By David Casalaspi
Each year, more and more states are requiring all high school students to take one of the two major college-entrance exams: the SAT or the ACT. While most of these states simply require their students to take the test as a way to boost college readiness, some states even use the test for the purpose of fulfilling federal accountability mandates, which require all 11th graders to be tested on their mastery of their state’s curriculum. A recent post by Education Week reported that in the current school year, 25 states are requiring high school students to take either the SAT or ACT, and 12 states are using the SAT or ACT for federal accountability purposes. Here in Michigan, we recently signed a contract with College Board to begin administering the SAT to all high school students, moving away from the ACT, which had previously been our required test.
In general, SAT For All plans like these are often lauded by policymakers as a way to spur interest in college among students who otherwise would be hesitant to apply. These plans also ensure that all students have completed a major requirement found on most college applications. On the other hand, opponents of SAT For All plans worry that as more states begin using the SAT for accountability purposes, the test may be inappropriately used to punish teachers and principals whose students don’t do well. After all, the SAT is designed to measure students’ college-readiness, not the effectiveness of their teacher.
The Performance Paradox
A bigger problem with SAT For All plans, however has to do with their effect on overall performance. In the 1980s, SAT scores were higher than they are today, in part because during that time only a relatively small portion of high-performing high school students took the test—those who were planning to attend college. As more and more students have begun taking the test, and the population of test takers has become more diverse, aggregate scores have dropped. This is the case even though the scores for some subpopulations (like White students, Asian students, and Black students) have either remained flat or increased.
Take the table below, for example. As the table shows, between 2005 and 2015 the percentage of SAT test-takers who were White shrank from 56% to 47% while the percentage of students who were Black increased from 10% to 13% and the percentage of students who were Hispanic increased from 10% to 20%. Because Black and Hispanic students usually have lower test scores than white students, the overall SAT scores were almost guaranteed to drop. At the same time, the portion of non-college-ready students taking the test within every racial group has increased as well, diluting the overall score for each population subgroup over time.
|SAT Reading Scores of College-Bound Seniors, by Race/Ethnicity|
|Race/ethnicity||Avg. Score||% of Test Takers||Avg. Score||% of Test Takers||Avg. Score||% of Test Takers|
|Note: Data from College Board and NCES.|
All of this is to say that while SAT For All plans are borne of noble aspirations—the desire to see more students qualify for college admission—there is an inherent tradeoff in the short-term in the form of lower aggregate performance. In recent years, there has been much handwringing about declining SAT scores, and every year it seems the release of the College Board’s annual score data is met with fervent calls for faster education reform (see here, here, and here). But until we are able to ensure that every student is college-ready, increasing the pool of students taking a college-readiness test will only result in lower scores. It is impossible to simultaneously expand the pool of test-takers to include people who are not college-ready and still expect universally high outcomes. Excellence and equality are inherently in conflict.
Contact David: firstname.lastname@example.org