SAT Scores Down Again, Wealth Up Again
Average scores on the SAT fell this year in critical reading, mathematics and writing. The writing test only has two years of scoring history, but for the other tests, this year's scores marked back-to-back years of score declines -- something that has not happened since 1991.
The mood at the Washington press conference of College Board officials was upbeat, with leaders playing down the significance of the declines, noting the increasing numbers and diversity of students taking the SAT, progress in encouraging students to take rigorous high school courses and progress (especially measured over the long term) for some racial and ethnic groups in their SAT score averages.
But other figures released Monday show large and growing gaps between black and other students on the SAT, a continuing growth in the wealth of those taking the SAT, a continued correlation between SAT scores and family income, and evidence that the ACT -- the SAT's archrival -- may be gaining strength.
With tons of data released, numbers can be found to make any number of points. But this year's data come at a crucial time for the SAT. An increasing number of colleges -- especially liberal arts institutions -- are ending requirements that applicants submit SAT scores. And as more states consider measures to ban affirmative action, standardized test results are being cited by critics of affirmative action to gain support, prompting many educators to want admissions systems that place less emphasis on test scores.
Here are the overall numbers, and the breakdowns by racial and ethnic group, for the SAT this year.
SAT Averages by Racial and Ethnic Group, 2007
|Group||Critical Reading Score||1-Year Change (Reading)||10-Year Change (Reading)||Math Score||1-Year Change (Math)||10-Year Change (Math)||Writing Score||1-Year Change (Writing)|
This year's total declines are all the more striking because they follow large decreases last year, when the five-point drop in critical reading, to 503, was the largest decline since 1975 and the two-point drop in mathematics, to 518, was the largest dip since 1978. Last year, SAT officials attributed the drops to a decline in the number of those who took the test more than once, and they denied strongly that changes in the SAT -- especially the much disliked lengthening of the exam time to make room for the new writing test -- had anything to do with the drop.
This year, College Board officials said that the repeat test-taking was back to normal, but the numbers didn't go back up. Laurence Bunin, general manager of the SAT, said that the scores were "within the expected range" and didn't signify any major changes or raise causes for alarm.
He urged reporters to take a "long term perspective" and not focus on short term drops, and he said that there was "extremely exciting news" in the gains of some minority groups over time. For instance, the gaps between the average scores of American Indian, black and Puerto Rican students and the averages for all test takers are at all time lows. That doesn't, of course, mean that they are small -- the gap on reading scores between black students and all test takers is 69 points.
And in the short term, the gap is growing significantly between black and Asian students taking the SAT. Across all three parts of the SAT, Asian students had the greatest gains, going up by 5 points, while black students fell by 4 points.
Critics of the College Board noted that SAT officials had promised repeatedly that introducing changes in the test would not result in any significant score changes. "The College Board failed to keep its promise that the revised SAT would remain a consistent measuring tool," said Robert Schaeffer, public education director of FairTest: the National Center for Fair & Open Testing. He said that the rationale offered last year (fewer repeat test takers) "does not hold up" this year, and that the "unexplained trend will further chip away at the exam's credibility."
While College Board officials stressed that the SAT is the top test for students to take when preparing to go to college, Schaeffer noted that in the two-plus years that the "new" version of the SAT has been in use, two dozen colleges have ended SAT requirements for applicants. While College Board officials correctly point out that most of the colleges that do not require the SAT aren't competitive in admissions, FairTest points out (also correctly) that the ranks of competitive colleges not requiring the SAT are also growing -- to 26 of the top 100 liberal arts colleges (as measured by U.S. News & World Report).
One of the other notable trends in recent years of SAT data has been that wealthier students appear to be making up larger shares of test takers. This year continued the trend, which attracts attention because there appears to be a clear relationship between family income and test scores. The means that follow are the totals of all three parts of the SAT.
SAT Population, and Mean Scores by Family Income Level, 2005-7
|Income Level||% of Test Takers 2005||% of Test Takers 2006||% of Test Takers 2007||Mean Total SAT Score|
|Less Than $10,000||5||4||4||1301|
|More than $100,000||21||24||26||1637|
The shares of test takers for those in the four categories up to $40,000 represented 32 percent of test takers just two years ago, and 27 percent of test takers today. During that same time period, the percentage of test takers with family incomes above $100,000 increased from 21 to 26 percent.
Wayne J. Camara, vice president for research and analysis of the College Board, said that the shift in income levels shouldn't be viewed as a concern, citing several reasons. He said that because so many students don't fill out that question or do so and don't really know their parents' income, this is the piece of demographic data in which the College Board has the "least faith." He also said that inflation may be at play, since many families see increases in income. And SAT officials noted that they were giving out more fee waivers for the test than ever before.
While the ACT has had some modest declines in test takers with the lowest incomes, it has not seen similar gains at the upper income levels. Only 10 percent of those taking the ACT are from families with incomes in the $100,000 and up level -- a proportion that has been constant for five years.
Schaefer said he was not surprised by the shift in income demographics of those taking the SAT. "The SAT is increasingly a wealth test, and it provides the highest scores for those who have the most opportunity in society," he said. "When colleges rely on the SAT, students from low income families are being doubly punished."
Another question being raised by the latest SAT figures is how the SAT is holding up against the ACT. The SAT had a drop last year, from which it bounced back and gained some ground. But in recent years, ACT totals have been going up at a higher rate. This year, more than 1.3 million graduating seniors took the ACT, up from 1.21 million the year before and 1.19 the prior year. For the SAT, comparable figures are 1.49 this year, up from 1.47 the year before and 1.48 the year before that.
Bunin said that the totals this year reinforce the SAT's role as "the established leader" in the field.
Schaeffer scoffed at that. He said that in the two years since the SAT unveiled its new model, the SAT grew by only 19,088 while ACT attracted 114,348 additional test takers.
Christine Parker, a lead SAT expert at the Princeton Review, said that more and more of those coming to her company for SAT coaching are also requesting ACT lessons. While Parker said that from an educational perspective, she doesn't like the idea of students preparing for multiple tests, more of them are insisting on it, fearing that SAT scores that aren't optimal will doom their futures. (Almost all colleges that require a standardized test will take either the SAT or the ACT and there is a scale that in theory makes scores comparable. But there are also many students who believe that they do "better" on the ACT. Many high school counselors report that students who do well in class, but don't "test well" end up with better ACT scores than SAT scores.)
Parker said that she has been especially struck by the way the ACT -- traditionally strongest in the Midwest and not much of a presence in the Northeast -- is all of the sudden being taken seriously be high school students there. In New Jersey, for example, the percentage of high school seniors taking the ACT in the last five years has increased from 6 to 11 percent. In New York, the increase has been from 15 to 21 percent. In Connecticut, the increase has been from 7 to 16 percent.
Schaeffer said that while he is pleased to see the SAT get challenged, people should think twice about all tests. "The ACT may be more consumer friendly, but it's just a different test," he said. "The truth is that no standardized exam is needed in the admissions process, as a growing number of schools have demonstrated."