The Chronicle of Higher Education, May 9, 2008, Friday
Copyright 2008 The Chronicle of Higher Education
All Rights Reserved
The Chronicle of Higher Education
May 9, 2008, Friday
SECTION: STUDENTS; Pg. 20 Vol. 54 No. 35
BYLINE: PETER SCHMIDT
BODY:
One after another at this time of year, elite colleges trumpet the outstanding SAT scores of the applicants they have admitted. The question often raised by such announcements is just how much those scores matter.
Two recent studies conclude that they matter quite a lot. The researchers assert that selective colleges give excessive weight to SAT scores for the sake of bolstering their college-guide rankings and, in doing so, greatly complicate their pursuit of diversity.
The studies give advocates for minority students cause for both hope and fear at a time when highly selective colleges are rejecting more applicants than ever, and race-conscious admissions policies are coming under renewed attack in the courts and the political arena.
The reports' authors argue that selective colleges do not necessarily have to consider applicants' ethnicity and race to promote diversity. Rather, colleges could increase their enrollments of minority and low-income students simply by giving more weight to admissions criteria other than standardized-test scores.
At the same time, however, the studies suggest that powerful market forces have selective colleges under pressure to give even more weight to standardized admissions tests, hurting the prospects of low-income and minority applicants who are less likely to post high scores.
One of the reports, to be published next month in the book Realizing Bakke's Legacy: Affirmative Action, Equal Opportunity, and Access to Higher Education, charts sharp increases over the past few decades in the share of slots at 30 highly ranked colleges going to students with high SAT scores. While acknowledging that the trend is probably driven by a combination of forces -- including substantial long-term growth in the number of students taking the SAT -- it says the rise in the number of high scorers at many selective colleges is too steep to have taken place without at least some increase in the weight given to test scores.
The other study, summarized in the American Sociological Review late last year, likewise found that high scorers' share of selective-college enrollments has risen largely because of the institutions' "attempts to climb the pecking order of various college ranking systems." Based on statistical simulations of the impact of various changes in admissions policy, the study's authors conclude that selective colleges could achieve significant racial and ethnic diversity in their enrollments without race-conscious admissions policies if they would put more weight on factors such as class rank.
Many experts on college admissions say they doubt that selective colleges will be relying less on standardized tests anytime soon. Even if they did, the share of their entering classes with high scores might continue to rise as a result of trends beyond the colleges' control, such as increases in the overall number of students taking the tests, some of whom would probably add to the ranks of those performing well. The desire of many families to send their children to the most prestigious institutions they can get into also leads to high-scoring students' becoming more concentrated at colleges at the top of the rankings.
"I think these kinds of studies are very important to try to hold up the mirror for all of us," says Donald R. Hossler, a professor of educational leadership and policy studies at Indiana University at Bloomington, where he has served as vice chancellor for enrollment services.
But, Mr. Hossler says, "nobody is going to push the pendulum back as far as some of these studies suggest it should be pushed back."
Aiming High
The authors of the study being published in Realizing Bakke's Legacy are Catherine L. Horn, an assistant professor of educational leadership and cultural studies at the University of Houston, who served as one of the book's editors, and John T. Yun, an assistant professor of education at the University of California at Santa Barbara. They focused their analysis on the 30 colleges that U.S. News & World Report ranked highest in 2007 and examined long-term trends in the proportion of entering freshmen who exceeded the score thresholds of 500, 600, and 700 points on the SAT's verbal and math sections.
Their study has some major limitations. Because it relied on the information that the institutions chose to report to leading college guides, it does not have numbers from all 30 colleges for each of the years covered, which go back in roughly 10-year intervals to 1969. Perhaps most significantly, the two researchers were unable to adjust their findings to take into account a 1996 change in how the SAT was scored. The change resulted in substantially higher verbal-test scores for the same level of performance, so that today's scores of 700 to 750 are equivalent to pre-1996 scores of 640 to 690. As a result, the study's use of point-based thresholds exaggerates the growth in the share of each studied college's entering freshmen who were top scorers on the verbal portion of the SAT.
