ETS Creating Demographic Index for SAT
A program under development by the maker of the SAT has added to the debate over whether a student's race should be considered in the college-application process and how much weight admissions officers should give test scores.
The Educational Testing Service is creating an index that would compare students' SAT scores against the averages of other test-takers with similar racial and demographic backgrounds. Those who outscored their demographic peers would be dubbed "Strivers" because they surpassed the expectations of young men and women of similar race, poverty, family, and educational backgrounds.
The tool could help admissions officers identify disadvantaged and minority students who were deemed the most likely to succeed because they had overcome odds against them.
"What we're trying to do—not just with the Strivers program but with others—is to look at different ways of capturing a student's strengths," Wayne Camara, the executive director of the College Board, which sponsors the SAT, said at a Washington news conference announcing the admissions-test results for the class of 1999.
With the new index, admissions counselors might be able to determine a student's ability to study independently and to overcome difficult circumstances, Mr. Camara said. "These are skills that the SAT doesn't measure," he said.
ETS officials say that the Strivers program is still being developed and that details are forthcoming. "In six to eight weeks, we'll have something for you based on a complete report," said Kevin Gonzalez, a spokesman for the giant nonprofit testing organization, based in Princeton, N.J.
According to data released last week by the College Board--which contracts with the ETS to run the SAT—the gap in scores between whites and most minority students persists. This year's results show that whites scored an average of 527 on the verbal section of the SAT I: Reasoning Test and 528 on the math portion, out of a possible 800 on each section. African-Americans scored at 434 on verbal and 422 on math.
In the past 10 years, the black-white gap narrowed by 2 points on the verbal exam, but rose 12 points on the math section.
The composite SAT I scores stayed the same in verbal this year, but fell 1 point in mathematics. During the past decade, verbal achievement rose 1 point, and math scores jumped 9 points.
The two portions of the SAT I assess a student's general abilities in such areas as vocabulary, reading comprehension, and early algebra. The New York City- based College Board also offers the SAT II, which tests students in such subjects as literature, American history, and biology.
Scores from the SAT are an important factor in a college application, but they shouldn't be a determining factor for admission, testing officials say.
To help admissions officers, the ETS is planning to offer two different indexes, according to a story in the Aug. 31 Wall Street Journal. One would consider a test-taker's race, socioeconomic status, language background, the past performance of students at his or her school, the school's location and demographic makeup, and the employment status of the student's mother. The second would disregard the test-taker's race.
If a student scored 200 or more points higher than others in his or her demographic-peer group, he or she would be called a Striver.
College- admission officials say the Strivers program would be most helpful to larger universities that process thousands of applications. Most of them, however, already have their own procedures for identifying promising students who would add diversity to their enrollments.
"With a fairly sophisticated admissions office, my hunch is there would be no gain" from using the Strivers index, said Kati Haycock, the director of the Education Trust, a Washington-based policy organization that promotes higher achievement by disadvantaged students. "Where there may be some benefit is in colleges that don't have generously funded admissions programs or haven't thought about [diversity] in this way," she said.
In California, where a voter initiative prohibited most racial preferences in public institutions, the University of California system considers such factors as students' ability to overcome obstacles, the determination they show to succeed in school, and their school leadership roles, according to Michael W. Kirst, a professor of education at Stanford University. The UC system also gives more weight to the SAT II than to the SAT I on the grounds that the former is better at measuring school success.
"That strikes me as a whole lot more sophisticated" than the Strivers concept, Mr. Kirst said. "
Small schools may be better equipped to handle the workload of analyzing student applications to determine which disadvantaged students might excel.
"When we look at SAT scores, we try to take all of that information into account anyway," said Richard L. Nesbitt, the acting director of admissions at Williams College, a highly regarded, 2,000-student private school in Williamstown, Mass. "Our operation is small enough that we can look at each individual closely."
Williams received 5,000 applications last year and accepted 1,150 of them, Mr. Nesbitt said.
Critics of affirmative action said last week that the Strivers program could raise questions about the legality of factoring race into admissions decisions at publicly financed universities.
"The version that relies on race is highly suspect," said Terence J. Pell, a senior counsel for the Washington-based Center for Individual Rights. "It's race norming dressed up in new clothes."
Mr. Pell represents white applicants who say they were denied admission to the University of Michigan's undergraduate colleges and law school because of separate standards for students of different races. He won a similar suit at the federal appellate level against the University of Texas law school, which had the effect of outlawing race-based admissions in public institutions in Texas, Louisiana, and Mississippi.
Ms. Haycock, a former University of California admissions officer, said that schools barred from using race-based admissions policies would be unlikely to rely on a Strivers index that analyzed scores by race.
"It's not about to be a backdoor effort to do affirmative action under a different name," she said. "It's not likely to change the overall patterns" of college enrollment.
Vol. 19, Issue 1, Pages 1, 12Published in Print: September 8, 1999, as ETS Creating Demographic Index for SAT