores improve.
North Carolina's college-bound high school seniors saw their average verbal SAT scores rise 15 points and math scores leap 25 points during the past decade. It was the biggest overall improvement in the US among the 23 states where half or more graduates had taken the test.
Some attribute the improvement to the fact that, for about five years now, North Carolina has been paying bonuses to teachers whose students score higher on statewide tests - a tactic that seems to carry over into SAT scores.
But to Jo Dover, that big improvement is linked to something else. For three of the past four years, she was on a squad of consultants assigned to help turn failing schools around.
"It's really been due to the concentration we've given to reading, writing and math - going back to basics," she says. "That's what did it."