But the long-term growth in the share of high scorers at colleges is striking even when it comes to the math portion of the test, on which the 1996 adjustment did not cause scores to substantially change. Among the 22 institutions that reported how many students with SAT math scores over 700 they admitted in 1989 and in 2007, the proportion of students with scores that high rose about 25 percentage points, on average.
The researchers put the colleges they examined into three categories: those that have always been prestigious enough to fill their classes with high scorers, so that the increases can be attributed largely to their giving more weight to the SAT; those where the increases seem attributable mainly to rises in the number of high-scoring applicants; and those that brought about big increases by aggressively recruiting top-scoring applicants and accepting them at high rates.
The second study was carried out by Marta Tienda, a professor of demographic studies, sociology, and public affairs at Princeton University, and Sigal Alon, an assistant professor of sociology at Tel Aviv University. They conducted simulations of how various shifts in admissions criteria would affect enrollments, using student data from the federal "High School and Beyond" study, the National Education Longitudinal Survey, the College and Beyond database, and the University of Texas at Austin. The researchers concluded that selective colleges created their own need to use race-conscious admissions policies to promote diversity by placing so much emphasis on standardized tests. "The apparent tension between merit and diversity exists only where merit is narrowly defined by test scores," they argue.
On the basis of their analysis of data from the Texas flagship, which is required by state law to accept any state resident in the top tenth of his or her high-school class, Ms. Tienda and Ms. Alon argue that "using class rank as a measure of merit requires smaller or no race preferences to achieve diversity" and that shifting to admissions policies based on class rank would not hurt graduation rates.
Knocks on the Door
According to the National Association for College Admission Counseling, the proportion of colleges reporting that they give standardized-test scores "considerable importance" in their admissions decisions has risen from about 50 percent to about 60 percent over the past decade. As of the association's most recent member survey, in 2006, the only traits that colleges were likelier to weigh as heavily were an applicant's grades in college-preparatory courses and the strength of that student's high-school curriculum. Fewer than a fourth of the institutions surveyed that year reported giving such weight to class rank.
But two other figures strongly suggest that colleges' admissions policies are hardly the only forces at work. Over the past decade, the total number of students taking the SAT has increased by more than 40 percent, to about 1.5 million. And of the nation's nearly 2,200 private four-year colleges, the 30 highly ranked institutions studied by Ms. Horn and Mr. Yun enrolled about 25,000 -- or more than a third -- of the 74,000 students who earned SAT verbal scores above 700 last year.
Some college officials and admissions experts suggest that the rise in the number of high scorers at top-ranked colleges may not be a result of the colleges' trying harder to enroll such students so much as the students' trying harder to get into top colleges.
"All of the top-achieving kids in the country are more and more focusing their attention on a smaller and smaller group of schools," says Jeff B. Brenzel, dean of undergraduate admissions at Yale University.
Ronald G. Ehrenberg, director of the Cornell Higher Education Research Institute, says rising levels of economic inequality have caused many Americans to conclude that it is important "to try to buy the best" when it comes to their children's college educations.
Thomas E. Lifka, associate vice chancellor for student academic services at the University of California at Los Angeles, says UCLA's enrollments of high scorers would rise even if it gave no consideration at all to the SAT and ACT, simply because so many more talented applicants are knocking on its doors.
Still, given how strongly standardized-test performance is correlated with race and family income, Thomas G. Mortenson, a senior scholar at the Pell Institute for the Study of Opportunity in Higher Education, says that by letting high-scoring students dominate their campuses, selective colleges "are shirking their responsibility to educate low-income students."
"The demography of this country is changing, and some of these high-SAT institutions are fighting it," he says. "The population they are serving is the population with the greatest inherited wealth and privilege to begin with."
LOAD-DATE: May 7, 2008
<< Home