SAT Scores Decline in the District
Drop Reflects National Trend After Recent Changes to College Entrance Exam

By Daniel de Vise
Washington Post Staff Writer
Wednesday, August 29, 2007

SAT stores dropped 30 points in the District with the Class of 2007, reflecting a downward trend nationwide in the two years since the college entrance test was revised and expanded.

Data released by the College Board yesterday showed that scores declined on all three sections of the test -- reading, writing and math -- in the District, Maryland and Virginia. Performance in the District, including at private schools, fell 9 points in reading, 10 points in math and 11 points in writing to a combined 1411, reflecting the second consecutive year of decline.

Maryland's scores fell 13 points overall, and Virginia's dipped five points. Only Loudoun and Arlington counties reported higher scores locally, and a few districts, including Prince George's and Arundel counties, did not release data.

Nationally, the Class of 2007 was the largest and most diverse group of students to take the SAT, which has evolved over the decades from an elite sorting tool into almost a rite of college preparation, the College Board said.

For the first time, an entire state, Maine, reported 100 percent participation in the SAT, which is required of students to promote rigor and college preparation. Baltimore educators pay the fee for high school students to take the test, resulting in near-universal participation but lower scores, district officials say.

Montgomery students took the test in record numbers, Superintendent Jerry D. Weast said. The county's performance was "a perfect Simpson's paradox," a statistical scenario in which the success of individual groups reverses when they are combined, he said.

Scores rose for Hispanic and white students in Montgomery and declined for black and Asian students. But the overall county average was down 10 points.

The county average is pulled down, Weast said, by students who are new to college-entrance testing and score low because of poverty and inadequate preparation. Participation on the test reached 79 percent in the county, an all-time high, with black and Hispanic students accounting for three-quarters of the increase.

"Eighty percent of any group taking anything is pretty high," Weast said. "We aren't eliminating students who aren't prepared."

Officials with the College Board, which administers the SAT, said the scores reflect greater participation among students from lower socioeconomic backgrounds who previously might not have aspired to college, let alone a college-entrance exam.

They dismissed two other factors cited by students and educators: a "fatigue factor" caused by the addition of a third section, writing, to the test, and a declining share of students choosing to take the longer SAT a second and third time to burnish their scores.

"The one thing we can say is, we know that this group of students . . . is slightly more diverse," reflecting populations "that have not been going to college at as high a rate as others," said Wayne J. Camara, College Board vice president for research.

More black, Hispanic and Asian students took the SAT this spring than in any previous year; minorities accounted for two-fifths of all test-takers.

The country as a whole hasn't performed this poorly on the SAT since 1999, when the combined reading and math score was 1016. Scores rose steadily in the first half of this decade but plunged for the Class of 2006 with the added writing section.

Since the peak with the Class of 2005, the national average has dropped six points, to 502 in verbal ability; five points, to 515 in math; and three points, to 494 on the writing section.

Critics of standardized testing point to the decline -- and the shortage of plausible explanations -- as evidence the new SAT is flawed.

If a more diverse testing pool is responsible for lower performance, they ask, then why did scores steadily rise in the first part of this decade, when the test-taking population was steadily diversifying?

The erosion "indicates that the College Board failed to keep its promise that the revised SAT would be comparable to the old test," said Robert Schaeffer, public education director of the National Center for Fair and Open Testing. He said that participation in the rival ACT has increased by almost 10 percent since the revised SAT was introduced and that SAT participation has increased about 1 percent.

Students consistently report they are tired after sitting through the new test, which lasts 3 hours and 45 minutes. Last year, College Board officials noted a related decline in the number of students retaking the test, a trend that would seem to yield lower scores as SAT performance tends to rise on a student's second attempt.

Christopher Garran, principal of high-performing Walter Johnson High School in Bethesda, said students are choosing not to retake the test "if their scores are around the average required by the colleges in which they are interested, because the test is so long."

Garran also said students are opting to take the SAT once and then take the ACT, rather than take the SAT two or three times. Last year, for the first time, Walter Johnson officials reported ACT averages on the school profile they send to colleges.

Sean Bulson, principal of neighboring Bethesda-Chevy Chase High School, said his students "are hearing from college recruiters and the media that colleges are looking more at the whole student." In that environment, he said, "it has been difficult to maintain a sense of urgency among our very busy students to focus on these tests the way they did just a few years ago."

College Board officials said the one-year drop is statistically insignificant: The average score declined no more than three points on any section of the test, a quantity smaller than the value of a correct answer on one SAT multiple-choice question.

Staff writers Jay Mathews and Philip Rucker and researcher Meg Smith contributed to this report.