Blog Post by: Benjamin Landy , on February 21, 2014
This year, about 3 million students will graduate from high school. Of the 2 million who go on to attend college, more than 1.5 million will take the SAT. And, as usual, colleges will lean heavily on the results to assemble America's next educational elite.
Here are some other things we know about the SAT:
· SAT scores are a better predictor of a students’ socioeconomic background than how well they will perform in college.
· SAT scores are a worse predictor of how well students perform in college than high school GPA.
· Colleges’ use of the SAT has a disproportionate adverse impact on black and Hispanic students, as well as low-income students of all races.
These are inconvenient facts for the unregulated, $4 billion-a-year testing industry. Elite colleges and universities are deeply invested in the notion that standardized testing constitutes a fair, practical, and even scientific method for determining which applicants rise and fall in America’s increasingly stratified higher education system. It’s nearly impossible to talk about the modern meritocracy without talking about the SAT.
But academics have known for years that the SAT (and its counterpart, the ACT) are poor predictors of students’ undergraduate success. Even the College Board acknowledges that high school grades are a better measure of students’ qualifications; they recommend colleges not “overuse” test results, which increase predictive validity only in combination with high school GPA.
The response from higher education has been tepid. A number of competitive colleges have adopted “test-optional” admissions policies, including liberal arts colleges like Bowdoin, Smith and Bates, and national universities like Wake Forest. But, so far, not one top-ranked institution has moved to reduce their reliance on the SAT or ACT. In fact, surveys indicate that increasing competition among the most elite schools—Harvard, Columbia, Stanford, and so on—has resulted in an even higher premium on testing.
One reason for this overreliance on the SAT is institutional prestige. “Colleges fear that dropping their ACT/SAT requirements might signal potential applicants and other important stakeholders that they are lowering academic standards,” writes FairTest’s Robert Schaeffer. “College rankings, particularly those from U.S. News & World Report magazine, which include average test scores in their calculations, help reinforce this concern.”
The signaling issue is complex, and won’t be easily resolved. But the actual research on test-optional admissions is clear: colleges that have dropped their SAT or ACT requirement have seen virtually no difference in academic performance between students who did and did not submit test scores as part of their application.
Earlier this month, the NACAC released the largest-ever study of test-optional admission policies, covering 123,000 students and 33 colleges of varying types. They found the following:
· High school grades do a better job of predicting college success than test scores alone.
· Students with low grades in high school, but high test scores, generally receive lower college grades than students with high grades in high school and low test scores.
· College students who did not submit test scores had an average college GPA just 0.05 lower than students who did submit test scores.
· The difference in graduation rates between submitters and non-submitters was 0.6 percent.
· Non-submitters were more likely to be non-white, the first in their families to go to college, female, or Pell Grant recipients (that is, low-income).
Those results align with empirical evidence from individual schools such as Wake Forest University, which professor Joseph A. Soares documents in SAT Wars: The Case for Test-Optional College Admissions, and numerous other studies (look here, here, and here) that have found little or no adverse impact at colleges that reduced or eliminated their testing requirement as a criterion for admissions.
Unfortunately, few educational institutions have yet felt sufficient pressure from policymakers and stakeholders to modernize their admissions policies. One problem, as noted earlier, is that colleges use high SAT scores to signal institutional prestige. Another is the fact that testing benefits colleges’ bottom line, by screening out otherwise talented applicants from poor socioeconomic backgrounds. (You know—those who would need financial aid.)
It should be noted that affirmative action is also one reason more colleges haven’t felt the pressure to go “test-optional.” The use of race-based preferences in college admissions has effectively softened the SAT’s disparate impact on minorities and low-income students. But if the courts continue to turn against racial affirmative action, as was indicated by the Supreme Court’s 2013 decision in Fisher v. University of Texas, colleges may be forced to respond to testing’s discriminatory effect. Already, the NAACP Legal Defense Fund has filed a federal civil rights complaint against eight specialized New York City high schools, including Stuyvesant and Bronx Science, on the grounds that their entrance exam violates Title VI of the Civil Rights Act. The legal outcome of that complaint could pave the way for similar challenges to the use of standardized testing in higher education.
Until that happens, colleges are likely to go on ignoring the flaws in their ostensibly meritocratic admissions process. The "test-optional" movement is gaining momentum, but for now, our best and brightest will continue to be sorted more by their backgrounds than their abilities.